diff options
Diffstat (limited to 'third_party/python/setuptools/pkg_resources')
43 files changed, 18981 insertions, 0 deletions
diff --git a/third_party/python/setuptools/pkg_resources/__init__.py b/third_party/python/setuptools/pkg_resources/__init__.py new file mode 100644 index 0000000000..3fc97af475 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/__init__.py @@ -0,0 +1,3361 @@ +""" +Package resource API +-------------------- + +A resource is a logical file contained within a package, or a logical +subdirectory thereof. The package resource API expects resource names +to have their path parts separated with ``/``, *not* whatever the local +path separator is. Do not use os.path operations to manipulate resource +names being passed into the API. + +The package resource API is designed to work with normal filesystem packages, +.egg files, and unpacked .egg files. It can also work in a limited way with +.zip files and with custom PEP 302 loaders that support the ``get_data()`` +method. + +This module is deprecated. Users are directed to :mod:`importlib.resources`, +:mod:`importlib.metadata` and :pypi:`packaging` instead. +""" + +import sys +import os +import io +import time +import re +import types +import zipfile +import zipimport +import warnings +import stat +import functools +import pkgutil +import operator +import platform +import collections +import plistlib +import email.parser +import errno +import tempfile +import textwrap +import inspect +import ntpath +import posixpath +import importlib +from pkgutil import get_importer + +try: + import _imp +except ImportError: + # Python 3.2 compatibility + import imp as _imp + +try: + FileExistsError +except NameError: + FileExistsError = OSError + +# capture these to bypass sandboxing +from os import utime + +try: + from os import mkdir, rename, unlink + + WRITE_SUPPORT = True +except ImportError: + # no write support, probably under GAE + WRITE_SUPPORT = False + +from os import open as os_open +from os.path import isdir, split + +try: + import importlib.machinery as importlib_machinery + + # access attribute to force import under delayed import mechanisms. + importlib_machinery.__name__ +except ImportError: + importlib_machinery = None + +from pkg_resources.extern.jaraco.text import ( + yield_lines, + drop_comment, + join_continuation, +) + +from pkg_resources.extern import platformdirs +from pkg_resources.extern import packaging + +__import__('pkg_resources.extern.packaging.version') +__import__('pkg_resources.extern.packaging.specifiers') +__import__('pkg_resources.extern.packaging.requirements') +__import__('pkg_resources.extern.packaging.markers') +__import__('pkg_resources.extern.packaging.utils') + +if sys.version_info < (3, 5): + raise RuntimeError("Python 3.5 or later is required") + +# declare some globals that will be defined later to +# satisfy the linters. +require = None +working_set = None +add_activation_listener = None +resources_stream = None +cleanup_resources = None +resource_dir = None +resource_stream = None +set_extraction_path = None +resource_isdir = None +resource_string = None +iter_entry_points = None +resource_listdir = None +resource_filename = None +resource_exists = None +_distribution_finders = None +_namespace_handlers = None +_namespace_packages = None + + +warnings.warn( + "pkg_resources is deprecated as an API. " + "See https://setuptools.pypa.io/en/latest/pkg_resources.html", + DeprecationWarning, + stacklevel=2 +) + + +_PEP440_FALLBACK = re.compile(r"^v?(?P<safe>(?:[0-9]+!)?[0-9]+(?:\.[0-9]+)*)", re.I) + + +class PEP440Warning(RuntimeWarning): + """ + Used when there is an issue with a version or specifier not complying with + PEP 440. + """ + + +parse_version = packaging.version.Version + + +_state_vars = {} + + +def _declare_state(vartype, **kw): + globals().update(kw) + _state_vars.update(dict.fromkeys(kw, vartype)) + + +def __getstate__(): + state = {} + g = globals() + for k, v in _state_vars.items(): + state[k] = g['_sget_' + v](g[k]) + return state + + +def __setstate__(state): + g = globals() + for k, v in state.items(): + g['_sset_' + _state_vars[k]](k, g[k], v) + return state + + +def _sget_dict(val): + return val.copy() + + +def _sset_dict(key, ob, state): + ob.clear() + ob.update(state) + + +def _sget_object(val): + return val.__getstate__() + + +def _sset_object(key, ob, state): + ob.__setstate__(state) + + +_sget_none = _sset_none = lambda *args: None + + +def get_supported_platform(): + """Return this platform's maximum compatible version. + + distutils.util.get_platform() normally reports the minimum version + of macOS that would be required to *use* extensions produced by + distutils. But what we want when checking compatibility is to know the + version of macOS that we are *running*. To allow usage of packages that + explicitly require a newer version of macOS, we must also know the + current version of the OS. + + If this condition occurs for any other platform with a version in its + platform strings, this function should be extended accordingly. + """ + plat = get_build_platform() + m = macosVersionString.match(plat) + if m is not None and sys.platform == "darwin": + try: + plat = 'macosx-%s-%s' % ('.'.join(_macos_vers()[:2]), m.group(3)) + except ValueError: + # not macOS + pass + return plat + + +__all__ = [ + # Basic resource access and distribution/entry point discovery + 'require', + 'run_script', + 'get_provider', + 'get_distribution', + 'load_entry_point', + 'get_entry_map', + 'get_entry_info', + 'iter_entry_points', + 'resource_string', + 'resource_stream', + 'resource_filename', + 'resource_listdir', + 'resource_exists', + 'resource_isdir', + # Environmental control + 'declare_namespace', + 'working_set', + 'add_activation_listener', + 'find_distributions', + 'set_extraction_path', + 'cleanup_resources', + 'get_default_cache', + # Primary implementation classes + 'Environment', + 'WorkingSet', + 'ResourceManager', + 'Distribution', + 'Requirement', + 'EntryPoint', + # Exceptions + 'ResolutionError', + 'VersionConflict', + 'DistributionNotFound', + 'UnknownExtra', + 'ExtractionError', + # Warnings + 'PEP440Warning', + # Parsing functions and string utilities + 'parse_requirements', + 'parse_version', + 'safe_name', + 'safe_version', + 'get_platform', + 'compatible_platforms', + 'yield_lines', + 'split_sections', + 'safe_extra', + 'to_filename', + 'invalid_marker', + 'evaluate_marker', + # filesystem utilities + 'ensure_directory', + 'normalize_path', + # Distribution "precedence" constants + 'EGG_DIST', + 'BINARY_DIST', + 'SOURCE_DIST', + 'CHECKOUT_DIST', + 'DEVELOP_DIST', + # "Provider" interfaces, implementations, and registration/lookup APIs + 'IMetadataProvider', + 'IResourceProvider', + 'FileMetadata', + 'PathMetadata', + 'EggMetadata', + 'EmptyProvider', + 'empty_provider', + 'NullProvider', + 'EggProvider', + 'DefaultProvider', + 'ZipProvider', + 'register_finder', + 'register_namespace_handler', + 'register_loader_type', + 'fixup_namespace_packages', + 'get_importer', + # Warnings + 'PkgResourcesDeprecationWarning', + # Deprecated/backward compatibility only + 'run_main', + 'AvailableDistributions', +] + + +class ResolutionError(Exception): + """Abstract base for dependency resolution errors""" + + def __repr__(self): + return self.__class__.__name__ + repr(self.args) + + +class VersionConflict(ResolutionError): + """ + An already-installed version conflicts with the requested version. + + Should be initialized with the installed Distribution and the requested + Requirement. + """ + + _template = "{self.dist} is installed but {self.req} is required" + + @property + def dist(self): + return self.args[0] + + @property + def req(self): + return self.args[1] + + def report(self): + return self._template.format(**locals()) + + def with_context(self, required_by): + """ + If required_by is non-empty, return a version of self that is a + ContextualVersionConflict. + """ + if not required_by: + return self + args = self.args + (required_by,) + return ContextualVersionConflict(*args) + + +class ContextualVersionConflict(VersionConflict): + """ + A VersionConflict that accepts a third parameter, the set of the + requirements that required the installed Distribution. + """ + + _template = VersionConflict._template + ' by {self.required_by}' + + @property + def required_by(self): + return self.args[2] + + +class DistributionNotFound(ResolutionError): + """A requested distribution was not found""" + + _template = ( + "The '{self.req}' distribution was not found " + "and is required by {self.requirers_str}" + ) + + @property + def req(self): + return self.args[0] + + @property + def requirers(self): + return self.args[1] + + @property + def requirers_str(self): + if not self.requirers: + return 'the application' + return ', '.join(self.requirers) + + def report(self): + return self._template.format(**locals()) + + def __str__(self): + return self.report() + + +class UnknownExtra(ResolutionError): + """Distribution doesn't have an "extra feature" of the given name""" + + +_provider_factories = {} + +PY_MAJOR = '{}.{}'.format(*sys.version_info) +EGG_DIST = 3 +BINARY_DIST = 2 +SOURCE_DIST = 1 +CHECKOUT_DIST = 0 +DEVELOP_DIST = -1 + + +def register_loader_type(loader_type, provider_factory): + """Register `provider_factory` to make providers for `loader_type` + + `loader_type` is the type or class of a PEP 302 ``module.__loader__``, + and `provider_factory` is a function that, passed a *module* object, + returns an ``IResourceProvider`` for that module. + """ + _provider_factories[loader_type] = provider_factory + + +def get_provider(moduleOrReq): + """Return an IResourceProvider for the named module or requirement""" + if isinstance(moduleOrReq, Requirement): + return working_set.find(moduleOrReq) or require(str(moduleOrReq))[0] + try: + module = sys.modules[moduleOrReq] + except KeyError: + __import__(moduleOrReq) + module = sys.modules[moduleOrReq] + loader = getattr(module, '__loader__', None) + return _find_adapter(_provider_factories, loader)(module) + + +def _macos_vers(_cache=[]): + if not _cache: + version = platform.mac_ver()[0] + # fallback for MacPorts + if version == '': + plist = '/System/Library/CoreServices/SystemVersion.plist' + if os.path.exists(plist): + if hasattr(plistlib, 'readPlist'): + plist_content = plistlib.readPlist(plist) + if 'ProductVersion' in plist_content: + version = plist_content['ProductVersion'] + + _cache.append(version.split('.')) + return _cache[0] + + +def _macos_arch(machine): + return {'PowerPC': 'ppc', 'Power_Macintosh': 'ppc'}.get(machine, machine) + + +def get_build_platform(): + """Return this platform's string for platform-specific distributions + + XXX Currently this is the same as ``distutils.util.get_platform()``, but it + needs some hacks for Linux and macOS. + """ + from sysconfig import get_platform + + plat = get_platform() + if sys.platform == "darwin" and not plat.startswith('macosx-'): + try: + version = _macos_vers() + machine = os.uname()[4].replace(" ", "_") + return "macosx-%d.%d-%s" % ( + int(version[0]), + int(version[1]), + _macos_arch(machine), + ) + except ValueError: + # if someone is running a non-Mac darwin system, this will fall + # through to the default implementation + pass + return plat + + +macosVersionString = re.compile(r"macosx-(\d+)\.(\d+)-(.*)") +darwinVersionString = re.compile(r"darwin-(\d+)\.(\d+)\.(\d+)-(.*)") +# XXX backward compat +get_platform = get_build_platform + + +def compatible_platforms(provided, required): + """Can code for the `provided` platform run on the `required` platform? + + Returns true if either platform is ``None``, or the platforms are equal. + + XXX Needs compatibility checks for Linux and other unixy OSes. + """ + if provided is None or required is None or provided == required: + # easy case + return True + + # macOS special cases + reqMac = macosVersionString.match(required) + if reqMac: + provMac = macosVersionString.match(provided) + + # is this a Mac package? + if not provMac: + # this is backwards compatibility for packages built before + # setuptools 0.6. All packages built after this point will + # use the new macOS designation. + provDarwin = darwinVersionString.match(provided) + if provDarwin: + dversion = int(provDarwin.group(1)) + macosversion = "%s.%s" % (reqMac.group(1), reqMac.group(2)) + if ( + dversion == 7 + and macosversion >= "10.3" + or dversion == 8 + and macosversion >= "10.4" + ): + return True + # egg isn't macOS or legacy darwin + return False + + # are they the same major version and machine type? + if provMac.group(1) != reqMac.group(1) or provMac.group(3) != reqMac.group(3): + return False + + # is the required OS major update >= the provided one? + if int(provMac.group(2)) > int(reqMac.group(2)): + return False + + return True + + # XXX Linux and other platforms' special cases should go here + return False + + +def run_script(dist_spec, script_name): + """Locate distribution `dist_spec` and run its `script_name` script""" + ns = sys._getframe(1).f_globals + name = ns['__name__'] + ns.clear() + ns['__name__'] = name + require(dist_spec)[0].run_script(script_name, ns) + + +# backward compatibility +run_main = run_script + + +def get_distribution(dist): + """Return a current distribution object for a Requirement or string""" + if isinstance(dist, str): + dist = Requirement.parse(dist) + if isinstance(dist, Requirement): + dist = get_provider(dist) + if not isinstance(dist, Distribution): + raise TypeError("Expected string, Requirement, or Distribution", dist) + return dist + + +def load_entry_point(dist, group, name): + """Return `name` entry point of `group` for `dist` or raise ImportError""" + return get_distribution(dist).load_entry_point(group, name) + + +def get_entry_map(dist, group=None): + """Return the entry point map for `group`, or the full entry map""" + return get_distribution(dist).get_entry_map(group) + + +def get_entry_info(dist, group, name): + """Return the EntryPoint object for `group`+`name`, or ``None``""" + return get_distribution(dist).get_entry_info(group, name) + + +class IMetadataProvider: + def has_metadata(name): + """Does the package's distribution contain the named metadata?""" + + def get_metadata(name): + """The named metadata resource as a string""" + + def get_metadata_lines(name): + """Yield named metadata resource as list of non-blank non-comment lines + + Leading and trailing whitespace is stripped from each line, and lines + with ``#`` as the first non-blank character are omitted.""" + + def metadata_isdir(name): + """Is the named metadata a directory? (like ``os.path.isdir()``)""" + + def metadata_listdir(name): + """List of metadata names in the directory (like ``os.listdir()``)""" + + def run_script(script_name, namespace): + """Execute the named script in the supplied namespace dictionary""" + + +class IResourceProvider(IMetadataProvider): + """An object that provides access to package resources""" + + def get_resource_filename(manager, resource_name): + """Return a true filesystem path for `resource_name` + + `manager` must be an ``IResourceManager``""" + + def get_resource_stream(manager, resource_name): + """Return a readable file-like object for `resource_name` + + `manager` must be an ``IResourceManager``""" + + def get_resource_string(manager, resource_name): + """Return a string containing the contents of `resource_name` + + `manager` must be an ``IResourceManager``""" + + def has_resource(resource_name): + """Does the package contain the named resource?""" + + def resource_isdir(resource_name): + """Is the named resource a directory? (like ``os.path.isdir()``)""" + + def resource_listdir(resource_name): + """List of resource names in the directory (like ``os.listdir()``)""" + + +class WorkingSet: + """A collection of active distributions on sys.path (or a similar list)""" + + def __init__(self, entries=None): + """Create working set from list of path entries (default=sys.path)""" + self.entries = [] + self.entry_keys = {} + self.by_key = {} + self.normalized_to_canonical_keys = {} + self.callbacks = [] + + if entries is None: + entries = sys.path + + for entry in entries: + self.add_entry(entry) + + @classmethod + def _build_master(cls): + """ + Prepare the master working set. + """ + ws = cls() + try: + from __main__ import __requires__ + except ImportError: + # The main program does not list any requirements + return ws + + # ensure the requirements are met + try: + ws.require(__requires__) + except VersionConflict: + return cls._build_from_requirements(__requires__) + + return ws + + @classmethod + def _build_from_requirements(cls, req_spec): + """ + Build a working set from a requirement spec. Rewrites sys.path. + """ + # try it without defaults already on sys.path + # by starting with an empty path + ws = cls([]) + reqs = parse_requirements(req_spec) + dists = ws.resolve(reqs, Environment()) + for dist in dists: + ws.add(dist) + + # add any missing entries from sys.path + for entry in sys.path: + if entry not in ws.entries: + ws.add_entry(entry) + + # then copy back to sys.path + sys.path[:] = ws.entries + return ws + + def add_entry(self, entry): + """Add a path item to ``.entries``, finding any distributions on it + + ``find_distributions(entry, True)`` is used to find distributions + corresponding to the path entry, and they are added. `entry` is + always appended to ``.entries``, even if it is already present. + (This is because ``sys.path`` can contain the same value more than + once, and the ``.entries`` of the ``sys.path`` WorkingSet should always + equal ``sys.path``.) + """ + self.entry_keys.setdefault(entry, []) + self.entries.append(entry) + for dist in find_distributions(entry, True): + self.add(dist, entry, False) + + def __contains__(self, dist): + """True if `dist` is the active distribution for its project""" + return self.by_key.get(dist.key) == dist + + def find(self, req): + """Find a distribution matching requirement `req` + + If there is an active distribution for the requested project, this + returns it as long as it meets the version requirement specified by + `req`. But, if there is an active distribution for the project and it + does *not* meet the `req` requirement, ``VersionConflict`` is raised. + If there is no active distribution for the requested project, ``None`` + is returned. + """ + dist = self.by_key.get(req.key) + + if dist is None: + canonical_key = self.normalized_to_canonical_keys.get(req.key) + + if canonical_key is not None: + req.key = canonical_key + dist = self.by_key.get(canonical_key) + + if dist is not None and dist not in req: + # XXX add more info + raise VersionConflict(dist, req) + return dist + + def iter_entry_points(self, group, name=None): + """Yield entry point objects from `group` matching `name` + + If `name` is None, yields all entry points in `group` from all + distributions in the working set, otherwise only ones matching + both `group` and `name` are yielded (in distribution order). + """ + return ( + entry + for dist in self + for entry in dist.get_entry_map(group).values() + if name is None or name == entry.name + ) + + def run_script(self, requires, script_name): + """Locate distribution for `requires` and run `script_name` script""" + ns = sys._getframe(1).f_globals + name = ns['__name__'] + ns.clear() + ns['__name__'] = name + self.require(requires)[0].run_script(script_name, ns) + + def __iter__(self): + """Yield distributions for non-duplicate projects in the working set + + The yield order is the order in which the items' path entries were + added to the working set. + """ + seen = {} + for item in self.entries: + if item not in self.entry_keys: + # workaround a cache issue + continue + + for key in self.entry_keys[item]: + if key not in seen: + seen[key] = 1 + yield self.by_key[key] + + def add(self, dist, entry=None, insert=True, replace=False): + """Add `dist` to working set, associated with `entry` + + If `entry` is unspecified, it defaults to the ``.location`` of `dist`. + On exit from this routine, `entry` is added to the end of the working + set's ``.entries`` (if it wasn't already present). + + `dist` is only added to the working set if it's for a project that + doesn't already have a distribution in the set, unless `replace=True`. + If it's added, any callbacks registered with the ``subscribe()`` method + will be called. + """ + if insert: + dist.insert_on(self.entries, entry, replace=replace) + + if entry is None: + entry = dist.location + keys = self.entry_keys.setdefault(entry, []) + keys2 = self.entry_keys.setdefault(dist.location, []) + if not replace and dist.key in self.by_key: + # ignore hidden distros + return + + self.by_key[dist.key] = dist + normalized_name = packaging.utils.canonicalize_name(dist.key) + self.normalized_to_canonical_keys[normalized_name] = dist.key + if dist.key not in keys: + keys.append(dist.key) + if dist.key not in keys2: + keys2.append(dist.key) + self._added_new(dist) + + def resolve( + self, + requirements, + env=None, + installer=None, + replace_conflicting=False, + extras=None, + ): + """List all distributions needed to (recursively) meet `requirements` + + `requirements` must be a sequence of ``Requirement`` objects. `env`, + if supplied, should be an ``Environment`` instance. If + not supplied, it defaults to all distributions available within any + entry or distribution in the working set. `installer`, if supplied, + will be invoked with each requirement that cannot be met by an + already-installed distribution; it should return a ``Distribution`` or + ``None``. + + Unless `replace_conflicting=True`, raises a VersionConflict exception + if + any requirements are found on the path that have the correct name but + the wrong version. Otherwise, if an `installer` is supplied it will be + invoked to obtain the correct version of the requirement and activate + it. + + `extras` is a list of the extras to be used with these requirements. + This is important because extra requirements may look like `my_req; + extra = "my_extra"`, which would otherwise be interpreted as a purely + optional requirement. Instead, we want to be able to assert that these + requirements are truly required. + """ + + # set up the stack + requirements = list(requirements)[::-1] + # set of processed requirements + processed = {} + # key -> dist + best = {} + to_activate = [] + + req_extras = _ReqExtras() + + # Mapping of requirement to set of distributions that required it; + # useful for reporting info about conflicts. + required_by = collections.defaultdict(set) + + while requirements: + # process dependencies breadth-first + req = requirements.pop(0) + if req in processed: + # Ignore cyclic or redundant dependencies + continue + + if not req_extras.markers_pass(req, extras): + continue + + dist = self._resolve_dist( + req, best, replace_conflicting, env, installer, required_by, to_activate + ) + + # push the new requirements onto the stack + new_requirements = dist.requires(req.extras)[::-1] + requirements.extend(new_requirements) + + # Register the new requirements needed by req + for new_requirement in new_requirements: + required_by[new_requirement].add(req.project_name) + req_extras[new_requirement] = req.extras + + processed[req] = True + + # return list of distros to activate + return to_activate + + def _resolve_dist( + self, req, best, replace_conflicting, env, installer, required_by, to_activate + ): + dist = best.get(req.key) + if dist is None: + # Find the best distribution and add it to the map + dist = self.by_key.get(req.key) + if dist is None or (dist not in req and replace_conflicting): + ws = self + if env is None: + if dist is None: + env = Environment(self.entries) + else: + # Use an empty environment and workingset to avoid + # any further conflicts with the conflicting + # distribution + env = Environment([]) + ws = WorkingSet([]) + dist = best[req.key] = env.best_match( + req, ws, installer, replace_conflicting=replace_conflicting + ) + if dist is None: + requirers = required_by.get(req, None) + raise DistributionNotFound(req, requirers) + to_activate.append(dist) + if dist not in req: + # Oops, the "best" so far conflicts with a dependency + dependent_req = required_by[req] + raise VersionConflict(dist, req).with_context(dependent_req) + return dist + + def find_plugins(self, plugin_env, full_env=None, installer=None, fallback=True): + """Find all activatable distributions in `plugin_env` + + Example usage:: + + distributions, errors = working_set.find_plugins( + Environment(plugin_dirlist) + ) + # add plugins+libs to sys.path + map(working_set.add, distributions) + # display errors + print('Could not load', errors) + + The `plugin_env` should be an ``Environment`` instance that contains + only distributions that are in the project's "plugin directory" or + directories. The `full_env`, if supplied, should be an ``Environment`` + contains all currently-available distributions. If `full_env` is not + supplied, one is created automatically from the ``WorkingSet`` this + method is called on, which will typically mean that every directory on + ``sys.path`` will be scanned for distributions. + + `installer` is a standard installer callback as used by the + ``resolve()`` method. The `fallback` flag indicates whether we should + attempt to resolve older versions of a plugin if the newest version + cannot be resolved. + + This method returns a 2-tuple: (`distributions`, `error_info`), where + `distributions` is a list of the distributions found in `plugin_env` + that were loadable, along with any other distributions that are needed + to resolve their dependencies. `error_info` is a dictionary mapping + unloadable plugin distributions to an exception instance describing the + error that occurred. Usually this will be a ``DistributionNotFound`` or + ``VersionConflict`` instance. + """ + + plugin_projects = list(plugin_env) + # scan project names in alphabetic order + plugin_projects.sort() + + error_info = {} + distributions = {} + + if full_env is None: + env = Environment(self.entries) + env += plugin_env + else: + env = full_env + plugin_env + + shadow_set = self.__class__([]) + # put all our entries in shadow_set + list(map(shadow_set.add, self)) + + for project_name in plugin_projects: + for dist in plugin_env[project_name]: + req = [dist.as_requirement()] + + try: + resolvees = shadow_set.resolve(req, env, installer) + + except ResolutionError as v: + # save error info + error_info[dist] = v + if fallback: + # try the next older version of project + continue + else: + # give up on this project, keep going + break + + else: + list(map(shadow_set.add, resolvees)) + distributions.update(dict.fromkeys(resolvees)) + + # success, no need to try any more versions of this project + break + + distributions = list(distributions) + distributions.sort() + + return distributions, error_info + + def require(self, *requirements): + """Ensure that distributions matching `requirements` are activated + + `requirements` must be a string or a (possibly-nested) sequence + thereof, specifying the distributions and versions required. The + return value is a sequence of the distributions that needed to be + activated to fulfill the requirements; all relevant distributions are + included, even if they were already activated in this working set. + """ + needed = self.resolve(parse_requirements(requirements)) + + for dist in needed: + self.add(dist) + + return needed + + def subscribe(self, callback, existing=True): + """Invoke `callback` for all distributions + + If `existing=True` (default), + call on all existing ones, as well. + """ + if callback in self.callbacks: + return + self.callbacks.append(callback) + if not existing: + return + for dist in self: + callback(dist) + + def _added_new(self, dist): + for callback in self.callbacks: + callback(dist) + + def __getstate__(self): + return ( + self.entries[:], + self.entry_keys.copy(), + self.by_key.copy(), + self.normalized_to_canonical_keys.copy(), + self.callbacks[:], + ) + + def __setstate__(self, e_k_b_n_c): + entries, keys, by_key, normalized_to_canonical_keys, callbacks = e_k_b_n_c + self.entries = entries[:] + self.entry_keys = keys.copy() + self.by_key = by_key.copy() + self.normalized_to_canonical_keys = normalized_to_canonical_keys.copy() + self.callbacks = callbacks[:] + + +class _ReqExtras(dict): + """ + Map each requirement to the extras that demanded it. + """ + + def markers_pass(self, req, extras=None): + """ + Evaluate markers for req against each extra that + demanded it. + + Return False if the req has a marker and fails + evaluation. Otherwise, return True. + """ + extra_evals = ( + req.marker.evaluate({'extra': extra}) + for extra in self.get(req, ()) + (extras or (None,)) + ) + return not req.marker or any(extra_evals) + + +class Environment: + """Searchable snapshot of distributions on a search path""" + + def __init__( + self, search_path=None, platform=get_supported_platform(), python=PY_MAJOR + ): + """Snapshot distributions available on a search path + + Any distributions found on `search_path` are added to the environment. + `search_path` should be a sequence of ``sys.path`` items. If not + supplied, ``sys.path`` is used. + + `platform` is an optional string specifying the name of the platform + that platform-specific distributions must be compatible with. If + unspecified, it defaults to the current platform. `python` is an + optional string naming the desired version of Python (e.g. ``'3.6'``); + it defaults to the current version. + + You may explicitly set `platform` (and/or `python`) to ``None`` if you + wish to map *all* distributions, not just those compatible with the + running platform or Python version. + """ + self._distmap = {} + self.platform = platform + self.python = python + self.scan(search_path) + + def can_add(self, dist): + """Is distribution `dist` acceptable for this environment? + + The distribution must match the platform and python version + requirements specified when this environment was created, or False + is returned. + """ + py_compat = ( + self.python is None + or dist.py_version is None + or dist.py_version == self.python + ) + return py_compat and compatible_platforms(dist.platform, self.platform) + + def remove(self, dist): + """Remove `dist` from the environment""" + self._distmap[dist.key].remove(dist) + + def scan(self, search_path=None): + """Scan `search_path` for distributions usable in this environment + + Any distributions found are added to the environment. + `search_path` should be a sequence of ``sys.path`` items. If not + supplied, ``sys.path`` is used. Only distributions conforming to + the platform/python version defined at initialization are added. + """ + if search_path is None: + search_path = sys.path + + for item in search_path: + for dist in find_distributions(item): + self.add(dist) + + def __getitem__(self, project_name): + """Return a newest-to-oldest list of distributions for `project_name` + + Uses case-insensitive `project_name` comparison, assuming all the + project's distributions use their project's name converted to all + lowercase as their key. + + """ + distribution_key = project_name.lower() + return self._distmap.get(distribution_key, []) + + def add(self, dist): + """Add `dist` if we ``can_add()`` it and it has not already been added""" + if self.can_add(dist) and dist.has_version(): + dists = self._distmap.setdefault(dist.key, []) + if dist not in dists: + dists.append(dist) + dists.sort(key=operator.attrgetter('hashcmp'), reverse=True) + + def best_match(self, req, working_set, installer=None, replace_conflicting=False): + """Find distribution best matching `req` and usable on `working_set` + + This calls the ``find(req)`` method of the `working_set` to see if a + suitable distribution is already active. (This may raise + ``VersionConflict`` if an unsuitable version of the project is already + active in the specified `working_set`.) If a suitable distribution + isn't active, this method returns the newest distribution in the + environment that meets the ``Requirement`` in `req`. If no suitable + distribution is found, and `installer` is supplied, then the result of + calling the environment's ``obtain(req, installer)`` method will be + returned. + """ + try: + dist = working_set.find(req) + except VersionConflict: + if not replace_conflicting: + raise + dist = None + if dist is not None: + return dist + for dist in self[req.key]: + if dist in req: + return dist + # try to download/install + return self.obtain(req, installer) + + def obtain(self, requirement, installer=None): + """Obtain a distribution matching `requirement` (e.g. via download) + + Obtain a distro that matches requirement (e.g. via download). In the + base ``Environment`` class, this routine just returns + ``installer(requirement)``, unless `installer` is None, in which case + None is returned instead. This method is a hook that allows subclasses + to attempt other ways of obtaining a distribution before falling back + to the `installer` argument.""" + if installer is not None: + return installer(requirement) + + def __iter__(self): + """Yield the unique project names of the available distributions""" + for key in self._distmap.keys(): + if self[key]: + yield key + + def __iadd__(self, other): + """In-place addition of a distribution or environment""" + if isinstance(other, Distribution): + self.add(other) + elif isinstance(other, Environment): + for project in other: + for dist in other[project]: + self.add(dist) + else: + raise TypeError("Can't add %r to environment" % (other,)) + return self + + def __add__(self, other): + """Add an environment or distribution to an environment""" + new = self.__class__([], platform=None, python=None) + for env in self, other: + new += env + return new + + +# XXX backward compatibility +AvailableDistributions = Environment + + +class ExtractionError(RuntimeError): + """An error occurred extracting a resource + + The following attributes are available from instances of this exception: + + manager + The resource manager that raised this exception + + cache_path + The base directory for resource extraction + + original_error + The exception instance that caused extraction to fail + """ + + +class ResourceManager: + """Manage resource extraction and packages""" + + extraction_path = None + + def __init__(self): + self.cached_files = {} + + def resource_exists(self, package_or_requirement, resource_name): + """Does the named resource exist?""" + return get_provider(package_or_requirement).has_resource(resource_name) + + def resource_isdir(self, package_or_requirement, resource_name): + """Is the named resource an existing directory?""" + return get_provider(package_or_requirement).resource_isdir(resource_name) + + def resource_filename(self, package_or_requirement, resource_name): + """Return a true filesystem path for specified resource""" + return get_provider(package_or_requirement).get_resource_filename( + self, resource_name + ) + + def resource_stream(self, package_or_requirement, resource_name): + """Return a readable file-like object for specified resource""" + return get_provider(package_or_requirement).get_resource_stream( + self, resource_name + ) + + def resource_string(self, package_or_requirement, resource_name): + """Return specified resource as a string""" + return get_provider(package_or_requirement).get_resource_string( + self, resource_name + ) + + def resource_listdir(self, package_or_requirement, resource_name): + """List the contents of the named resource directory""" + return get_provider(package_or_requirement).resource_listdir(resource_name) + + def extraction_error(self): + """Give an error message for problems extracting file(s)""" + + old_exc = sys.exc_info()[1] + cache_path = self.extraction_path or get_default_cache() + + tmpl = textwrap.dedent( + """ + Can't extract file(s) to egg cache + + The following error occurred while trying to extract file(s) + to the Python egg cache: + + {old_exc} + + The Python egg cache directory is currently set to: + + {cache_path} + + Perhaps your account does not have write access to this directory? + You can change the cache directory by setting the PYTHON_EGG_CACHE + environment variable to point to an accessible directory. + """ + ).lstrip() + err = ExtractionError(tmpl.format(**locals())) + err.manager = self + err.cache_path = cache_path + err.original_error = old_exc + raise err + + def get_cache_path(self, archive_name, names=()): + """Return absolute location in cache for `archive_name` and `names` + + The parent directory of the resulting path will be created if it does + not already exist. `archive_name` should be the base filename of the + enclosing egg (which may not be the name of the enclosing zipfile!), + including its ".egg" extension. `names`, if provided, should be a + sequence of path name parts "under" the egg's extraction location. + + This method should only be called by resource providers that need to + obtain an extraction location, and only for names they intend to + extract, as it tracks the generated names for possible cleanup later. + """ + extract_path = self.extraction_path or get_default_cache() + target_path = os.path.join(extract_path, archive_name + '-tmp', *names) + try: + _bypass_ensure_directory(target_path) + except Exception: + self.extraction_error() + + self._warn_unsafe_extraction_path(extract_path) + + self.cached_files[target_path] = 1 + return target_path + + @staticmethod + def _warn_unsafe_extraction_path(path): + """ + If the default extraction path is overridden and set to an insecure + location, such as /tmp, it opens up an opportunity for an attacker to + replace an extracted file with an unauthorized payload. Warn the user + if a known insecure location is used. + + See Distribute #375 for more details. + """ + if os.name == 'nt' and not path.startswith(os.environ['windir']): + # On Windows, permissions are generally restrictive by default + # and temp directories are not writable by other users, so + # bypass the warning. + return + mode = os.stat(path).st_mode + if mode & stat.S_IWOTH or mode & stat.S_IWGRP: + msg = ( + "Extraction path is writable by group/others " + "and vulnerable to attack when " + "used with get_resource_filename ({path}). " + "Consider a more secure " + "location (set with .set_extraction_path or the " + "PYTHON_EGG_CACHE environment variable)." + ).format(**locals()) + warnings.warn(msg, UserWarning) + + def postprocess(self, tempname, filename): + """Perform any platform-specific postprocessing of `tempname` + + This is where Mac header rewrites should be done; other platforms don't + have anything special they should do. + + Resource providers should call this method ONLY after successfully + extracting a compressed resource. They must NOT call it on resources + that are already in the filesystem. + + `tempname` is the current (temporary) name of the file, and `filename` + is the name it will be renamed to by the caller after this routine + returns. + """ + + if os.name == 'posix': + # Make the resource executable + mode = ((os.stat(tempname).st_mode) | 0o555) & 0o7777 + os.chmod(tempname, mode) + + def set_extraction_path(self, path): + """Set the base path where resources will be extracted to, if needed. + + If you do not call this routine before any extractions take place, the + path defaults to the return value of ``get_default_cache()``. (Which + is based on the ``PYTHON_EGG_CACHE`` environment variable, with various + platform-specific fallbacks. See that routine's documentation for more + details.) + + Resources are extracted to subdirectories of this path based upon + information given by the ``IResourceProvider``. You may set this to a + temporary directory, but then you must call ``cleanup_resources()`` to + delete the extracted files when done. There is no guarantee that + ``cleanup_resources()`` will be able to remove all extracted files. + + (Note: you may not change the extraction path for a given resource + manager once resources have been extracted, unless you first call + ``cleanup_resources()``.) + """ + if self.cached_files: + raise ValueError("Can't change extraction path, files already extracted") + + self.extraction_path = path + + def cleanup_resources(self, force=False): + """ + Delete all extracted resource files and directories, returning a list + of the file and directory names that could not be successfully removed. + This function does not have any concurrency protection, so it should + generally only be called when the extraction path is a temporary + directory exclusive to a single process. This method is not + automatically called; you must call it explicitly or register it as an + ``atexit`` function if you wish to ensure cleanup of a temporary + directory used for extractions. + """ + # XXX + + +def get_default_cache(): + """ + Return the ``PYTHON_EGG_CACHE`` environment variable + or a platform-relevant user cache dir for an app + named "Python-Eggs". + """ + return os.environ.get('PYTHON_EGG_CACHE') or platformdirs.user_cache_dir( + appname='Python-Eggs' + ) + + +def safe_name(name): + """Convert an arbitrary string to a standard distribution name + + Any runs of non-alphanumeric/. characters are replaced with a single '-'. + """ + return re.sub('[^A-Za-z0-9.]+', '-', name) + + +def safe_version(version): + """ + Convert an arbitrary string to a standard version string + """ + try: + # normalize the version + return str(packaging.version.Version(version)) + except packaging.version.InvalidVersion: + version = version.replace(' ', '.') + return re.sub('[^A-Za-z0-9.]+', '-', version) + + +def _forgiving_version(version): + """Fallback when ``safe_version`` is not safe enough + >>> parse_version(_forgiving_version('0.23ubuntu1')) + <Version('0.23.dev0+sanitized.ubuntu1')> + >>> parse_version(_forgiving_version('0.23-')) + <Version('0.23.dev0+sanitized')> + >>> parse_version(_forgiving_version('0.-_')) + <Version('0.dev0+sanitized')> + >>> parse_version(_forgiving_version('42.+?1')) + <Version('42.dev0+sanitized.1')> + >>> parse_version(_forgiving_version('hello world')) + <Version('0.dev0+sanitized.hello.world')> + """ + version = version.replace(' ', '.') + match = _PEP440_FALLBACK.search(version) + if match: + safe = match["safe"] + rest = version[len(safe):] + else: + safe = "0" + rest = version + local = f"sanitized.{_safe_segment(rest)}".strip(".") + return f"{safe}.dev0+{local}" + + +def _safe_segment(segment): + """Convert an arbitrary string into a safe segment""" + segment = re.sub('[^A-Za-z0-9.]+', '-', segment) + segment = re.sub('-[^A-Za-z0-9]+', '-', segment) + return re.sub(r'\.[^A-Za-z0-9]+', '.', segment).strip(".-") + + +def safe_extra(extra): + """Convert an arbitrary string to a standard 'extra' name + + Any runs of non-alphanumeric characters are replaced with a single '_', + and the result is always lowercased. + """ + return re.sub('[^A-Za-z0-9.-]+', '_', extra).lower() + + +def to_filename(name): + """Convert a project or version name to its filename-escaped form + + Any '-' characters are currently replaced with '_'. + """ + return name.replace('-', '_') + + +def invalid_marker(text): + """ + Validate text as a PEP 508 environment marker; return an exception + if invalid or False otherwise. + """ + try: + evaluate_marker(text) + except SyntaxError as e: + e.filename = None + e.lineno = None + return e + return False + + +def evaluate_marker(text, extra=None): + """ + Evaluate a PEP 508 environment marker. + Return a boolean indicating the marker result in this environment. + Raise SyntaxError if marker is invalid. + + This implementation uses the 'pyparsing' module. + """ + try: + marker = packaging.markers.Marker(text) + return marker.evaluate() + except packaging.markers.InvalidMarker as e: + raise SyntaxError(e) from e + + +class NullProvider: + """Try to implement resources and metadata for arbitrary PEP 302 loaders""" + + egg_name = None + egg_info = None + loader = None + + def __init__(self, module): + self.loader = getattr(module, '__loader__', None) + self.module_path = os.path.dirname(getattr(module, '__file__', '')) + + def get_resource_filename(self, manager, resource_name): + return self._fn(self.module_path, resource_name) + + def get_resource_stream(self, manager, resource_name): + return io.BytesIO(self.get_resource_string(manager, resource_name)) + + def get_resource_string(self, manager, resource_name): + return self._get(self._fn(self.module_path, resource_name)) + + def has_resource(self, resource_name): + return self._has(self._fn(self.module_path, resource_name)) + + def _get_metadata_path(self, name): + return self._fn(self.egg_info, name) + + def has_metadata(self, name): + if not self.egg_info: + return self.egg_info + + path = self._get_metadata_path(name) + return self._has(path) + + def get_metadata(self, name): + if not self.egg_info: + return "" + path = self._get_metadata_path(name) + value = self._get(path) + try: + return value.decode('utf-8') + except UnicodeDecodeError as exc: + # Include the path in the error message to simplify + # troubleshooting, and without changing the exception type. + exc.reason += ' in {} file at path: {}'.format(name, path) + raise + + def get_metadata_lines(self, name): + return yield_lines(self.get_metadata(name)) + + def resource_isdir(self, resource_name): + return self._isdir(self._fn(self.module_path, resource_name)) + + def metadata_isdir(self, name): + return self.egg_info and self._isdir(self._fn(self.egg_info, name)) + + def resource_listdir(self, resource_name): + return self._listdir(self._fn(self.module_path, resource_name)) + + def metadata_listdir(self, name): + if self.egg_info: + return self._listdir(self._fn(self.egg_info, name)) + return [] + + def run_script(self, script_name, namespace): + script = 'scripts/' + script_name + if not self.has_metadata(script): + raise ResolutionError( + "Script {script!r} not found in metadata at {self.egg_info!r}".format( + **locals() + ), + ) + script_text = self.get_metadata(script).replace('\r\n', '\n') + script_text = script_text.replace('\r', '\n') + script_filename = self._fn(self.egg_info, script) + namespace['__file__'] = script_filename + if os.path.exists(script_filename): + with open(script_filename) as fid: + source = fid.read() + code = compile(source, script_filename, 'exec') + exec(code, namespace, namespace) + else: + from linecache import cache + + cache[script_filename] = ( + len(script_text), + 0, + script_text.split('\n'), + script_filename, + ) + script_code = compile(script_text, script_filename, 'exec') + exec(script_code, namespace, namespace) + + def _has(self, path): + raise NotImplementedError( + "Can't perform this operation for unregistered loader type" + ) + + def _isdir(self, path): + raise NotImplementedError( + "Can't perform this operation for unregistered loader type" + ) + + def _listdir(self, path): + raise NotImplementedError( + "Can't perform this operation for unregistered loader type" + ) + + def _fn(self, base, resource_name): + self._validate_resource_path(resource_name) + if resource_name: + return os.path.join(base, *resource_name.split('/')) + return base + + @staticmethod + def _validate_resource_path(path): + """ + Validate the resource paths according to the docs. + https://setuptools.pypa.io/en/latest/pkg_resources.html#basic-resource-access + + >>> warned = getfixture('recwarn') + >>> warnings.simplefilter('always') + >>> vrp = NullProvider._validate_resource_path + >>> vrp('foo/bar.txt') + >>> bool(warned) + False + >>> vrp('../foo/bar.txt') + >>> bool(warned) + True + >>> warned.clear() + >>> vrp('/foo/bar.txt') + >>> bool(warned) + True + >>> vrp('foo/../../bar.txt') + >>> bool(warned) + True + >>> warned.clear() + >>> vrp('foo/f../bar.txt') + >>> bool(warned) + False + + Windows path separators are straight-up disallowed. + >>> vrp(r'\\foo/bar.txt') + Traceback (most recent call last): + ... + ValueError: Use of .. or absolute path in a resource path \ +is not allowed. + + >>> vrp(r'C:\\foo/bar.txt') + Traceback (most recent call last): + ... + ValueError: Use of .. or absolute path in a resource path \ +is not allowed. + + Blank values are allowed + + >>> vrp('') + >>> bool(warned) + False + + Non-string values are not. + + >>> vrp(None) + Traceback (most recent call last): + ... + AttributeError: ... + """ + invalid = ( + os.path.pardir in path.split(posixpath.sep) + or posixpath.isabs(path) + or ntpath.isabs(path) + ) + if not invalid: + return + + msg = "Use of .. or absolute path in a resource path is not allowed." + + # Aggressively disallow Windows absolute paths + if ntpath.isabs(path) and not posixpath.isabs(path): + raise ValueError(msg) + + # for compatibility, warn; in future + # raise ValueError(msg) + issue_warning( + msg[:-1] + " and will raise exceptions in a future release.", + DeprecationWarning, + ) + + def _get(self, path): + if hasattr(self.loader, 'get_data'): + return self.loader.get_data(path) + raise NotImplementedError( + "Can't perform this operation for loaders without 'get_data()'" + ) + + +register_loader_type(object, NullProvider) + + +def _parents(path): + """ + yield all parents of path including path + """ + last = None + while path != last: + yield path + last = path + path, _ = os.path.split(path) + + +class EggProvider(NullProvider): + """Provider based on a virtual filesystem""" + + def __init__(self, module): + super().__init__(module) + self._setup_prefix() + + def _setup_prefix(self): + # Assume that metadata may be nested inside a "basket" + # of multiple eggs and use module_path instead of .archive. + eggs = filter(_is_egg_path, _parents(self.module_path)) + egg = next(eggs, None) + egg and self._set_egg(egg) + + def _set_egg(self, path): + self.egg_name = os.path.basename(path) + self.egg_info = os.path.join(path, 'EGG-INFO') + self.egg_root = path + + +class DefaultProvider(EggProvider): + """Provides access to package resources in the filesystem""" + + def _has(self, path): + return os.path.exists(path) + + def _isdir(self, path): + return os.path.isdir(path) + + def _listdir(self, path): + return os.listdir(path) + + def get_resource_stream(self, manager, resource_name): + return open(self._fn(self.module_path, resource_name), 'rb') + + def _get(self, path): + with open(path, 'rb') as stream: + return stream.read() + + @classmethod + def _register(cls): + loader_names = ( + 'SourceFileLoader', + 'SourcelessFileLoader', + ) + for name in loader_names: + loader_cls = getattr(importlib_machinery, name, type(None)) + register_loader_type(loader_cls, cls) + + +DefaultProvider._register() + + +class EmptyProvider(NullProvider): + """Provider that returns nothing for all requests""" + + module_path = None + + _isdir = _has = lambda self, path: False + + def _get(self, path): + return '' + + def _listdir(self, path): + return [] + + def __init__(self): + pass + + +empty_provider = EmptyProvider() + + +class ZipManifests(dict): + """ + zip manifest builder + """ + + @classmethod + def build(cls, path): + """ + Build a dictionary similar to the zipimport directory + caches, except instead of tuples, store ZipInfo objects. + + Use a platform-specific path separator (os.sep) for the path keys + for compatibility with pypy on Windows. + """ + with zipfile.ZipFile(path) as zfile: + items = ( + ( + name.replace('/', os.sep), + zfile.getinfo(name), + ) + for name in zfile.namelist() + ) + return dict(items) + + load = build + + +class MemoizedZipManifests(ZipManifests): + """ + Memoized zipfile manifests. + """ + + manifest_mod = collections.namedtuple('manifest_mod', 'manifest mtime') + + def load(self, path): + """ + Load a manifest at path or return a suitable manifest already loaded. + """ + path = os.path.normpath(path) + mtime = os.stat(path).st_mtime + + if path not in self or self[path].mtime != mtime: + manifest = self.build(path) + self[path] = self.manifest_mod(manifest, mtime) + + return self[path].manifest + + +class ZipProvider(EggProvider): + """Resource support for zips and eggs""" + + eagers = None + _zip_manifests = MemoizedZipManifests() + + def __init__(self, module): + super().__init__(module) + self.zip_pre = self.loader.archive + os.sep + + def _zipinfo_name(self, fspath): + # Convert a virtual filename (full path to file) into a zipfile subpath + # usable with the zipimport directory cache for our target archive + fspath = fspath.rstrip(os.sep) + if fspath == self.loader.archive: + return '' + if fspath.startswith(self.zip_pre): + return fspath[len(self.zip_pre) :] + raise AssertionError("%s is not a subpath of %s" % (fspath, self.zip_pre)) + + def _parts(self, zip_path): + # Convert a zipfile subpath into an egg-relative path part list. + # pseudo-fs path + fspath = self.zip_pre + zip_path + if fspath.startswith(self.egg_root + os.sep): + return fspath[len(self.egg_root) + 1 :].split(os.sep) + raise AssertionError("%s is not a subpath of %s" % (fspath, self.egg_root)) + + @property + def zipinfo(self): + return self._zip_manifests.load(self.loader.archive) + + def get_resource_filename(self, manager, resource_name): + if not self.egg_name: + raise NotImplementedError( + "resource_filename() only supported for .egg, not .zip" + ) + # no need to lock for extraction, since we use temp names + zip_path = self._resource_to_zip(resource_name) + eagers = self._get_eager_resources() + if '/'.join(self._parts(zip_path)) in eagers: + for name in eagers: + self._extract_resource(manager, self._eager_to_zip(name)) + return self._extract_resource(manager, zip_path) + + @staticmethod + def _get_date_and_size(zip_stat): + size = zip_stat.file_size + # ymdhms+wday, yday, dst + date_time = zip_stat.date_time + (0, 0, -1) + # 1980 offset already done + timestamp = time.mktime(date_time) + return timestamp, size + + # FIXME: 'ZipProvider._extract_resource' is too complex (12) + def _extract_resource(self, manager, zip_path): # noqa: C901 + if zip_path in self._index(): + for name in self._index()[zip_path]: + last = self._extract_resource(manager, os.path.join(zip_path, name)) + # return the extracted directory name + return os.path.dirname(last) + + timestamp, size = self._get_date_and_size(self.zipinfo[zip_path]) + + if not WRITE_SUPPORT: + raise IOError( + '"os.rename" and "os.unlink" are not supported ' 'on this platform' + ) + try: + real_path = manager.get_cache_path(self.egg_name, self._parts(zip_path)) + + if self._is_current(real_path, zip_path): + return real_path + + outf, tmpnam = _mkstemp( + ".$extract", + dir=os.path.dirname(real_path), + ) + os.write(outf, self.loader.get_data(zip_path)) + os.close(outf) + utime(tmpnam, (timestamp, timestamp)) + manager.postprocess(tmpnam, real_path) + + try: + rename(tmpnam, real_path) + + except os.error: + if os.path.isfile(real_path): + if self._is_current(real_path, zip_path): + # the file became current since it was checked above, + # so proceed. + return real_path + # Windows, del old file and retry + elif os.name == 'nt': + unlink(real_path) + rename(tmpnam, real_path) + return real_path + raise + + except os.error: + # report a user-friendly error + manager.extraction_error() + + return real_path + + def _is_current(self, file_path, zip_path): + """ + Return True if the file_path is current for this zip_path + """ + timestamp, size = self._get_date_and_size(self.zipinfo[zip_path]) + if not os.path.isfile(file_path): + return False + stat = os.stat(file_path) + if stat.st_size != size or stat.st_mtime != timestamp: + return False + # check that the contents match + zip_contents = self.loader.get_data(zip_path) + with open(file_path, 'rb') as f: + file_contents = f.read() + return zip_contents == file_contents + + def _get_eager_resources(self): + if self.eagers is None: + eagers = [] + for name in ('native_libs.txt', 'eager_resources.txt'): + if self.has_metadata(name): + eagers.extend(self.get_metadata_lines(name)) + self.eagers = eagers + return self.eagers + + def _index(self): + try: + return self._dirindex + except AttributeError: + ind = {} + for path in self.zipinfo: + parts = path.split(os.sep) + while parts: + parent = os.sep.join(parts[:-1]) + if parent in ind: + ind[parent].append(parts[-1]) + break + else: + ind[parent] = [parts.pop()] + self._dirindex = ind + return ind + + def _has(self, fspath): + zip_path = self._zipinfo_name(fspath) + return zip_path in self.zipinfo or zip_path in self._index() + + def _isdir(self, fspath): + return self._zipinfo_name(fspath) in self._index() + + def _listdir(self, fspath): + return list(self._index().get(self._zipinfo_name(fspath), ())) + + def _eager_to_zip(self, resource_name): + return self._zipinfo_name(self._fn(self.egg_root, resource_name)) + + def _resource_to_zip(self, resource_name): + return self._zipinfo_name(self._fn(self.module_path, resource_name)) + + +register_loader_type(zipimport.zipimporter, ZipProvider) + + +class FileMetadata(EmptyProvider): + """Metadata handler for standalone PKG-INFO files + + Usage:: + + metadata = FileMetadata("/path/to/PKG-INFO") + + This provider rejects all data and metadata requests except for PKG-INFO, + which is treated as existing, and will be the contents of the file at + the provided location. + """ + + def __init__(self, path): + self.path = path + + def _get_metadata_path(self, name): + return self.path + + def has_metadata(self, name): + return name == 'PKG-INFO' and os.path.isfile(self.path) + + def get_metadata(self, name): + if name != 'PKG-INFO': + raise KeyError("No metadata except PKG-INFO is available") + + with io.open(self.path, encoding='utf-8', errors="replace") as f: + metadata = f.read() + self._warn_on_replacement(metadata) + return metadata + + def _warn_on_replacement(self, metadata): + replacement_char = '�' + if replacement_char in metadata: + tmpl = "{self.path} could not be properly decoded in UTF-8" + msg = tmpl.format(**locals()) + warnings.warn(msg) + + def get_metadata_lines(self, name): + return yield_lines(self.get_metadata(name)) + + +class PathMetadata(DefaultProvider): + """Metadata provider for egg directories + + Usage:: + + # Development eggs: + + egg_info = "/path/to/PackageName.egg-info" + base_dir = os.path.dirname(egg_info) + metadata = PathMetadata(base_dir, egg_info) + dist_name = os.path.splitext(os.path.basename(egg_info))[0] + dist = Distribution(basedir, project_name=dist_name, metadata=metadata) + + # Unpacked egg directories: + + egg_path = "/path/to/PackageName-ver-pyver-etc.egg" + metadata = PathMetadata(egg_path, os.path.join(egg_path,'EGG-INFO')) + dist = Distribution.from_filename(egg_path, metadata=metadata) + """ + + def __init__(self, path, egg_info): + self.module_path = path + self.egg_info = egg_info + + +class EggMetadata(ZipProvider): + """Metadata provider for .egg files""" + + def __init__(self, importer): + """Create a metadata provider from a zipimporter""" + + self.zip_pre = importer.archive + os.sep + self.loader = importer + if importer.prefix: + self.module_path = os.path.join(importer.archive, importer.prefix) + else: + self.module_path = importer.archive + self._setup_prefix() + + +_declare_state('dict', _distribution_finders={}) + + +def register_finder(importer_type, distribution_finder): + """Register `distribution_finder` to find distributions in sys.path items + + `importer_type` is the type or class of a PEP 302 "Importer" (sys.path item + handler), and `distribution_finder` is a callable that, passed a path + item and the importer instance, yields ``Distribution`` instances found on + that path item. See ``pkg_resources.find_on_path`` for an example.""" + _distribution_finders[importer_type] = distribution_finder + + +def find_distributions(path_item, only=False): + """Yield distributions accessible via `path_item`""" + importer = get_importer(path_item) + finder = _find_adapter(_distribution_finders, importer) + return finder(importer, path_item, only) + + +def find_eggs_in_zip(importer, path_item, only=False): + """ + Find eggs in zip files; possibly multiple nested eggs. + """ + if importer.archive.endswith('.whl'): + # wheels are not supported with this finder + # they don't have PKG-INFO metadata, and won't ever contain eggs + return + metadata = EggMetadata(importer) + if metadata.has_metadata('PKG-INFO'): + yield Distribution.from_filename(path_item, metadata=metadata) + if only: + # don't yield nested distros + return + for subitem in metadata.resource_listdir(''): + if _is_egg_path(subitem): + subpath = os.path.join(path_item, subitem) + dists = find_eggs_in_zip(zipimport.zipimporter(subpath), subpath) + for dist in dists: + yield dist + elif subitem.lower().endswith(('.dist-info', '.egg-info')): + subpath = os.path.join(path_item, subitem) + submeta = EggMetadata(zipimport.zipimporter(subpath)) + submeta.egg_info = subpath + yield Distribution.from_location(path_item, subitem, submeta) + + +register_finder(zipimport.zipimporter, find_eggs_in_zip) + + +def find_nothing(importer, path_item, only=False): + return () + + +register_finder(object, find_nothing) + + +def find_on_path(importer, path_item, only=False): + """Yield distributions accessible on a sys.path directory""" + path_item = _normalize_cached(path_item) + + if _is_unpacked_egg(path_item): + yield Distribution.from_filename( + path_item, + metadata=PathMetadata(path_item, os.path.join(path_item, 'EGG-INFO')), + ) + return + + entries = (os.path.join(path_item, child) for child in safe_listdir(path_item)) + + # scan for .egg and .egg-info in directory + for entry in sorted(entries): + fullpath = os.path.join(path_item, entry) + factory = dist_factory(path_item, entry, only) + for dist in factory(fullpath): + yield dist + + +def dist_factory(path_item, entry, only): + """Return a dist_factory for the given entry.""" + lower = entry.lower() + is_egg_info = lower.endswith('.egg-info') + is_dist_info = lower.endswith('.dist-info') and os.path.isdir( + os.path.join(path_item, entry) + ) + is_meta = is_egg_info or is_dist_info + return ( + distributions_from_metadata + if is_meta + else find_distributions + if not only and _is_egg_path(entry) + else resolve_egg_link + if not only and lower.endswith('.egg-link') + else NoDists() + ) + + +class NoDists: + """ + >>> bool(NoDists()) + False + + >>> list(NoDists()('anything')) + [] + """ + + def __bool__(self): + return False + + def __call__(self, fullpath): + return iter(()) + + +def safe_listdir(path): + """ + Attempt to list contents of path, but suppress some exceptions. + """ + try: + return os.listdir(path) + except (PermissionError, NotADirectoryError): + pass + except OSError as e: + # Ignore the directory if does not exist, not a directory or + # permission denied + if e.errno not in (errno.ENOTDIR, errno.EACCES, errno.ENOENT): + raise + return () + + +def distributions_from_metadata(path): + root = os.path.dirname(path) + if os.path.isdir(path): + if len(os.listdir(path)) == 0: + # empty metadata dir; skip + return + metadata = PathMetadata(root, path) + else: + metadata = FileMetadata(path) + entry = os.path.basename(path) + yield Distribution.from_location( + root, + entry, + metadata, + precedence=DEVELOP_DIST, + ) + + +def non_empty_lines(path): + """ + Yield non-empty lines from file at path + """ + with open(path) as f: + for line in f: + line = line.strip() + if line: + yield line + + +def resolve_egg_link(path): + """ + Given a path to an .egg-link, resolve distributions + present in the referenced path. + """ + referenced_paths = non_empty_lines(path) + resolved_paths = ( + os.path.join(os.path.dirname(path), ref) for ref in referenced_paths + ) + dist_groups = map(find_distributions, resolved_paths) + return next(dist_groups, ()) + + +if hasattr(pkgutil, 'ImpImporter'): + register_finder(pkgutil.ImpImporter, find_on_path) + +register_finder(importlib_machinery.FileFinder, find_on_path) + +_declare_state('dict', _namespace_handlers={}) +_declare_state('dict', _namespace_packages={}) + + +def register_namespace_handler(importer_type, namespace_handler): + """Register `namespace_handler` to declare namespace packages + + `importer_type` is the type or class of a PEP 302 "Importer" (sys.path item + handler), and `namespace_handler` is a callable like this:: + + def namespace_handler(importer, path_entry, moduleName, module): + # return a path_entry to use for child packages + + Namespace handlers are only called if the importer object has already + agreed that it can handle the relevant path item, and they should only + return a subpath if the module __path__ does not already contain an + equivalent subpath. For an example namespace handler, see + ``pkg_resources.file_ns_handler``. + """ + _namespace_handlers[importer_type] = namespace_handler + + +def _handle_ns(packageName, path_item): + """Ensure that named package includes a subpath of path_item (if needed)""" + + importer = get_importer(path_item) + if importer is None: + return None + + # use find_spec (PEP 451) and fall-back to find_module (PEP 302) + try: + spec = importer.find_spec(packageName) + except AttributeError: + # capture warnings due to #1111 + with warnings.catch_warnings(): + warnings.simplefilter("ignore") + loader = importer.find_module(packageName) + else: + loader = spec.loader if spec else None + + if loader is None: + return None + module = sys.modules.get(packageName) + if module is None: + module = sys.modules[packageName] = types.ModuleType(packageName) + module.__path__ = [] + _set_parent_ns(packageName) + elif not hasattr(module, '__path__'): + raise TypeError("Not a package:", packageName) + handler = _find_adapter(_namespace_handlers, importer) + subpath = handler(importer, path_item, packageName, module) + if subpath is not None: + path = module.__path__ + path.append(subpath) + importlib.import_module(packageName) + _rebuild_mod_path(path, packageName, module) + return subpath + + +def _rebuild_mod_path(orig_path, package_name, module): + """ + Rebuild module.__path__ ensuring that all entries are ordered + corresponding to their sys.path order + """ + sys_path = [_normalize_cached(p) for p in sys.path] + + def safe_sys_path_index(entry): + """ + Workaround for #520 and #513. + """ + try: + return sys_path.index(entry) + except ValueError: + return float('inf') + + def position_in_sys_path(path): + """ + Return the ordinal of the path based on its position in sys.path + """ + path_parts = path.split(os.sep) + module_parts = package_name.count('.') + 1 + parts = path_parts[:-module_parts] + return safe_sys_path_index(_normalize_cached(os.sep.join(parts))) + + new_path = sorted(orig_path, key=position_in_sys_path) + new_path = [_normalize_cached(p) for p in new_path] + + if isinstance(module.__path__, list): + module.__path__[:] = new_path + else: + module.__path__ = new_path + + +def declare_namespace(packageName): + """Declare that package 'packageName' is a namespace package""" + + msg = ( + f"Deprecated call to `pkg_resources.declare_namespace({packageName!r})`.\n" + "Implementing implicit namespace packages (as specified in PEP 420) " + "is preferred to `pkg_resources.declare_namespace`. " + "See https://setuptools.pypa.io/en/latest/references/" + "keywords.html#keyword-namespace-packages" + ) + warnings.warn(msg, DeprecationWarning, stacklevel=2) + + _imp.acquire_lock() + try: + if packageName in _namespace_packages: + return + + path = sys.path + parent, _, _ = packageName.rpartition('.') + + if parent: + declare_namespace(parent) + if parent not in _namespace_packages: + __import__(parent) + try: + path = sys.modules[parent].__path__ + except AttributeError as e: + raise TypeError("Not a package:", parent) from e + + # Track what packages are namespaces, so when new path items are added, + # they can be updated + _namespace_packages.setdefault(parent or None, []).append(packageName) + _namespace_packages.setdefault(packageName, []) + + for path_item in path: + # Ensure all the parent's path items are reflected in the child, + # if they apply + _handle_ns(packageName, path_item) + + finally: + _imp.release_lock() + + +def fixup_namespace_packages(path_item, parent=None): + """Ensure that previously-declared namespace packages include path_item""" + _imp.acquire_lock() + try: + for package in _namespace_packages.get(parent, ()): + subpath = _handle_ns(package, path_item) + if subpath: + fixup_namespace_packages(subpath, package) + finally: + _imp.release_lock() + + +def file_ns_handler(importer, path_item, packageName, module): + """Compute an ns-package subpath for a filesystem or zipfile importer""" + + subpath = os.path.join(path_item, packageName.split('.')[-1]) + normalized = _normalize_cached(subpath) + for item in module.__path__: + if _normalize_cached(item) == normalized: + break + else: + # Only return the path if it's not already there + return subpath + + +if hasattr(pkgutil, 'ImpImporter'): + register_namespace_handler(pkgutil.ImpImporter, file_ns_handler) + +register_namespace_handler(zipimport.zipimporter, file_ns_handler) +register_namespace_handler(importlib_machinery.FileFinder, file_ns_handler) + + +def null_ns_handler(importer, path_item, packageName, module): + return None + + +register_namespace_handler(object, null_ns_handler) + + +def normalize_path(filename): + """Normalize a file/dir name for comparison purposes""" + return os.path.normcase(os.path.realpath(os.path.normpath(_cygwin_patch(filename)))) + + +def _cygwin_patch(filename): # pragma: nocover + """ + Contrary to POSIX 2008, on Cygwin, getcwd (3) contains + symlink components. Using + os.path.abspath() works around this limitation. A fix in os.getcwd() + would probably better, in Cygwin even more so, except + that this seems to be by design... + """ + return os.path.abspath(filename) if sys.platform == 'cygwin' else filename + + +def _normalize_cached(filename, _cache={}): + try: + return _cache[filename] + except KeyError: + _cache[filename] = result = normalize_path(filename) + return result + + +def _is_egg_path(path): + """ + Determine if given path appears to be an egg. + """ + return _is_zip_egg(path) or _is_unpacked_egg(path) + + +def _is_zip_egg(path): + return ( + path.lower().endswith('.egg') + and os.path.isfile(path) + and zipfile.is_zipfile(path) + ) + + +def _is_unpacked_egg(path): + """ + Determine if given path appears to be an unpacked egg. + """ + return path.lower().endswith('.egg') and os.path.isfile( + os.path.join(path, 'EGG-INFO', 'PKG-INFO') + ) + + +def _set_parent_ns(packageName): + parts = packageName.split('.') + name = parts.pop() + if parts: + parent = '.'.join(parts) + setattr(sys.modules[parent], name, sys.modules[packageName]) + + +MODULE = re.compile(r"\w+(\.\w+)*$").match +EGG_NAME = re.compile( + r""" + (?P<name>[^-]+) ( + -(?P<ver>[^-]+) ( + -py(?P<pyver>[^-]+) ( + -(?P<plat>.+) + )? + )? + )? + """, + re.VERBOSE | re.IGNORECASE, +).match + + +class EntryPoint: + """Object representing an advertised importable object""" + + def __init__(self, name, module_name, attrs=(), extras=(), dist=None): + if not MODULE(module_name): + raise ValueError("Invalid module name", module_name) + self.name = name + self.module_name = module_name + self.attrs = tuple(attrs) + self.extras = tuple(extras) + self.dist = dist + + def __str__(self): + s = "%s = %s" % (self.name, self.module_name) + if self.attrs: + s += ':' + '.'.join(self.attrs) + if self.extras: + s += ' [%s]' % ','.join(self.extras) + return s + + def __repr__(self): + return "EntryPoint.parse(%r)" % str(self) + + def load(self, require=True, *args, **kwargs): + """ + Require packages for this EntryPoint, then resolve it. + """ + if not require or args or kwargs: + warnings.warn( + "Parameters to load are deprecated. Call .resolve and " + ".require separately.", + PkgResourcesDeprecationWarning, + stacklevel=2, + ) + if require: + self.require(*args, **kwargs) + return self.resolve() + + def resolve(self): + """ + Resolve the entry point from its module and attrs. + """ + module = __import__(self.module_name, fromlist=['__name__'], level=0) + try: + return functools.reduce(getattr, self.attrs, module) + except AttributeError as exc: + raise ImportError(str(exc)) from exc + + def require(self, env=None, installer=None): + if self.extras and not self.dist: + raise UnknownExtra("Can't require() without a distribution", self) + + # Get the requirements for this entry point with all its extras and + # then resolve them. We have to pass `extras` along when resolving so + # that the working set knows what extras we want. Otherwise, for + # dist-info distributions, the working set will assume that the + # requirements for that extra are purely optional and skip over them. + reqs = self.dist.requires(self.extras) + items = working_set.resolve(reqs, env, installer, extras=self.extras) + list(map(working_set.add, items)) + + pattern = re.compile( + r'\s*' + r'(?P<name>.+?)\s*' + r'=\s*' + r'(?P<module>[\w.]+)\s*' + r'(:\s*(?P<attr>[\w.]+))?\s*' + r'(?P<extras>\[.*\])?\s*$' + ) + + @classmethod + def parse(cls, src, dist=None): + """Parse a single entry point from string `src` + + Entry point syntax follows the form:: + + name = some.module:some.attr [extra1, extra2] + + The entry name and module name are required, but the ``:attrs`` and + ``[extras]`` parts are optional + """ + m = cls.pattern.match(src) + if not m: + msg = "EntryPoint must be in 'name=module:attrs [extras]' format" + raise ValueError(msg, src) + res = m.groupdict() + extras = cls._parse_extras(res['extras']) + attrs = res['attr'].split('.') if res['attr'] else () + return cls(res['name'], res['module'], attrs, extras, dist) + + @classmethod + def _parse_extras(cls, extras_spec): + if not extras_spec: + return () + req = Requirement.parse('x' + extras_spec) + if req.specs: + raise ValueError() + return req.extras + + @classmethod + def parse_group(cls, group, lines, dist=None): + """Parse an entry point group""" + if not MODULE(group): + raise ValueError("Invalid group name", group) + this = {} + for line in yield_lines(lines): + ep = cls.parse(line, dist) + if ep.name in this: + raise ValueError("Duplicate entry point", group, ep.name) + this[ep.name] = ep + return this + + @classmethod + def parse_map(cls, data, dist=None): + """Parse a map of entry point groups""" + if isinstance(data, dict): + data = data.items() + else: + data = split_sections(data) + maps = {} + for group, lines in data: + if group is None: + if not lines: + continue + raise ValueError("Entry points must be listed in groups") + group = group.strip() + if group in maps: + raise ValueError("Duplicate group name", group) + maps[group] = cls.parse_group(group, lines, dist) + return maps + + +def _version_from_file(lines): + """ + Given an iterable of lines from a Metadata file, return + the value of the Version field, if present, or None otherwise. + """ + + def is_version_line(line): + return line.lower().startswith('version:') + + version_lines = filter(is_version_line, lines) + line = next(iter(version_lines), '') + _, _, value = line.partition(':') + return safe_version(value.strip()) or None + + +class Distribution: + """Wrap an actual or potential sys.path entry w/metadata""" + + PKG_INFO = 'PKG-INFO' + + def __init__( + self, + location=None, + metadata=None, + project_name=None, + version=None, + py_version=PY_MAJOR, + platform=None, + precedence=EGG_DIST, + ): + self.project_name = safe_name(project_name or 'Unknown') + if version is not None: + self._version = safe_version(version) + self.py_version = py_version + self.platform = platform + self.location = location + self.precedence = precedence + self._provider = metadata or empty_provider + + @classmethod + def from_location(cls, location, basename, metadata=None, **kw): + project_name, version, py_version, platform = [None] * 4 + basename, ext = os.path.splitext(basename) + if ext.lower() in _distributionImpl: + cls = _distributionImpl[ext.lower()] + + match = EGG_NAME(basename) + if match: + project_name, version, py_version, platform = match.group( + 'name', 'ver', 'pyver', 'plat' + ) + return cls( + location, + metadata, + project_name=project_name, + version=version, + py_version=py_version, + platform=platform, + **kw, + )._reload_version() + + def _reload_version(self): + return self + + @property + def hashcmp(self): + return ( + self._forgiving_parsed_version, + self.precedence, + self.key, + self.location, + self.py_version or '', + self.platform or '', + ) + + def __hash__(self): + return hash(self.hashcmp) + + def __lt__(self, other): + return self.hashcmp < other.hashcmp + + def __le__(self, other): + return self.hashcmp <= other.hashcmp + + def __gt__(self, other): + return self.hashcmp > other.hashcmp + + def __ge__(self, other): + return self.hashcmp >= other.hashcmp + + def __eq__(self, other): + if not isinstance(other, self.__class__): + # It's not a Distribution, so they are not equal + return False + return self.hashcmp == other.hashcmp + + def __ne__(self, other): + return not self == other + + # These properties have to be lazy so that we don't have to load any + # metadata until/unless it's actually needed. (i.e., some distributions + # may not know their name or version without loading PKG-INFO) + + @property + def key(self): + try: + return self._key + except AttributeError: + self._key = key = self.project_name.lower() + return key + + @property + def parsed_version(self): + if not hasattr(self, "_parsed_version"): + try: + self._parsed_version = parse_version(self.version) + except packaging.version.InvalidVersion as ex: + info = f"(package: {self.project_name})" + if hasattr(ex, "add_note"): + ex.add_note(info) # PEP 678 + raise + raise packaging.version.InvalidVersion(f"{str(ex)} {info}") from None + + return self._parsed_version + + @property + def _forgiving_parsed_version(self): + try: + return self.parsed_version + except packaging.version.InvalidVersion as ex: + self._parsed_version = parse_version(_forgiving_version(self.version)) + + notes = "\n".join(getattr(ex, "__notes__", [])) # PEP 678 + msg = f"""!!\n\n + ************************************************************************* + {str(ex)}\n{notes} + + This is a long overdue deprecation. + For the time being, `pkg_resources` will use `{self._parsed_version}` + as a replacement to avoid breaking existing environments, + but no future compatibility is guaranteed. + + If you maintain package {self.project_name} you should implement + the relevant changes to adequate the project to PEP 440 immediately. + ************************************************************************* + \n\n!! + """ + warnings.warn(msg, DeprecationWarning) + + return self._parsed_version + + @property + def version(self): + try: + return self._version + except AttributeError as e: + version = self._get_version() + if version is None: + path = self._get_metadata_path_for_display(self.PKG_INFO) + msg = ("Missing 'Version:' header and/or {} file at path: {}").format( + self.PKG_INFO, path + ) + raise ValueError(msg, self) from e + + return version + + @property + def _dep_map(self): + """ + A map of extra to its list of (direct) requirements + for this distribution, including the null extra. + """ + try: + return self.__dep_map + except AttributeError: + self.__dep_map = self._filter_extras(self._build_dep_map()) + return self.__dep_map + + @staticmethod + def _filter_extras(dm): + """ + Given a mapping of extras to dependencies, strip off + environment markers and filter out any dependencies + not matching the markers. + """ + for extra in list(filter(None, dm)): + new_extra = extra + reqs = dm.pop(extra) + new_extra, _, marker = extra.partition(':') + fails_marker = marker and ( + invalid_marker(marker) or not evaluate_marker(marker) + ) + if fails_marker: + reqs = [] + new_extra = safe_extra(new_extra) or None + + dm.setdefault(new_extra, []).extend(reqs) + return dm + + def _build_dep_map(self): + dm = {} + for name in 'requires.txt', 'depends.txt': + for extra, reqs in split_sections(self._get_metadata(name)): + dm.setdefault(extra, []).extend(parse_requirements(reqs)) + return dm + + def requires(self, extras=()): + """List of Requirements needed for this distro if `extras` are used""" + dm = self._dep_map + deps = [] + deps.extend(dm.get(None, ())) + for ext in extras: + try: + deps.extend(dm[safe_extra(ext)]) + except KeyError as e: + raise UnknownExtra( + "%s has no such extra feature %r" % (self, ext) + ) from e + return deps + + def _get_metadata_path_for_display(self, name): + """ + Return the path to the given metadata file, if available. + """ + try: + # We need to access _get_metadata_path() on the provider object + # directly rather than through this class's __getattr__() + # since _get_metadata_path() is marked private. + path = self._provider._get_metadata_path(name) + + # Handle exceptions e.g. in case the distribution's metadata + # provider doesn't support _get_metadata_path(). + except Exception: + return '[could not detect]' + + return path + + def _get_metadata(self, name): + if self.has_metadata(name): + for line in self.get_metadata_lines(name): + yield line + + def _get_version(self): + lines = self._get_metadata(self.PKG_INFO) + version = _version_from_file(lines) + + return version + + def activate(self, path=None, replace=False): + """Ensure distribution is importable on `path` (default=sys.path)""" + if path is None: + path = sys.path + self.insert_on(path, replace=replace) + if path is sys.path: + fixup_namespace_packages(self.location) + for pkg in self._get_metadata('namespace_packages.txt'): + if pkg in sys.modules: + declare_namespace(pkg) + + def egg_name(self): + """Return what this distribution's standard .egg filename should be""" + filename = "%s-%s-py%s" % ( + to_filename(self.project_name), + to_filename(self.version), + self.py_version or PY_MAJOR, + ) + + if self.platform: + filename += '-' + self.platform + return filename + + def __repr__(self): + if self.location: + return "%s (%s)" % (self, self.location) + else: + return str(self) + + def __str__(self): + try: + version = getattr(self, 'version', None) + except ValueError: + version = None + version = version or "[unknown version]" + return "%s %s" % (self.project_name, version) + + def __getattr__(self, attr): + """Delegate all unrecognized public attributes to .metadata provider""" + if attr.startswith('_'): + raise AttributeError(attr) + return getattr(self._provider, attr) + + def __dir__(self): + return list( + set(super(Distribution, self).__dir__()) + | set(attr for attr in self._provider.__dir__() if not attr.startswith('_')) + ) + + @classmethod + def from_filename(cls, filename, metadata=None, **kw): + return cls.from_location( + _normalize_cached(filename), os.path.basename(filename), metadata, **kw + ) + + def as_requirement(self): + """Return a ``Requirement`` that matches this distribution exactly""" + if isinstance(self.parsed_version, packaging.version.Version): + spec = "%s==%s" % (self.project_name, self.parsed_version) + else: + spec = "%s===%s" % (self.project_name, self.parsed_version) + + return Requirement.parse(spec) + + def load_entry_point(self, group, name): + """Return the `name` entry point of `group` or raise ImportError""" + ep = self.get_entry_info(group, name) + if ep is None: + raise ImportError("Entry point %r not found" % ((group, name),)) + return ep.load() + + def get_entry_map(self, group=None): + """Return the entry point map for `group`, or the full entry map""" + try: + ep_map = self._ep_map + except AttributeError: + ep_map = self._ep_map = EntryPoint.parse_map( + self._get_metadata('entry_points.txt'), self + ) + if group is not None: + return ep_map.get(group, {}) + return ep_map + + def get_entry_info(self, group, name): + """Return the EntryPoint object for `group`+`name`, or ``None``""" + return self.get_entry_map(group).get(name) + + # FIXME: 'Distribution.insert_on' is too complex (13) + def insert_on(self, path, loc=None, replace=False): # noqa: C901 + """Ensure self.location is on path + + If replace=False (default): + - If location is already in path anywhere, do nothing. + - Else: + - If it's an egg and its parent directory is on path, + insert just ahead of the parent. + - Else: add to the end of path. + If replace=True: + - If location is already on path anywhere (not eggs) + or higher priority than its parent (eggs) + do nothing. + - Else: + - If it's an egg and its parent directory is on path, + insert just ahead of the parent, + removing any lower-priority entries. + - Else: add it to the front of path. + """ + + loc = loc or self.location + if not loc: + return + + nloc = _normalize_cached(loc) + bdir = os.path.dirname(nloc) + npath = [(p and _normalize_cached(p) or p) for p in path] + + for p, item in enumerate(npath): + if item == nloc: + if replace: + break + else: + # don't modify path (even removing duplicates) if + # found and not replace + return + elif item == bdir and self.precedence == EGG_DIST: + # if it's an .egg, give it precedence over its directory + # UNLESS it's already been added to sys.path and replace=False + if (not replace) and nloc in npath[p:]: + return + if path is sys.path: + self.check_version_conflict() + path.insert(p, loc) + npath.insert(p, nloc) + break + else: + if path is sys.path: + self.check_version_conflict() + if replace: + path.insert(0, loc) + else: + path.append(loc) + return + + # p is the spot where we found or inserted loc; now remove duplicates + while True: + try: + np = npath.index(nloc, p + 1) + except ValueError: + break + else: + del npath[np], path[np] + # ha! + p = np + + return + + def check_version_conflict(self): + if self.key == 'setuptools': + # ignore the inevitable setuptools self-conflicts :( + return + + nsp = dict.fromkeys(self._get_metadata('namespace_packages.txt')) + loc = normalize_path(self.location) + for modname in self._get_metadata('top_level.txt'): + if ( + modname not in sys.modules + or modname in nsp + or modname in _namespace_packages + ): + continue + if modname in ('pkg_resources', 'setuptools', 'site'): + continue + fn = getattr(sys.modules[modname], '__file__', None) + if fn and ( + normalize_path(fn).startswith(loc) or fn.startswith(self.location) + ): + continue + issue_warning( + "Module %s was already imported from %s, but %s is being added" + " to sys.path" % (modname, fn, self.location), + ) + + def has_version(self): + try: + self.version + except ValueError: + issue_warning("Unbuilt egg for " + repr(self)) + return False + except SystemError: + # TODO: remove this except clause when python/cpython#103632 is fixed. + return False + return True + + def clone(self, **kw): + """Copy this distribution, substituting in any changed keyword args""" + names = 'project_name version py_version platform location precedence' + for attr in names.split(): + kw.setdefault(attr, getattr(self, attr, None)) + kw.setdefault('metadata', self._provider) + return self.__class__(**kw) + + @property + def extras(self): + return [dep for dep in self._dep_map if dep] + + +class EggInfoDistribution(Distribution): + def _reload_version(self): + """ + Packages installed by distutils (e.g. numpy or scipy), + which uses an old safe_version, and so + their version numbers can get mangled when + converted to filenames (e.g., 1.11.0.dev0+2329eae to + 1.11.0.dev0_2329eae). These distributions will not be + parsed properly + downstream by Distribution and safe_version, so + take an extra step and try to get the version number from + the metadata file itself instead of the filename. + """ + md_version = self._get_version() + if md_version: + self._version = md_version + return self + + +class DistInfoDistribution(Distribution): + """ + Wrap an actual or potential sys.path entry + w/metadata, .dist-info style. + """ + + PKG_INFO = 'METADATA' + EQEQ = re.compile(r"([\(,])\s*(\d.*?)\s*([,\)])") + + @property + def _parsed_pkg_info(self): + """Parse and cache metadata""" + try: + return self._pkg_info + except AttributeError: + metadata = self.get_metadata(self.PKG_INFO) + self._pkg_info = email.parser.Parser().parsestr(metadata) + return self._pkg_info + + @property + def _dep_map(self): + try: + return self.__dep_map + except AttributeError: + self.__dep_map = self._compute_dependencies() + return self.__dep_map + + def _compute_dependencies(self): + """Recompute this distribution's dependencies.""" + dm = self.__dep_map = {None: []} + + reqs = [] + # Including any condition expressions + for req in self._parsed_pkg_info.get_all('Requires-Dist') or []: + reqs.extend(parse_requirements(req)) + + def reqs_for_extra(extra): + for req in reqs: + if not req.marker or req.marker.evaluate({'extra': extra}): + yield req + + common = types.MappingProxyType(dict.fromkeys(reqs_for_extra(None))) + dm[None].extend(common) + + for extra in self._parsed_pkg_info.get_all('Provides-Extra') or []: + s_extra = safe_extra(extra.strip()) + dm[s_extra] = [r for r in reqs_for_extra(extra) if r not in common] + + return dm + + +_distributionImpl = { + '.egg': Distribution, + '.egg-info': EggInfoDistribution, + '.dist-info': DistInfoDistribution, +} + + +def issue_warning(*args, **kw): + level = 1 + g = globals() + try: + # find the first stack frame that is *not* code in + # the pkg_resources module, to use for the warning + while sys._getframe(level).f_globals is g: + level += 1 + except ValueError: + pass + warnings.warn(stacklevel=level + 1, *args, **kw) + + +def parse_requirements(strs): + """ + Yield ``Requirement`` objects for each specification in `strs`. + + `strs` must be a string, or a (possibly-nested) iterable thereof. + """ + return map(Requirement, join_continuation(map(drop_comment, yield_lines(strs)))) + + +class RequirementParseError(packaging.requirements.InvalidRequirement): + "Compatibility wrapper for InvalidRequirement" + + +class Requirement(packaging.requirements.Requirement): + def __init__(self, requirement_string): + """DO NOT CALL THIS UNDOCUMENTED METHOD; use Requirement.parse()!""" + super(Requirement, self).__init__(requirement_string) + self.unsafe_name = self.name + project_name = safe_name(self.name) + self.project_name, self.key = project_name, project_name.lower() + self.specs = [(spec.operator, spec.version) for spec in self.specifier] + self.extras = tuple(map(safe_extra, self.extras)) + self.hashCmp = ( + self.key, + self.url, + self.specifier, + frozenset(self.extras), + str(self.marker) if self.marker else None, + ) + self.__hash = hash(self.hashCmp) + + def __eq__(self, other): + return isinstance(other, Requirement) and self.hashCmp == other.hashCmp + + def __ne__(self, other): + return not self == other + + def __contains__(self, item): + if isinstance(item, Distribution): + if item.key != self.key: + return False + + item = item.version + + # Allow prereleases always in order to match the previous behavior of + # this method. In the future this should be smarter and follow PEP 440 + # more accurately. + return self.specifier.contains(item, prereleases=True) + + def __hash__(self): + return self.__hash + + def __repr__(self): + return "Requirement.parse(%r)" % str(self) + + @staticmethod + def parse(s): + (req,) = parse_requirements(s) + return req + + +def _always_object(classes): + """ + Ensure object appears in the mro even + for old-style classes. + """ + if object not in classes: + return classes + (object,) + return classes + + +def _find_adapter(registry, ob): + """Return an adapter factory for `ob` from `registry`""" + types = _always_object(inspect.getmro(getattr(ob, '__class__', type(ob)))) + for t in types: + if t in registry: + return registry[t] + + +def ensure_directory(path): + """Ensure that the parent directory of `path` exists""" + dirname = os.path.dirname(path) + os.makedirs(dirname, exist_ok=True) + + +def _bypass_ensure_directory(path): + """Sandbox-bypassing version of ensure_directory()""" + if not WRITE_SUPPORT: + raise IOError('"os.mkdir" not supported on this platform.') + dirname, filename = split(path) + if dirname and filename and not isdir(dirname): + _bypass_ensure_directory(dirname) + try: + mkdir(dirname, 0o755) + except FileExistsError: + pass + + +def split_sections(s): + """Split a string or iterable thereof into (section, content) pairs + + Each ``section`` is a stripped version of the section header ("[section]") + and each ``content`` is a list of stripped lines excluding blank lines and + comment-only lines. If there are any such lines before the first section + header, they're returned in a first ``section`` of ``None``. + """ + section = None + content = [] + for line in yield_lines(s): + if line.startswith("["): + if line.endswith("]"): + if section or content: + yield section, content + section = line[1:-1].strip() + content = [] + else: + raise ValueError("Invalid section heading", line) + else: + content.append(line) + + # wrap up last segment + yield section, content + + +def _mkstemp(*args, **kw): + old_open = os.open + try: + # temporarily bypass sandboxing + os.open = os_open + return tempfile.mkstemp(*args, **kw) + finally: + # and then put it back + os.open = old_open + + +# Silence the PEP440Warning by default, so that end users don't get hit by it +# randomly just because they use pkg_resources. We want to append the rule +# because we want earlier uses of filterwarnings to take precedence over this +# one. +warnings.filterwarnings("ignore", category=PEP440Warning, append=True) + + +# from jaraco.functools 1.3 +def _call_aside(f, *args, **kwargs): + f(*args, **kwargs) + return f + + +@_call_aside +def _initialize(g=globals()): + "Set up global resource manager (deliberately not state-saved)" + manager = ResourceManager() + g['_manager'] = manager + g.update( + (name, getattr(manager, name)) + for name in dir(manager) + if not name.startswith('_') + ) + + +class PkgResourcesDeprecationWarning(Warning): + """ + Base class for warning about deprecations in ``pkg_resources`` + + This class is not derived from ``DeprecationWarning``, and as such is + visible by default. + """ + + +@_call_aside +def _initialize_master_working_set(): + """ + Prepare the master working set and make the ``require()`` + API available. + + This function has explicit effects on the global state + of pkg_resources. It is intended to be invoked once at + the initialization of this module. + + Invocation by other packages is unsupported and done + at their own risk. + """ + working_set = WorkingSet._build_master() + _declare_state('object', working_set=working_set) + + require = working_set.require + iter_entry_points = working_set.iter_entry_points + add_activation_listener = working_set.subscribe + run_script = working_set.run_script + # backward compatibility + run_main = run_script + # Activate all distributions already on sys.path with replace=False and + # ensure that all distributions added to the working set in the future + # (e.g. by calling ``require()``) will get activated as well, + # with higher priority (replace=True). + tuple(dist.activate(replace=False) for dist in working_set) + add_activation_listener( + lambda dist: dist.activate(replace=True), + existing=False, + ) + working_set.entries = [] + # match order + list(map(working_set.add_entry, sys.path)) + globals().update(locals()) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/__init__.py new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/__init__.py diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/__init__.py new file mode 100644 index 0000000000..34e3a9950c --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/__init__.py @@ -0,0 +1,36 @@ +"""Read resources contained within a package.""" + +from ._common import ( + as_file, + files, + Package, +) + +from ._legacy import ( + contents, + open_binary, + read_binary, + open_text, + read_text, + is_resource, + path, + Resource, +) + +from .abc import ResourceReader + + +__all__ = [ + 'Package', + 'Resource', + 'ResourceReader', + 'as_file', + 'contents', + 'files', + 'is_resource', + 'open_binary', + 'open_text', + 'path', + 'read_binary', + 'read_text', +] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_adapters.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_adapters.py new file mode 100644 index 0000000000..ea363d86a5 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_adapters.py @@ -0,0 +1,170 @@ +from contextlib import suppress +from io import TextIOWrapper + +from . import abc + + +class SpecLoaderAdapter: + """ + Adapt a package spec to adapt the underlying loader. + """ + + def __init__(self, spec, adapter=lambda spec: spec.loader): + self.spec = spec + self.loader = adapter(spec) + + def __getattr__(self, name): + return getattr(self.spec, name) + + +class TraversableResourcesLoader: + """ + Adapt a loader to provide TraversableResources. + """ + + def __init__(self, spec): + self.spec = spec + + def get_resource_reader(self, name): + return CompatibilityFiles(self.spec)._native() + + +def _io_wrapper(file, mode='r', *args, **kwargs): + if mode == 'r': + return TextIOWrapper(file, *args, **kwargs) + elif mode == 'rb': + return file + raise ValueError( + "Invalid mode value '{}', only 'r' and 'rb' are supported".format(mode) + ) + + +class CompatibilityFiles: + """ + Adapter for an existing or non-existent resource reader + to provide a compatibility .files(). + """ + + class SpecPath(abc.Traversable): + """ + Path tied to a module spec. + Can be read and exposes the resource reader children. + """ + + def __init__(self, spec, reader): + self._spec = spec + self._reader = reader + + def iterdir(self): + if not self._reader: + return iter(()) + return iter( + CompatibilityFiles.ChildPath(self._reader, path) + for path in self._reader.contents() + ) + + def is_file(self): + return False + + is_dir = is_file + + def joinpath(self, other): + if not self._reader: + return CompatibilityFiles.OrphanPath(other) + return CompatibilityFiles.ChildPath(self._reader, other) + + @property + def name(self): + return self._spec.name + + def open(self, mode='r', *args, **kwargs): + return _io_wrapper(self._reader.open_resource(None), mode, *args, **kwargs) + + class ChildPath(abc.Traversable): + """ + Path tied to a resource reader child. + Can be read but doesn't expose any meaningful children. + """ + + def __init__(self, reader, name): + self._reader = reader + self._name = name + + def iterdir(self): + return iter(()) + + def is_file(self): + return self._reader.is_resource(self.name) + + def is_dir(self): + return not self.is_file() + + def joinpath(self, other): + return CompatibilityFiles.OrphanPath(self.name, other) + + @property + def name(self): + return self._name + + def open(self, mode='r', *args, **kwargs): + return _io_wrapper( + self._reader.open_resource(self.name), mode, *args, **kwargs + ) + + class OrphanPath(abc.Traversable): + """ + Orphan path, not tied to a module spec or resource reader. + Can't be read and doesn't expose any meaningful children. + """ + + def __init__(self, *path_parts): + if len(path_parts) < 1: + raise ValueError('Need at least one path part to construct a path') + self._path = path_parts + + def iterdir(self): + return iter(()) + + def is_file(self): + return False + + is_dir = is_file + + def joinpath(self, other): + return CompatibilityFiles.OrphanPath(*self._path, other) + + @property + def name(self): + return self._path[-1] + + def open(self, mode='r', *args, **kwargs): + raise FileNotFoundError("Can't open orphan path") + + def __init__(self, spec): + self.spec = spec + + @property + def _reader(self): + with suppress(AttributeError): + return self.spec.loader.get_resource_reader(self.spec.name) + + def _native(self): + """ + Return the native reader if it supports files(). + """ + reader = self._reader + return reader if hasattr(reader, 'files') else self + + def __getattr__(self, attr): + return getattr(self._reader, attr) + + def files(self): + return CompatibilityFiles.SpecPath(self.spec, self._reader) + + +def wrap_spec(package): + """ + Construct a package spec with traversable compatibility + on the spec/loader/reader. + """ + return SpecLoaderAdapter(package.__spec__, TraversableResourcesLoader) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_common.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_common.py new file mode 100644 index 0000000000..3c6de1cfb2 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_common.py @@ -0,0 +1,207 @@ +import os +import pathlib +import tempfile +import functools +import contextlib +import types +import importlib +import inspect +import warnings +import itertools + +from typing import Union, Optional, cast +from .abc import ResourceReader, Traversable + +from ._compat import wrap_spec + +Package = Union[types.ModuleType, str] +Anchor = Package + + +def package_to_anchor(func): + """ + Replace 'package' parameter as 'anchor' and warn about the change. + + Other errors should fall through. + + >>> files('a', 'b') + Traceback (most recent call last): + TypeError: files() takes from 0 to 1 positional arguments but 2 were given + """ + undefined = object() + + @functools.wraps(func) + def wrapper(anchor=undefined, package=undefined): + if package is not undefined: + if anchor is not undefined: + return func(anchor, package) + warnings.warn( + "First parameter to files is renamed to 'anchor'", + DeprecationWarning, + stacklevel=2, + ) + return func(package) + elif anchor is undefined: + return func() + return func(anchor) + + return wrapper + + +@package_to_anchor +def files(anchor: Optional[Anchor] = None) -> Traversable: + """ + Get a Traversable resource for an anchor. + """ + return from_package(resolve(anchor)) + + +def get_resource_reader(package: types.ModuleType) -> Optional[ResourceReader]: + """ + Return the package's loader if it's a ResourceReader. + """ + # We can't use + # a issubclass() check here because apparently abc.'s __subclasscheck__() + # hook wants to create a weak reference to the object, but + # zipimport.zipimporter does not support weak references, resulting in a + # TypeError. That seems terrible. + spec = package.__spec__ + reader = getattr(spec.loader, 'get_resource_reader', None) # type: ignore + if reader is None: + return None + return reader(spec.name) # type: ignore + + +@functools.singledispatch +def resolve(cand: Optional[Anchor]) -> types.ModuleType: + return cast(types.ModuleType, cand) + + +@resolve.register +def _(cand: str) -> types.ModuleType: + return importlib.import_module(cand) + + +@resolve.register +def _(cand: None) -> types.ModuleType: + return resolve(_infer_caller().f_globals['__name__']) + + +def _infer_caller(): + """ + Walk the stack and find the frame of the first caller not in this module. + """ + + def is_this_file(frame_info): + return frame_info.filename == __file__ + + def is_wrapper(frame_info): + return frame_info.function == 'wrapper' + + not_this_file = itertools.filterfalse(is_this_file, inspect.stack()) + # also exclude 'wrapper' due to singledispatch in the call stack + callers = itertools.filterfalse(is_wrapper, not_this_file) + return next(callers).frame + + +def from_package(package: types.ModuleType): + """ + Return a Traversable object for the given package. + + """ + spec = wrap_spec(package) + reader = spec.loader.get_resource_reader(spec.name) + return reader.files() + + +@contextlib.contextmanager +def _tempfile( + reader, + suffix='', + # gh-93353: Keep a reference to call os.remove() in late Python + # finalization. + *, + _os_remove=os.remove, +): + # Not using tempfile.NamedTemporaryFile as it leads to deeper 'try' + # blocks due to the need to close the temporary file to work on Windows + # properly. + fd, raw_path = tempfile.mkstemp(suffix=suffix) + try: + try: + os.write(fd, reader()) + finally: + os.close(fd) + del reader + yield pathlib.Path(raw_path) + finally: + try: + _os_remove(raw_path) + except FileNotFoundError: + pass + + +def _temp_file(path): + return _tempfile(path.read_bytes, suffix=path.name) + + +def _is_present_dir(path: Traversable) -> bool: + """ + Some Traversables implement ``is_dir()`` to raise an + exception (i.e. ``FileNotFoundError``) when the + directory doesn't exist. This function wraps that call + to always return a boolean and only return True + if there's a dir and it exists. + """ + with contextlib.suppress(FileNotFoundError): + return path.is_dir() + return False + + +@functools.singledispatch +def as_file(path): + """ + Given a Traversable object, return that object as a + path on the local file system in a context manager. + """ + return _temp_dir(path) if _is_present_dir(path) else _temp_file(path) + + +@as_file.register(pathlib.Path) +@contextlib.contextmanager +def _(path): + """ + Degenerate behavior for pathlib.Path objects. + """ + yield path + + +@contextlib.contextmanager +def _temp_path(dir: tempfile.TemporaryDirectory): + """ + Wrap tempfile.TemporyDirectory to return a pathlib object. + """ + with dir as result: + yield pathlib.Path(result) + + +@contextlib.contextmanager +def _temp_dir(path): + """ + Given a traversable dir, recursively replicate the whole tree + to the file system in a context manager. + """ + assert path.is_dir() + with _temp_path(tempfile.TemporaryDirectory()) as temp_dir: + yield _write_contents(temp_dir, path) + + +def _write_contents(target, source): + child = target.joinpath(source.name) + if source.is_dir(): + child.mkdir() + for item in source.iterdir(): + _write_contents(child, item) + else: + child.write_bytes(source.read_bytes()) + return child diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_compat.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_compat.py new file mode 100644 index 0000000000..8b5b1d280f --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_compat.py @@ -0,0 +1,108 @@ +# flake8: noqa + +import abc +import os +import sys +import pathlib +from contextlib import suppress +from typing import Union + + +if sys.version_info >= (3, 10): + from zipfile import Path as ZipPath # type: ignore +else: + from ..zipp import Path as ZipPath # type: ignore + + +try: + from typing import runtime_checkable # type: ignore +except ImportError: + + def runtime_checkable(cls): # type: ignore + return cls + + +try: + from typing import Protocol # type: ignore +except ImportError: + Protocol = abc.ABC # type: ignore + + +class TraversableResourcesLoader: + """ + Adapt loaders to provide TraversableResources and other + compatibility. + + Used primarily for Python 3.9 and earlier where the native + loaders do not yet implement TraversableResources. + """ + + def __init__(self, spec): + self.spec = spec + + @property + def path(self): + return self.spec.origin + + def get_resource_reader(self, name): + from . import readers, _adapters + + def _zip_reader(spec): + with suppress(AttributeError): + return readers.ZipReader(spec.loader, spec.name) + + def _namespace_reader(spec): + with suppress(AttributeError, ValueError): + return readers.NamespaceReader(spec.submodule_search_locations) + + def _available_reader(spec): + with suppress(AttributeError): + return spec.loader.get_resource_reader(spec.name) + + def _native_reader(spec): + reader = _available_reader(spec) + return reader if hasattr(reader, 'files') else None + + def _file_reader(spec): + try: + path = pathlib.Path(self.path) + except TypeError: + return None + if path.exists(): + return readers.FileReader(self) + + return ( + # native reader if it supplies 'files' + _native_reader(self.spec) + or + # local ZipReader if a zip module + _zip_reader(self.spec) + or + # local NamespaceReader if a namespace module + _namespace_reader(self.spec) + or + # local FileReader + _file_reader(self.spec) + # fallback - adapt the spec ResourceReader to TraversableReader + or _adapters.CompatibilityFiles(self.spec) + ) + + +def wrap_spec(package): + """ + Construct a package spec with traversable compatibility + on the spec/loader/reader. + + Supersedes _adapters.wrap_spec to use TraversableResourcesLoader + from above for older Python compatibility (<3.10). + """ + from . import _adapters + + return _adapters.SpecLoaderAdapter(package.__spec__, TraversableResourcesLoader) + + +if sys.version_info >= (3, 9): + StrPath = Union[str, os.PathLike[str]] +else: + # PathLike is only subscriptable at runtime in 3.9+ + StrPath = Union[str, "os.PathLike[str]"] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_itertools.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_itertools.py new file mode 100644 index 0000000000..cce05582ff --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_itertools.py @@ -0,0 +1,35 @@ +from itertools import filterfalse + +from typing import ( + Callable, + Iterable, + Iterator, + Optional, + Set, + TypeVar, + Union, +) + +# Type and type variable definitions +_T = TypeVar('_T') +_U = TypeVar('_U') + + +def unique_everseen( + iterable: Iterable[_T], key: Optional[Callable[[_T], _U]] = None +) -> Iterator[_T]: + "List unique elements, preserving order. Remember all elements ever seen." + # unique_everseen('AAAABBBCCDAABBB') --> A B C D + # unique_everseen('ABBCcAD', str.lower) --> A B C D + seen: Set[Union[_T, _U]] = set() + seen_add = seen.add + if key is None: + for element in filterfalse(seen.__contains__, iterable): + seen_add(element) + yield element + else: + for element in iterable: + k = key(element) + if k not in seen: + seen_add(k) + yield element diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_legacy.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_legacy.py new file mode 100644 index 0000000000..b1ea8105da --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/_legacy.py @@ -0,0 +1,120 @@ +import functools +import os +import pathlib +import types +import warnings + +from typing import Union, Iterable, ContextManager, BinaryIO, TextIO, Any + +from . import _common + +Package = Union[types.ModuleType, str] +Resource = str + + +def deprecated(func): + @functools.wraps(func) + def wrapper(*args, **kwargs): + warnings.warn( + f"{func.__name__} is deprecated. Use files() instead. " + "Refer to https://importlib-resources.readthedocs.io" + "/en/latest/using.html#migrating-from-legacy for migration advice.", + DeprecationWarning, + stacklevel=2, + ) + return func(*args, **kwargs) + + return wrapper + + +def normalize_path(path: Any) -> str: + """Normalize a path by ensuring it is a string. + + If the resulting string contains path separators, an exception is raised. + """ + str_path = str(path) + parent, file_name = os.path.split(str_path) + if parent: + raise ValueError(f'{path!r} must be only a file name') + return file_name + + +@deprecated +def open_binary(package: Package, resource: Resource) -> BinaryIO: + """Return a file-like object opened for binary reading of the resource.""" + return (_common.files(package) / normalize_path(resource)).open('rb') + + +@deprecated +def read_binary(package: Package, resource: Resource) -> bytes: + """Return the binary contents of the resource.""" + return (_common.files(package) / normalize_path(resource)).read_bytes() + + +@deprecated +def open_text( + package: Package, + resource: Resource, + encoding: str = 'utf-8', + errors: str = 'strict', +) -> TextIO: + """Return a file-like object opened for text reading of the resource.""" + return (_common.files(package) / normalize_path(resource)).open( + 'r', encoding=encoding, errors=errors + ) + + +@deprecated +def read_text( + package: Package, + resource: Resource, + encoding: str = 'utf-8', + errors: str = 'strict', +) -> str: + """Return the decoded string of the resource. + + The decoding-related arguments have the same semantics as those of + bytes.decode(). + """ + with open_text(package, resource, encoding, errors) as fp: + return fp.read() + + +@deprecated +def contents(package: Package) -> Iterable[str]: + """Return an iterable of entries in `package`. + + Note that not all entries are resources. Specifically, directories are + not considered resources. Use `is_resource()` on each entry returned here + to check if it is a resource or not. + """ + return [path.name for path in _common.files(package).iterdir()] + + +@deprecated +def is_resource(package: Package, name: str) -> bool: + """True if `name` is a resource inside `package`. + + Directories are *not* resources. + """ + resource = normalize_path(name) + return any( + traversable.name == resource and traversable.is_file() + for traversable in _common.files(package).iterdir() + ) + + +@deprecated +def path( + package: Package, + resource: Resource, +) -> ContextManager[pathlib.Path]: + """A context manager providing a file path object to the resource. + + If the resource does not already exist on its own on the file system, + a temporary file will be created. If the file was created, the file + will be deleted upon exiting the context manager (no exception is + raised if the file was deleted prior to the context manager + exiting). + """ + return _common.as_file(_common.files(package) / normalize_path(resource)) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/abc.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/abc.py new file mode 100644 index 0000000000..23b6aeafe4 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/abc.py @@ -0,0 +1,170 @@ +import abc +import io +import itertools +import pathlib +from typing import Any, BinaryIO, Iterable, Iterator, NoReturn, Text, Optional + +from ._compat import runtime_checkable, Protocol, StrPath + + +__all__ = ["ResourceReader", "Traversable", "TraversableResources"] + + +class ResourceReader(metaclass=abc.ABCMeta): + """Abstract base class for loaders to provide resource reading support.""" + + @abc.abstractmethod + def open_resource(self, resource: Text) -> BinaryIO: + """Return an opened, file-like object for binary reading. + + The 'resource' argument is expected to represent only a file name. + If the resource cannot be found, FileNotFoundError is raised. + """ + # This deliberately raises FileNotFoundError instead of + # NotImplementedError so that if this method is accidentally called, + # it'll still do the right thing. + raise FileNotFoundError + + @abc.abstractmethod + def resource_path(self, resource: Text) -> Text: + """Return the file system path to the specified resource. + + The 'resource' argument is expected to represent only a file name. + If the resource does not exist on the file system, raise + FileNotFoundError. + """ + # This deliberately raises FileNotFoundError instead of + # NotImplementedError so that if this method is accidentally called, + # it'll still do the right thing. + raise FileNotFoundError + + @abc.abstractmethod + def is_resource(self, path: Text) -> bool: + """Return True if the named 'path' is a resource. + + Files are resources, directories are not. + """ + raise FileNotFoundError + + @abc.abstractmethod + def contents(self) -> Iterable[str]: + """Return an iterable of entries in `package`.""" + raise FileNotFoundError + + +class TraversalError(Exception): + pass + + +@runtime_checkable +class Traversable(Protocol): + """ + An object with a subset of pathlib.Path methods suitable for + traversing directories and opening files. + + Any exceptions that occur when accessing the backing resource + may propagate unaltered. + """ + + @abc.abstractmethod + def iterdir(self) -> Iterator["Traversable"]: + """ + Yield Traversable objects in self + """ + + def read_bytes(self) -> bytes: + """ + Read contents of self as bytes + """ + with self.open('rb') as strm: + return strm.read() + + def read_text(self, encoding: Optional[str] = None) -> str: + """ + Read contents of self as text + """ + with self.open(encoding=encoding) as strm: + return strm.read() + + @abc.abstractmethod + def is_dir(self) -> bool: + """ + Return True if self is a directory + """ + + @abc.abstractmethod + def is_file(self) -> bool: + """ + Return True if self is a file + """ + + def joinpath(self, *descendants: StrPath) -> "Traversable": + """ + Return Traversable resolved with any descendants applied. + + Each descendant should be a path segment relative to self + and each may contain multiple levels separated by + ``posixpath.sep`` (``/``). + """ + if not descendants: + return self + names = itertools.chain.from_iterable( + path.parts for path in map(pathlib.PurePosixPath, descendants) + ) + target = next(names) + matches = ( + traversable for traversable in self.iterdir() if traversable.name == target + ) + try: + match = next(matches) + except StopIteration: + raise TraversalError( + "Target not found during traversal.", target, list(names) + ) + return match.joinpath(*names) + + def __truediv__(self, child: StrPath) -> "Traversable": + """ + Return Traversable child in self + """ + return self.joinpath(child) + + @abc.abstractmethod + def open(self, mode='r', *args, **kwargs): + """ + mode may be 'r' or 'rb' to open as text or binary. Return a handle + suitable for reading (same as pathlib.Path.open). + + When opening as text, accepts encoding parameters such as those + accepted by io.TextIOWrapper. + """ + + @property + @abc.abstractmethod + def name(self) -> str: + """ + The base name of this object without any parent references. + """ + + +class TraversableResources(ResourceReader): + """ + The required interface for providing traversable + resources. + """ + + @abc.abstractmethod + def files(self) -> "Traversable": + """Return a Traversable object for the loaded package.""" + + def open_resource(self, resource: StrPath) -> io.BufferedReader: + return self.files().joinpath(resource).open('rb') + + def resource_path(self, resource: Any) -> NoReturn: + raise FileNotFoundError(resource) + + def is_resource(self, path: StrPath) -> bool: + return self.files().joinpath(path).is_file() + + def contents(self) -> Iterator[str]: + return (item.name for item in self.files().iterdir()) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/readers.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/readers.py new file mode 100644 index 0000000000..ab34db7409 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/readers.py @@ -0,0 +1,120 @@ +import collections +import pathlib +import operator + +from . import abc + +from ._itertools import unique_everseen +from ._compat import ZipPath + + +def remove_duplicates(items): + return iter(collections.OrderedDict.fromkeys(items)) + + +class FileReader(abc.TraversableResources): + def __init__(self, loader): + self.path = pathlib.Path(loader.path).parent + + def resource_path(self, resource): + """ + Return the file system path to prevent + `resources.path()` from creating a temporary + copy. + """ + return str(self.path.joinpath(resource)) + + def files(self): + return self.path + + +class ZipReader(abc.TraversableResources): + def __init__(self, loader, module): + _, _, name = module.rpartition('.') + self.prefix = loader.prefix.replace('\\', '/') + name + '/' + self.archive = loader.archive + + def open_resource(self, resource): + try: + return super().open_resource(resource) + except KeyError as exc: + raise FileNotFoundError(exc.args[0]) + + def is_resource(self, path): + # workaround for `zipfile.Path.is_file` returning true + # for non-existent paths. + target = self.files().joinpath(path) + return target.is_file() and target.exists() + + def files(self): + return ZipPath(self.archive, self.prefix) + + +class MultiplexedPath(abc.Traversable): + """ + Given a series of Traversable objects, implement a merged + version of the interface across all objects. Useful for + namespace packages which may be multihomed at a single + name. + """ + + def __init__(self, *paths): + self._paths = list(map(pathlib.Path, remove_duplicates(paths))) + if not self._paths: + message = 'MultiplexedPath must contain at least one path' + raise FileNotFoundError(message) + if not all(path.is_dir() for path in self._paths): + raise NotADirectoryError('MultiplexedPath only supports directories') + + def iterdir(self): + files = (file for path in self._paths for file in path.iterdir()) + return unique_everseen(files, key=operator.attrgetter('name')) + + def read_bytes(self): + raise FileNotFoundError(f'{self} is not a file') + + def read_text(self, *args, **kwargs): + raise FileNotFoundError(f'{self} is not a file') + + def is_dir(self): + return True + + def is_file(self): + return False + + def joinpath(self, *descendants): + try: + return super().joinpath(*descendants) + except abc.TraversalError: + # One of the paths did not resolve (a directory does not exist). + # Just return something that will not exist. + return self._paths[0].joinpath(*descendants) + + def open(self, *args, **kwargs): + raise FileNotFoundError(f'{self} is not a file') + + @property + def name(self): + return self._paths[0].name + + def __repr__(self): + paths = ', '.join(f"'{path}'" for path in self._paths) + return f'MultiplexedPath({paths})' + + +class NamespaceReader(abc.TraversableResources): + def __init__(self, namespace_path): + if 'NamespacePath' not in str(namespace_path): + raise ValueError('Invalid path') + self.path = MultiplexedPath(*list(namespace_path)) + + def resource_path(self, resource): + """ + Return the file system path to prevent + `resources.path()` from creating a temporary + copy. + """ + return str(self.path.joinpath(resource)) + + def files(self): + return self.path diff --git a/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/simple.py b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/simple.py new file mode 100644 index 0000000000..7770c922c8 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/importlib_resources/simple.py @@ -0,0 +1,106 @@ +""" +Interface adapters for low-level readers. +""" + +import abc +import io +import itertools +from typing import BinaryIO, List + +from .abc import Traversable, TraversableResources + + +class SimpleReader(abc.ABC): + """ + The minimum, low-level interface required from a resource + provider. + """ + + @property + @abc.abstractmethod + def package(self) -> str: + """ + The name of the package for which this reader loads resources. + """ + + @abc.abstractmethod + def children(self) -> List['SimpleReader']: + """ + Obtain an iterable of SimpleReader for available + child containers (e.g. directories). + """ + + @abc.abstractmethod + def resources(self) -> List[str]: + """ + Obtain available named resources for this virtual package. + """ + + @abc.abstractmethod + def open_binary(self, resource: str) -> BinaryIO: + """ + Obtain a File-like for a named resource. + """ + + @property + def name(self): + return self.package.split('.')[-1] + + +class ResourceContainer(Traversable): + """ + Traversable container for a package's resources via its reader. + """ + + def __init__(self, reader: SimpleReader): + self.reader = reader + + def is_dir(self): + return True + + def is_file(self): + return False + + def iterdir(self): + files = (ResourceHandle(self, name) for name in self.reader.resources) + dirs = map(ResourceContainer, self.reader.children()) + return itertools.chain(files, dirs) + + def open(self, *args, **kwargs): + raise IsADirectoryError() + + +class ResourceHandle(Traversable): + """ + Handle to a named resource in a ResourceReader. + """ + + def __init__(self, parent: ResourceContainer, name: str): + self.parent = parent + self.name = name # type: ignore + + def is_file(self): + return True + + def is_dir(self): + return False + + def open(self, mode='r', *args, **kwargs): + stream = self.parent.reader.open_binary(self.name) + if 'b' not in mode: + stream = io.TextIOWrapper(*args, **kwargs) + return stream + + def joinpath(self, name): + raise RuntimeError("Cannot traverse into a resource") + + +class TraversableReader(TraversableResources, SimpleReader): + """ + A TraversableResources based on SimpleReader. Resource providers + may derive from this class to provide the TraversableResources + interface by supplying the SimpleReader interface. + """ + + def files(self): + return ResourceContainer(self) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/jaraco/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/__init__.py new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/__init__.py diff --git a/third_party/python/setuptools/pkg_resources/_vendor/jaraco/context.py b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/context.py new file mode 100644 index 0000000000..b0d1ef37cb --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/context.py @@ -0,0 +1,288 @@ +import os +import subprocess +import contextlib +import functools +import tempfile +import shutil +import operator +import warnings + + +@contextlib.contextmanager +def pushd(dir): + """ + >>> tmp_path = getfixture('tmp_path') + >>> with pushd(tmp_path): + ... assert os.getcwd() == os.fspath(tmp_path) + >>> assert os.getcwd() != os.fspath(tmp_path) + """ + + orig = os.getcwd() + os.chdir(dir) + try: + yield dir + finally: + os.chdir(orig) + + +@contextlib.contextmanager +def tarball_context(url, target_dir=None, runner=None, pushd=pushd): + """ + Get a tarball, extract it, change to that directory, yield, then + clean up. + `runner` is the function to invoke commands. + `pushd` is a context manager for changing the directory. + """ + if target_dir is None: + target_dir = os.path.basename(url).replace('.tar.gz', '').replace('.tgz', '') + if runner is None: + runner = functools.partial(subprocess.check_call, shell=True) + else: + warnings.warn("runner parameter is deprecated", DeprecationWarning) + # In the tar command, use --strip-components=1 to strip the first path and + # then + # use -C to cause the files to be extracted to {target_dir}. This ensures + # that we always know where the files were extracted. + runner('mkdir {target_dir}'.format(**vars())) + try: + getter = 'wget {url} -O -' + extract = 'tar x{compression} --strip-components=1 -C {target_dir}' + cmd = ' | '.join((getter, extract)) + runner(cmd.format(compression=infer_compression(url), **vars())) + with pushd(target_dir): + yield target_dir + finally: + runner('rm -Rf {target_dir}'.format(**vars())) + + +def infer_compression(url): + """ + Given a URL or filename, infer the compression code for tar. + + >>> infer_compression('http://foo/bar.tar.gz') + 'z' + >>> infer_compression('http://foo/bar.tgz') + 'z' + >>> infer_compression('file.bz') + 'j' + >>> infer_compression('file.xz') + 'J' + """ + # cheat and just assume it's the last two characters + compression_indicator = url[-2:] + mapping = dict(gz='z', bz='j', xz='J') + # Assume 'z' (gzip) if no match + return mapping.get(compression_indicator, 'z') + + +@contextlib.contextmanager +def temp_dir(remover=shutil.rmtree): + """ + Create a temporary directory context. Pass a custom remover + to override the removal behavior. + + >>> import pathlib + >>> with temp_dir() as the_dir: + ... assert os.path.isdir(the_dir) + ... _ = pathlib.Path(the_dir).joinpath('somefile').write_text('contents') + >>> assert not os.path.exists(the_dir) + """ + temp_dir = tempfile.mkdtemp() + try: + yield temp_dir + finally: + remover(temp_dir) + + +@contextlib.contextmanager +def repo_context(url, branch=None, quiet=True, dest_ctx=temp_dir): + """ + Check out the repo indicated by url. + + If dest_ctx is supplied, it should be a context manager + to yield the target directory for the check out. + """ + exe = 'git' if 'git' in url else 'hg' + with dest_ctx() as repo_dir: + cmd = [exe, 'clone', url, repo_dir] + if branch: + cmd.extend(['--branch', branch]) + devnull = open(os.path.devnull, 'w') + stdout = devnull if quiet else None + subprocess.check_call(cmd, stdout=stdout) + yield repo_dir + + +@contextlib.contextmanager +def null(): + """ + A null context suitable to stand in for a meaningful context. + + >>> with null() as value: + ... assert value is None + """ + yield + + +class ExceptionTrap: + """ + A context manager that will catch certain exceptions and provide an + indication they occurred. + + >>> with ExceptionTrap() as trap: + ... raise Exception() + >>> bool(trap) + True + + >>> with ExceptionTrap() as trap: + ... pass + >>> bool(trap) + False + + >>> with ExceptionTrap(ValueError) as trap: + ... raise ValueError("1 + 1 is not 3") + >>> bool(trap) + True + >>> trap.value + ValueError('1 + 1 is not 3') + >>> trap.tb + <traceback object at ...> + + >>> with ExceptionTrap(ValueError) as trap: + ... raise Exception() + Traceback (most recent call last): + ... + Exception + + >>> bool(trap) + False + """ + + exc_info = None, None, None + + def __init__(self, exceptions=(Exception,)): + self.exceptions = exceptions + + def __enter__(self): + return self + + @property + def type(self): + return self.exc_info[0] + + @property + def value(self): + return self.exc_info[1] + + @property + def tb(self): + return self.exc_info[2] + + def __exit__(self, *exc_info): + type = exc_info[0] + matches = type and issubclass(type, self.exceptions) + if matches: + self.exc_info = exc_info + return matches + + def __bool__(self): + return bool(self.type) + + def raises(self, func, *, _test=bool): + """ + Wrap func and replace the result with the truth + value of the trap (True if an exception occurred). + + First, give the decorator an alias to support Python 3.8 + Syntax. + + >>> raises = ExceptionTrap(ValueError).raises + + Now decorate a function that always fails. + + >>> @raises + ... def fail(): + ... raise ValueError('failed') + >>> fail() + True + """ + + @functools.wraps(func) + def wrapper(*args, **kwargs): + with ExceptionTrap(self.exceptions) as trap: + func(*args, **kwargs) + return _test(trap) + + return wrapper + + def passes(self, func): + """ + Wrap func and replace the result with the truth + value of the trap (True if no exception). + + First, give the decorator an alias to support Python 3.8 + Syntax. + + >>> passes = ExceptionTrap(ValueError).passes + + Now decorate a function that always fails. + + >>> @passes + ... def fail(): + ... raise ValueError('failed') + + >>> fail() + False + """ + return self.raises(func, _test=operator.not_) + + +class suppress(contextlib.suppress, contextlib.ContextDecorator): + """ + A version of contextlib.suppress with decorator support. + + >>> @suppress(KeyError) + ... def key_error(): + ... {}[''] + >>> key_error() + """ + + +class on_interrupt(contextlib.ContextDecorator): + """ + Replace a KeyboardInterrupt with SystemExit(1) + + >>> def do_interrupt(): + ... raise KeyboardInterrupt() + >>> on_interrupt('error')(do_interrupt)() + Traceback (most recent call last): + ... + SystemExit: 1 + >>> on_interrupt('error', code=255)(do_interrupt)() + Traceback (most recent call last): + ... + SystemExit: 255 + >>> on_interrupt('suppress')(do_interrupt)() + >>> with __import__('pytest').raises(KeyboardInterrupt): + ... on_interrupt('ignore')(do_interrupt)() + """ + + def __init__( + self, + action='error', + # py3.7 compat + # /, + code=1, + ): + self.action = action + self.code = code + + def __enter__(self): + return self + + def __exit__(self, exctype, excinst, exctb): + if exctype is not KeyboardInterrupt or self.action == 'ignore': + return + elif self.action == 'error': + raise SystemExit(self.code) from excinst + return self.action == 'suppress' diff --git a/third_party/python/setuptools/pkg_resources/_vendor/jaraco/functools.py b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/functools.py new file mode 100644 index 0000000000..67aeadc353 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/functools.py @@ -0,0 +1,556 @@ +import functools +import time +import inspect +import collections +import types +import itertools +import warnings + +import pkg_resources.extern.more_itertools + +from typing import Callable, TypeVar + + +CallableT = TypeVar("CallableT", bound=Callable[..., object]) + + +def compose(*funcs): + """ + Compose any number of unary functions into a single unary function. + + >>> import textwrap + >>> expected = str.strip(textwrap.dedent(compose.__doc__)) + >>> strip_and_dedent = compose(str.strip, textwrap.dedent) + >>> strip_and_dedent(compose.__doc__) == expected + True + + Compose also allows the innermost function to take arbitrary arguments. + + >>> round_three = lambda x: round(x, ndigits=3) + >>> f = compose(round_three, int.__truediv__) + >>> [f(3*x, x+1) for x in range(1,10)] + [1.5, 2.0, 2.25, 2.4, 2.5, 2.571, 2.625, 2.667, 2.7] + """ + + def compose_two(f1, f2): + return lambda *args, **kwargs: f1(f2(*args, **kwargs)) + + return functools.reduce(compose_two, funcs) + + +def method_caller(method_name, *args, **kwargs): + """ + Return a function that will call a named method on the + target object with optional positional and keyword + arguments. + + >>> lower = method_caller('lower') + >>> lower('MyString') + 'mystring' + """ + + def call_method(target): + func = getattr(target, method_name) + return func(*args, **kwargs) + + return call_method + + +def once(func): + """ + Decorate func so it's only ever called the first time. + + This decorator can ensure that an expensive or non-idempotent function + will not be expensive on subsequent calls and is idempotent. + + >>> add_three = once(lambda a: a+3) + >>> add_three(3) + 6 + >>> add_three(9) + 6 + >>> add_three('12') + 6 + + To reset the stored value, simply clear the property ``saved_result``. + + >>> del add_three.saved_result + >>> add_three(9) + 12 + >>> add_three(8) + 12 + + Or invoke 'reset()' on it. + + >>> add_three.reset() + >>> add_three(-3) + 0 + >>> add_three(0) + 0 + """ + + @functools.wraps(func) + def wrapper(*args, **kwargs): + if not hasattr(wrapper, 'saved_result'): + wrapper.saved_result = func(*args, **kwargs) + return wrapper.saved_result + + wrapper.reset = lambda: vars(wrapper).__delitem__('saved_result') + return wrapper + + +def method_cache( + method: CallableT, + cache_wrapper: Callable[ + [CallableT], CallableT + ] = functools.lru_cache(), # type: ignore[assignment] +) -> CallableT: + """ + Wrap lru_cache to support storing the cache data in the object instances. + + Abstracts the common paradigm where the method explicitly saves an + underscore-prefixed protected property on first call and returns that + subsequently. + + >>> class MyClass: + ... calls = 0 + ... + ... @method_cache + ... def method(self, value): + ... self.calls += 1 + ... return value + + >>> a = MyClass() + >>> a.method(3) + 3 + >>> for x in range(75): + ... res = a.method(x) + >>> a.calls + 75 + + Note that the apparent behavior will be exactly like that of lru_cache + except that the cache is stored on each instance, so values in one + instance will not flush values from another, and when an instance is + deleted, so are the cached values for that instance. + + >>> b = MyClass() + >>> for x in range(35): + ... res = b.method(x) + >>> b.calls + 35 + >>> a.method(0) + 0 + >>> a.calls + 75 + + Note that if method had been decorated with ``functools.lru_cache()``, + a.calls would have been 76 (due to the cached value of 0 having been + flushed by the 'b' instance). + + Clear the cache with ``.cache_clear()`` + + >>> a.method.cache_clear() + + Same for a method that hasn't yet been called. + + >>> c = MyClass() + >>> c.method.cache_clear() + + Another cache wrapper may be supplied: + + >>> cache = functools.lru_cache(maxsize=2) + >>> MyClass.method2 = method_cache(lambda self: 3, cache_wrapper=cache) + >>> a = MyClass() + >>> a.method2() + 3 + + Caution - do not subsequently wrap the method with another decorator, such + as ``@property``, which changes the semantics of the function. + + See also + http://code.activestate.com/recipes/577452-a-memoize-decorator-for-instance-methods/ + for another implementation and additional justification. + """ + + def wrapper(self: object, *args: object, **kwargs: object) -> object: + # it's the first call, replace the method with a cached, bound method + bound_method: CallableT = types.MethodType( # type: ignore[assignment] + method, self + ) + cached_method = cache_wrapper(bound_method) + setattr(self, method.__name__, cached_method) + return cached_method(*args, **kwargs) + + # Support cache clear even before cache has been created. + wrapper.cache_clear = lambda: None # type: ignore[attr-defined] + + return ( # type: ignore[return-value] + _special_method_cache(method, cache_wrapper) or wrapper + ) + + +def _special_method_cache(method, cache_wrapper): + """ + Because Python treats special methods differently, it's not + possible to use instance attributes to implement the cached + methods. + + Instead, install the wrapper method under a different name + and return a simple proxy to that wrapper. + + https://github.com/jaraco/jaraco.functools/issues/5 + """ + name = method.__name__ + special_names = '__getattr__', '__getitem__' + if name not in special_names: + return + + wrapper_name = '__cached' + name + + def proxy(self, *args, **kwargs): + if wrapper_name not in vars(self): + bound = types.MethodType(method, self) + cache = cache_wrapper(bound) + setattr(self, wrapper_name, cache) + else: + cache = getattr(self, wrapper_name) + return cache(*args, **kwargs) + + return proxy + + +def apply(transform): + """ + Decorate a function with a transform function that is + invoked on results returned from the decorated function. + + >>> @apply(reversed) + ... def get_numbers(start): + ... "doc for get_numbers" + ... return range(start, start+3) + >>> list(get_numbers(4)) + [6, 5, 4] + >>> get_numbers.__doc__ + 'doc for get_numbers' + """ + + def wrap(func): + return functools.wraps(func)(compose(transform, func)) + + return wrap + + +def result_invoke(action): + r""" + Decorate a function with an action function that is + invoked on the results returned from the decorated + function (for its side-effect), then return the original + result. + + >>> @result_invoke(print) + ... def add_two(a, b): + ... return a + b + >>> x = add_two(2, 3) + 5 + >>> x + 5 + """ + + def wrap(func): + @functools.wraps(func) + def wrapper(*args, **kwargs): + result = func(*args, **kwargs) + action(result) + return result + + return wrapper + + return wrap + + +def invoke(f, *args, **kwargs): + """ + Call a function for its side effect after initialization. + + The benefit of using the decorator instead of simply invoking a function + after defining it is that it makes explicit the author's intent for the + function to be called immediately. Whereas if one simply calls the + function immediately, it's less obvious if that was intentional or + incidental. It also avoids repeating the name - the two actions, defining + the function and calling it immediately are modeled separately, but linked + by the decorator construct. + + The benefit of having a function construct (opposed to just invoking some + behavior inline) is to serve as a scope in which the behavior occurs. It + avoids polluting the global namespace with local variables, provides an + anchor on which to attach documentation (docstring), keeps the behavior + logically separated (instead of conceptually separated or not separated at + all), and provides potential to re-use the behavior for testing or other + purposes. + + This function is named as a pithy way to communicate, "call this function + primarily for its side effect", or "while defining this function, also + take it aside and call it". It exists because there's no Python construct + for "define and call" (nor should there be, as decorators serve this need + just fine). The behavior happens immediately and synchronously. + + >>> @invoke + ... def func(): print("called") + called + >>> func() + called + + Use functools.partial to pass parameters to the initial call + + >>> @functools.partial(invoke, name='bingo') + ... def func(name): print("called with", name) + called with bingo + """ + f(*args, **kwargs) + return f + + +def call_aside(*args, **kwargs): + """ + Deprecated name for invoke. + """ + warnings.warn("call_aside is deprecated, use invoke", DeprecationWarning) + return invoke(*args, **kwargs) + + +class Throttler: + """ + Rate-limit a function (or other callable) + """ + + def __init__(self, func, max_rate=float('Inf')): + if isinstance(func, Throttler): + func = func.func + self.func = func + self.max_rate = max_rate + self.reset() + + def reset(self): + self.last_called = 0 + + def __call__(self, *args, **kwargs): + self._wait() + return self.func(*args, **kwargs) + + def _wait(self): + "ensure at least 1/max_rate seconds from last call" + elapsed = time.time() - self.last_called + must_wait = 1 / self.max_rate - elapsed + time.sleep(max(0, must_wait)) + self.last_called = time.time() + + def __get__(self, obj, type=None): + return first_invoke(self._wait, functools.partial(self.func, obj)) + + +def first_invoke(func1, func2): + """ + Return a function that when invoked will invoke func1 without + any parameters (for its side-effect) and then invoke func2 + with whatever parameters were passed, returning its result. + """ + + def wrapper(*args, **kwargs): + func1() + return func2(*args, **kwargs) + + return wrapper + + +def retry_call(func, cleanup=lambda: None, retries=0, trap=()): + """ + Given a callable func, trap the indicated exceptions + for up to 'retries' times, invoking cleanup on the + exception. On the final attempt, allow any exceptions + to propagate. + """ + attempts = itertools.count() if retries == float('inf') else range(retries) + for attempt in attempts: + try: + return func() + except trap: + cleanup() + + return func() + + +def retry(*r_args, **r_kwargs): + """ + Decorator wrapper for retry_call. Accepts arguments to retry_call + except func and then returns a decorator for the decorated function. + + Ex: + + >>> @retry(retries=3) + ... def my_func(a, b): + ... "this is my funk" + ... print(a, b) + >>> my_func.__doc__ + 'this is my funk' + """ + + def decorate(func): + @functools.wraps(func) + def wrapper(*f_args, **f_kwargs): + bound = functools.partial(func, *f_args, **f_kwargs) + return retry_call(bound, *r_args, **r_kwargs) + + return wrapper + + return decorate + + +def print_yielded(func): + """ + Convert a generator into a function that prints all yielded elements + + >>> @print_yielded + ... def x(): + ... yield 3; yield None + >>> x() + 3 + None + """ + print_all = functools.partial(map, print) + print_results = compose(more_itertools.consume, print_all, func) + return functools.wraps(func)(print_results) + + +def pass_none(func): + """ + Wrap func so it's not called if its first param is None + + >>> print_text = pass_none(print) + >>> print_text('text') + text + >>> print_text(None) + """ + + @functools.wraps(func) + def wrapper(param, *args, **kwargs): + if param is not None: + return func(param, *args, **kwargs) + + return wrapper + + +def assign_params(func, namespace): + """ + Assign parameters from namespace where func solicits. + + >>> def func(x, y=3): + ... print(x, y) + >>> assigned = assign_params(func, dict(x=2, z=4)) + >>> assigned() + 2 3 + + The usual errors are raised if a function doesn't receive + its required parameters: + + >>> assigned = assign_params(func, dict(y=3, z=4)) + >>> assigned() + Traceback (most recent call last): + TypeError: func() ...argument... + + It even works on methods: + + >>> class Handler: + ... def meth(self, arg): + ... print(arg) + >>> assign_params(Handler().meth, dict(arg='crystal', foo='clear'))() + crystal + """ + sig = inspect.signature(func) + params = sig.parameters.keys() + call_ns = {k: namespace[k] for k in params if k in namespace} + return functools.partial(func, **call_ns) + + +def save_method_args(method): + """ + Wrap a method such that when it is called, the args and kwargs are + saved on the method. + + >>> class MyClass: + ... @save_method_args + ... def method(self, a, b): + ... print(a, b) + >>> my_ob = MyClass() + >>> my_ob.method(1, 2) + 1 2 + >>> my_ob._saved_method.args + (1, 2) + >>> my_ob._saved_method.kwargs + {} + >>> my_ob.method(a=3, b='foo') + 3 foo + >>> my_ob._saved_method.args + () + >>> my_ob._saved_method.kwargs == dict(a=3, b='foo') + True + + The arguments are stored on the instance, allowing for + different instance to save different args. + + >>> your_ob = MyClass() + >>> your_ob.method({str('x'): 3}, b=[4]) + {'x': 3} [4] + >>> your_ob._saved_method.args + ({'x': 3},) + >>> my_ob._saved_method.args + () + """ + args_and_kwargs = collections.namedtuple('args_and_kwargs', 'args kwargs') + + @functools.wraps(method) + def wrapper(self, *args, **kwargs): + attr_name = '_saved_' + method.__name__ + attr = args_and_kwargs(args, kwargs) + setattr(self, attr_name, attr) + return method(self, *args, **kwargs) + + return wrapper + + +def except_(*exceptions, replace=None, use=None): + """ + Replace the indicated exceptions, if raised, with the indicated + literal replacement or evaluated expression (if present). + + >>> safe_int = except_(ValueError)(int) + >>> safe_int('five') + >>> safe_int('5') + 5 + + Specify a literal replacement with ``replace``. + + >>> safe_int_r = except_(ValueError, replace=0)(int) + >>> safe_int_r('five') + 0 + + Provide an expression to ``use`` to pass through particular parameters. + + >>> safe_int_pt = except_(ValueError, use='args[0]')(int) + >>> safe_int_pt('five') + 'five' + + """ + + def decorate(func): + @functools.wraps(func) + def wrapper(*args, **kwargs): + try: + return func(*args, **kwargs) + except exceptions: + try: + return eval(use) + except TypeError: + return replace + + return wrapper + + return decorate diff --git a/third_party/python/setuptools/pkg_resources/_vendor/jaraco/text/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/text/__init__.py new file mode 100644 index 0000000000..c466378ceb --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/jaraco/text/__init__.py @@ -0,0 +1,599 @@ +import re +import itertools +import textwrap +import functools + +try: + from importlib.resources import files # type: ignore +except ImportError: # pragma: nocover + from pkg_resources.extern.importlib_resources import files # type: ignore + +from pkg_resources.extern.jaraco.functools import compose, method_cache +from pkg_resources.extern.jaraco.context import ExceptionTrap + + +def substitution(old, new): + """ + Return a function that will perform a substitution on a string + """ + return lambda s: s.replace(old, new) + + +def multi_substitution(*substitutions): + """ + Take a sequence of pairs specifying substitutions, and create + a function that performs those substitutions. + + >>> multi_substitution(('foo', 'bar'), ('bar', 'baz'))('foo') + 'baz' + """ + substitutions = itertools.starmap(substitution, substitutions) + # compose function applies last function first, so reverse the + # substitutions to get the expected order. + substitutions = reversed(tuple(substitutions)) + return compose(*substitutions) + + +class FoldedCase(str): + """ + A case insensitive string class; behaves just like str + except compares equal when the only variation is case. + + >>> s = FoldedCase('hello world') + + >>> s == 'Hello World' + True + + >>> 'Hello World' == s + True + + >>> s != 'Hello World' + False + + >>> s.index('O') + 4 + + >>> s.split('O') + ['hell', ' w', 'rld'] + + >>> sorted(map(FoldedCase, ['GAMMA', 'alpha', 'Beta'])) + ['alpha', 'Beta', 'GAMMA'] + + Sequence membership is straightforward. + + >>> "Hello World" in [s] + True + >>> s in ["Hello World"] + True + + You may test for set inclusion, but candidate and elements + must both be folded. + + >>> FoldedCase("Hello World") in {s} + True + >>> s in {FoldedCase("Hello World")} + True + + String inclusion works as long as the FoldedCase object + is on the right. + + >>> "hello" in FoldedCase("Hello World") + True + + But not if the FoldedCase object is on the left: + + >>> FoldedCase('hello') in 'Hello World' + False + + In that case, use ``in_``: + + >>> FoldedCase('hello').in_('Hello World') + True + + >>> FoldedCase('hello') > FoldedCase('Hello') + False + """ + + def __lt__(self, other): + return self.lower() < other.lower() + + def __gt__(self, other): + return self.lower() > other.lower() + + def __eq__(self, other): + return self.lower() == other.lower() + + def __ne__(self, other): + return self.lower() != other.lower() + + def __hash__(self): + return hash(self.lower()) + + def __contains__(self, other): + return super().lower().__contains__(other.lower()) + + def in_(self, other): + "Does self appear in other?" + return self in FoldedCase(other) + + # cache lower since it's likely to be called frequently. + @method_cache + def lower(self): + return super().lower() + + def index(self, sub): + return self.lower().index(sub.lower()) + + def split(self, splitter=' ', maxsplit=0): + pattern = re.compile(re.escape(splitter), re.I) + return pattern.split(self, maxsplit) + + +# Python 3.8 compatibility +_unicode_trap = ExceptionTrap(UnicodeDecodeError) + + +@_unicode_trap.passes +def is_decodable(value): + r""" + Return True if the supplied value is decodable (using the default + encoding). + + >>> is_decodable(b'\xff') + False + >>> is_decodable(b'\x32') + True + """ + value.decode() + + +def is_binary(value): + r""" + Return True if the value appears to be binary (that is, it's a byte + string and isn't decodable). + + >>> is_binary(b'\xff') + True + >>> is_binary('\xff') + False + """ + return isinstance(value, bytes) and not is_decodable(value) + + +def trim(s): + r""" + Trim something like a docstring to remove the whitespace that + is common due to indentation and formatting. + + >>> trim("\n\tfoo = bar\n\t\tbar = baz\n") + 'foo = bar\n\tbar = baz' + """ + return textwrap.dedent(s).strip() + + +def wrap(s): + """ + Wrap lines of text, retaining existing newlines as + paragraph markers. + + >>> print(wrap(lorem_ipsum)) + Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do + eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad + minim veniam, quis nostrud exercitation ullamco laboris nisi ut + aliquip ex ea commodo consequat. Duis aute irure dolor in + reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla + pariatur. Excepteur sint occaecat cupidatat non proident, sunt in + culpa qui officia deserunt mollit anim id est laborum. + <BLANKLINE> + Curabitur pretium tincidunt lacus. Nulla gravida orci a odio. Nullam + varius, turpis et commodo pharetra, est eros bibendum elit, nec luctus + magna felis sollicitudin mauris. Integer in mauris eu nibh euismod + gravida. Duis ac tellus et risus vulputate vehicula. Donec lobortis + risus a elit. Etiam tempor. Ut ullamcorper, ligula eu tempor congue, + eros est euismod turpis, id tincidunt sapien risus a quam. Maecenas + fermentum consequat mi. Donec fermentum. Pellentesque malesuada nulla + a mi. Duis sapien sem, aliquet nec, commodo eget, consequat quis, + neque. Aliquam faucibus, elit ut dictum aliquet, felis nisl adipiscing + sapien, sed malesuada diam lacus eget erat. Cras mollis scelerisque + nunc. Nullam arcu. Aliquam consequat. Curabitur augue lorem, dapibus + quis, laoreet et, pretium ac, nisi. Aenean magna nisl, mollis quis, + molestie eu, feugiat in, orci. In hac habitasse platea dictumst. + """ + paragraphs = s.splitlines() + wrapped = ('\n'.join(textwrap.wrap(para)) for para in paragraphs) + return '\n\n'.join(wrapped) + + +def unwrap(s): + r""" + Given a multi-line string, return an unwrapped version. + + >>> wrapped = wrap(lorem_ipsum) + >>> wrapped.count('\n') + 20 + >>> unwrapped = unwrap(wrapped) + >>> unwrapped.count('\n') + 1 + >>> print(unwrapped) + Lorem ipsum dolor sit amet, consectetur adipiscing ... + Curabitur pretium tincidunt lacus. Nulla gravida orci ... + + """ + paragraphs = re.split(r'\n\n+', s) + cleaned = (para.replace('\n', ' ') for para in paragraphs) + return '\n'.join(cleaned) + + + + +class Splitter(object): + """object that will split a string with the given arguments for each call + + >>> s = Splitter(',') + >>> s('hello, world, this is your, master calling') + ['hello', ' world', ' this is your', ' master calling'] + """ + + def __init__(self, *args): + self.args = args + + def __call__(self, s): + return s.split(*self.args) + + +def indent(string, prefix=' ' * 4): + """ + >>> indent('foo') + ' foo' + """ + return prefix + string + + +class WordSet(tuple): + """ + Given an identifier, return the words that identifier represents, + whether in camel case, underscore-separated, etc. + + >>> WordSet.parse("camelCase") + ('camel', 'Case') + + >>> WordSet.parse("under_sep") + ('under', 'sep') + + Acronyms should be retained + + >>> WordSet.parse("firstSNL") + ('first', 'SNL') + + >>> WordSet.parse("you_and_I") + ('you', 'and', 'I') + + >>> WordSet.parse("A simple test") + ('A', 'simple', 'test') + + Multiple caps should not interfere with the first cap of another word. + + >>> WordSet.parse("myABCClass") + ('my', 'ABC', 'Class') + + The result is a WordSet, so you can get the form you need. + + >>> WordSet.parse("myABCClass").underscore_separated() + 'my_ABC_Class' + + >>> WordSet.parse('a-command').camel_case() + 'ACommand' + + >>> WordSet.parse('someIdentifier').lowered().space_separated() + 'some identifier' + + Slices of the result should return another WordSet. + + >>> WordSet.parse('taken-out-of-context')[1:].underscore_separated() + 'out_of_context' + + >>> WordSet.from_class_name(WordSet()).lowered().space_separated() + 'word set' + + >>> example = WordSet.parse('figured it out') + >>> example.headless_camel_case() + 'figuredItOut' + >>> example.dash_separated() + 'figured-it-out' + + """ + + _pattern = re.compile('([A-Z]?[a-z]+)|([A-Z]+(?![a-z]))') + + def capitalized(self): + return WordSet(word.capitalize() for word in self) + + def lowered(self): + return WordSet(word.lower() for word in self) + + def camel_case(self): + return ''.join(self.capitalized()) + + def headless_camel_case(self): + words = iter(self) + first = next(words).lower() + new_words = itertools.chain((first,), WordSet(words).camel_case()) + return ''.join(new_words) + + def underscore_separated(self): + return '_'.join(self) + + def dash_separated(self): + return '-'.join(self) + + def space_separated(self): + return ' '.join(self) + + def trim_right(self, item): + """ + Remove the item from the end of the set. + + >>> WordSet.parse('foo bar').trim_right('foo') + ('foo', 'bar') + >>> WordSet.parse('foo bar').trim_right('bar') + ('foo',) + >>> WordSet.parse('').trim_right('bar') + () + """ + return self[:-1] if self and self[-1] == item else self + + def trim_left(self, item): + """ + Remove the item from the beginning of the set. + + >>> WordSet.parse('foo bar').trim_left('foo') + ('bar',) + >>> WordSet.parse('foo bar').trim_left('bar') + ('foo', 'bar') + >>> WordSet.parse('').trim_left('bar') + () + """ + return self[1:] if self and self[0] == item else self + + def trim(self, item): + """ + >>> WordSet.parse('foo bar').trim('foo') + ('bar',) + """ + return self.trim_left(item).trim_right(item) + + def __getitem__(self, item): + result = super(WordSet, self).__getitem__(item) + if isinstance(item, slice): + result = WordSet(result) + return result + + @classmethod + def parse(cls, identifier): + matches = cls._pattern.finditer(identifier) + return WordSet(match.group(0) for match in matches) + + @classmethod + def from_class_name(cls, subject): + return cls.parse(subject.__class__.__name__) + + +# for backward compatibility +words = WordSet.parse + + +def simple_html_strip(s): + r""" + Remove HTML from the string `s`. + + >>> str(simple_html_strip('')) + '' + + >>> print(simple_html_strip('A <bold>stormy</bold> day in paradise')) + A stormy day in paradise + + >>> print(simple_html_strip('Somebody <!-- do not --> tell the truth.')) + Somebody tell the truth. + + >>> print(simple_html_strip('What about<br/>\nmultiple lines?')) + What about + multiple lines? + """ + html_stripper = re.compile('(<!--.*?-->)|(<[^>]*>)|([^<]+)', re.DOTALL) + texts = (match.group(3) or '' for match in html_stripper.finditer(s)) + return ''.join(texts) + + +class SeparatedValues(str): + """ + A string separated by a separator. Overrides __iter__ for getting + the values. + + >>> list(SeparatedValues('a,b,c')) + ['a', 'b', 'c'] + + Whitespace is stripped and empty values are discarded. + + >>> list(SeparatedValues(' a, b , c, ')) + ['a', 'b', 'c'] + """ + + separator = ',' + + def __iter__(self): + parts = self.split(self.separator) + return filter(None, (part.strip() for part in parts)) + + +class Stripper: + r""" + Given a series of lines, find the common prefix and strip it from them. + + >>> lines = [ + ... 'abcdefg\n', + ... 'abc\n', + ... 'abcde\n', + ... ] + >>> res = Stripper.strip_prefix(lines) + >>> res.prefix + 'abc' + >>> list(res.lines) + ['defg\n', '\n', 'de\n'] + + If no prefix is common, nothing should be stripped. + + >>> lines = [ + ... 'abcd\n', + ... '1234\n', + ... ] + >>> res = Stripper.strip_prefix(lines) + >>> res.prefix = '' + >>> list(res.lines) + ['abcd\n', '1234\n'] + """ + + def __init__(self, prefix, lines): + self.prefix = prefix + self.lines = map(self, lines) + + @classmethod + def strip_prefix(cls, lines): + prefix_lines, lines = itertools.tee(lines) + prefix = functools.reduce(cls.common_prefix, prefix_lines) + return cls(prefix, lines) + + def __call__(self, line): + if not self.prefix: + return line + null, prefix, rest = line.partition(self.prefix) + return rest + + @staticmethod + def common_prefix(s1, s2): + """ + Return the common prefix of two lines. + """ + index = min(len(s1), len(s2)) + while s1[:index] != s2[:index]: + index -= 1 + return s1[:index] + + +def remove_prefix(text, prefix): + """ + Remove the prefix from the text if it exists. + + >>> remove_prefix('underwhelming performance', 'underwhelming ') + 'performance' + + >>> remove_prefix('something special', 'sample') + 'something special' + """ + null, prefix, rest = text.rpartition(prefix) + return rest + + +def remove_suffix(text, suffix): + """ + Remove the suffix from the text if it exists. + + >>> remove_suffix('name.git', '.git') + 'name' + + >>> remove_suffix('something special', 'sample') + 'something special' + """ + rest, suffix, null = text.partition(suffix) + return rest + + +def normalize_newlines(text): + r""" + Replace alternate newlines with the canonical newline. + + >>> normalize_newlines('Lorem Ipsum\u2029') + 'Lorem Ipsum\n' + >>> normalize_newlines('Lorem Ipsum\r\n') + 'Lorem Ipsum\n' + >>> normalize_newlines('Lorem Ipsum\x85') + 'Lorem Ipsum\n' + """ + newlines = ['\r\n', '\r', '\n', '\u0085', '\u2028', '\u2029'] + pattern = '|'.join(newlines) + return re.sub(pattern, '\n', text) + + +def _nonblank(str): + return str and not str.startswith('#') + + +@functools.singledispatch +def yield_lines(iterable): + r""" + Yield valid lines of a string or iterable. + + >>> list(yield_lines('')) + [] + >>> list(yield_lines(['foo', 'bar'])) + ['foo', 'bar'] + >>> list(yield_lines('foo\nbar')) + ['foo', 'bar'] + >>> list(yield_lines('\nfoo\n#bar\nbaz #comment')) + ['foo', 'baz #comment'] + >>> list(yield_lines(['foo\nbar', 'baz', 'bing\n\n\n'])) + ['foo', 'bar', 'baz', 'bing'] + """ + return itertools.chain.from_iterable(map(yield_lines, iterable)) + + +@yield_lines.register(str) +def _(text): + return filter(_nonblank, map(str.strip, text.splitlines())) + + +def drop_comment(line): + """ + Drop comments. + + >>> drop_comment('foo # bar') + 'foo' + + A hash without a space may be in a URL. + + >>> drop_comment('http://example.com/foo#bar') + 'http://example.com/foo#bar' + """ + return line.partition(' #')[0] + + +def join_continuation(lines): + r""" + Join lines continued by a trailing backslash. + + >>> list(join_continuation(['foo \\', 'bar', 'baz'])) + ['foobar', 'baz'] + >>> list(join_continuation(['foo \\', 'bar', 'baz'])) + ['foobar', 'baz'] + >>> list(join_continuation(['foo \\', 'bar \\', 'baz'])) + ['foobarbaz'] + + Not sure why, but... + The character preceeding the backslash is also elided. + + >>> list(join_continuation(['goo\\', 'dly'])) + ['godly'] + + A terrible idea, but... + If no line is available to continue, suppress the lines. + + >>> list(join_continuation(['foo', 'bar\\', 'baz\\'])) + ['foo'] + """ + lines = iter(lines) + for item in lines: + while item.endswith('\\'): + try: + item = item[:-2].strip() + next(lines) + except StopIteration: + return + yield item diff --git a/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/__init__.py new file mode 100644 index 0000000000..66443971df --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/__init__.py @@ -0,0 +1,6 @@ +"""More routines for operating on iterables, beyond itertools""" + +from .more import * # noqa +from .recipes import * # noqa + +__version__ = '9.1.0' diff --git a/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/more.py b/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/more.py new file mode 100644 index 0000000000..e0e2d3de92 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/more.py @@ -0,0 +1,4391 @@ +import warnings + +from collections import Counter, defaultdict, deque, abc +from collections.abc import Sequence +from functools import partial, reduce, wraps +from heapq import heapify, heapreplace, heappop +from itertools import ( + chain, + compress, + count, + cycle, + dropwhile, + groupby, + islice, + repeat, + starmap, + takewhile, + tee, + zip_longest, +) +from math import exp, factorial, floor, log +from queue import Empty, Queue +from random import random, randrange, uniform +from operator import itemgetter, mul, sub, gt, lt, ge, le +from sys import hexversion, maxsize +from time import monotonic + +from .recipes import ( + _marker, + _zip_equal, + UnequalIterablesError, + consume, + flatten, + pairwise, + powerset, + take, + unique_everseen, + all_equal, +) + +__all__ = [ + 'AbortThread', + 'SequenceView', + 'UnequalIterablesError', + 'adjacent', + 'all_unique', + 'always_iterable', + 'always_reversible', + 'bucket', + 'callback_iter', + 'chunked', + 'chunked_even', + 'circular_shifts', + 'collapse', + 'combination_index', + 'consecutive_groups', + 'constrained_batches', + 'consumer', + 'count_cycle', + 'countable', + 'difference', + 'distinct_combinations', + 'distinct_permutations', + 'distribute', + 'divide', + 'duplicates_everseen', + 'duplicates_justseen', + 'exactly_n', + 'filter_except', + 'first', + 'gray_product', + 'groupby_transform', + 'ichunked', + 'iequals', + 'ilen', + 'interleave', + 'interleave_evenly', + 'interleave_longest', + 'intersperse', + 'is_sorted', + 'islice_extended', + 'iterate', + 'last', + 'locate', + 'longest_common_prefix', + 'lstrip', + 'make_decorator', + 'map_except', + 'map_if', + 'map_reduce', + 'mark_ends', + 'minmax', + 'nth_or_last', + 'nth_permutation', + 'nth_product', + 'numeric_range', + 'one', + 'only', + 'padded', + 'partitions', + 'peekable', + 'permutation_index', + 'product_index', + 'raise_', + 'repeat_each', + 'repeat_last', + 'replace', + 'rlocate', + 'rstrip', + 'run_length', + 'sample', + 'seekable', + 'set_partitions', + 'side_effect', + 'sliced', + 'sort_together', + 'split_after', + 'split_at', + 'split_before', + 'split_into', + 'split_when', + 'spy', + 'stagger', + 'strip', + 'strictly_n', + 'substrings', + 'substrings_indexes', + 'time_limited', + 'unique_in_window', + 'unique_to_each', + 'unzip', + 'value_chain', + 'windowed', + 'windowed_complete', + 'with_iter', + 'zip_broadcast', + 'zip_equal', + 'zip_offset', +] + + +def chunked(iterable, n, strict=False): + """Break *iterable* into lists of length *n*: + + >>> list(chunked([1, 2, 3, 4, 5, 6], 3)) + [[1, 2, 3], [4, 5, 6]] + + By the default, the last yielded list will have fewer than *n* elements + if the length of *iterable* is not divisible by *n*: + + >>> list(chunked([1, 2, 3, 4, 5, 6, 7, 8], 3)) + [[1, 2, 3], [4, 5, 6], [7, 8]] + + To use a fill-in value instead, see the :func:`grouper` recipe. + + If the length of *iterable* is not divisible by *n* and *strict* is + ``True``, then ``ValueError`` will be raised before the last + list is yielded. + + """ + iterator = iter(partial(take, n, iter(iterable)), []) + if strict: + if n is None: + raise ValueError('n must not be None when using strict mode.') + + def ret(): + for chunk in iterator: + if len(chunk) != n: + raise ValueError('iterable is not divisible by n.') + yield chunk + + return iter(ret()) + else: + return iterator + + +def first(iterable, default=_marker): + """Return the first item of *iterable*, or *default* if *iterable* is + empty. + + >>> first([0, 1, 2, 3]) + 0 + >>> first([], 'some default') + 'some default' + + If *default* is not provided and there are no items in the iterable, + raise ``ValueError``. + + :func:`first` is useful when you have a generator of expensive-to-retrieve + values and want any arbitrary one. It is marginally shorter than + ``next(iter(iterable), default)``. + + """ + try: + return next(iter(iterable)) + except StopIteration as e: + if default is _marker: + raise ValueError( + 'first() was called on an empty iterable, and no ' + 'default value was provided.' + ) from e + return default + + +def last(iterable, default=_marker): + """Return the last item of *iterable*, or *default* if *iterable* is + empty. + + >>> last([0, 1, 2, 3]) + 3 + >>> last([], 'some default') + 'some default' + + If *default* is not provided and there are no items in the iterable, + raise ``ValueError``. + """ + try: + if isinstance(iterable, Sequence): + return iterable[-1] + # Work around https://bugs.python.org/issue38525 + elif hasattr(iterable, '__reversed__') and (hexversion != 0x030800F0): + return next(reversed(iterable)) + else: + return deque(iterable, maxlen=1)[-1] + except (IndexError, TypeError, StopIteration): + if default is _marker: + raise ValueError( + 'last() was called on an empty iterable, and no default was ' + 'provided.' + ) + return default + + +def nth_or_last(iterable, n, default=_marker): + """Return the nth or the last item of *iterable*, + or *default* if *iterable* is empty. + + >>> nth_or_last([0, 1, 2, 3], 2) + 2 + >>> nth_or_last([0, 1], 2) + 1 + >>> nth_or_last([], 0, 'some default') + 'some default' + + If *default* is not provided and there are no items in the iterable, + raise ``ValueError``. + """ + return last(islice(iterable, n + 1), default=default) + + +class peekable: + """Wrap an iterator to allow lookahead and prepending elements. + + Call :meth:`peek` on the result to get the value that will be returned + by :func:`next`. This won't advance the iterator: + + >>> p = peekable(['a', 'b']) + >>> p.peek() + 'a' + >>> next(p) + 'a' + + Pass :meth:`peek` a default value to return that instead of raising + ``StopIteration`` when the iterator is exhausted. + + >>> p = peekable([]) + >>> p.peek('hi') + 'hi' + + peekables also offer a :meth:`prepend` method, which "inserts" items + at the head of the iterable: + + >>> p = peekable([1, 2, 3]) + >>> p.prepend(10, 11, 12) + >>> next(p) + 10 + >>> p.peek() + 11 + >>> list(p) + [11, 12, 1, 2, 3] + + peekables can be indexed. Index 0 is the item that will be returned by + :func:`next`, index 1 is the item after that, and so on: + The values up to the given index will be cached. + + >>> p = peekable(['a', 'b', 'c', 'd']) + >>> p[0] + 'a' + >>> p[1] + 'b' + >>> next(p) + 'a' + + Negative indexes are supported, but be aware that they will cache the + remaining items in the source iterator, which may require significant + storage. + + To check whether a peekable is exhausted, check its truth value: + + >>> p = peekable(['a', 'b']) + >>> if p: # peekable has items + ... list(p) + ['a', 'b'] + >>> if not p: # peekable is exhausted + ... list(p) + [] + + """ + + def __init__(self, iterable): + self._it = iter(iterable) + self._cache = deque() + + def __iter__(self): + return self + + def __bool__(self): + try: + self.peek() + except StopIteration: + return False + return True + + def peek(self, default=_marker): + """Return the item that will be next returned from ``next()``. + + Return ``default`` if there are no items left. If ``default`` is not + provided, raise ``StopIteration``. + + """ + if not self._cache: + try: + self._cache.append(next(self._it)) + except StopIteration: + if default is _marker: + raise + return default + return self._cache[0] + + def prepend(self, *items): + """Stack up items to be the next ones returned from ``next()`` or + ``self.peek()``. The items will be returned in + first in, first out order:: + + >>> p = peekable([1, 2, 3]) + >>> p.prepend(10, 11, 12) + >>> next(p) + 10 + >>> list(p) + [11, 12, 1, 2, 3] + + It is possible, by prepending items, to "resurrect" a peekable that + previously raised ``StopIteration``. + + >>> p = peekable([]) + >>> next(p) + Traceback (most recent call last): + ... + StopIteration + >>> p.prepend(1) + >>> next(p) + 1 + >>> next(p) + Traceback (most recent call last): + ... + StopIteration + + """ + self._cache.extendleft(reversed(items)) + + def __next__(self): + if self._cache: + return self._cache.popleft() + + return next(self._it) + + def _get_slice(self, index): + # Normalize the slice's arguments + step = 1 if (index.step is None) else index.step + if step > 0: + start = 0 if (index.start is None) else index.start + stop = maxsize if (index.stop is None) else index.stop + elif step < 0: + start = -1 if (index.start is None) else index.start + stop = (-maxsize - 1) if (index.stop is None) else index.stop + else: + raise ValueError('slice step cannot be zero') + + # If either the start or stop index is negative, we'll need to cache + # the rest of the iterable in order to slice from the right side. + if (start < 0) or (stop < 0): + self._cache.extend(self._it) + # Otherwise we'll need to find the rightmost index and cache to that + # point. + else: + n = min(max(start, stop) + 1, maxsize) + cache_len = len(self._cache) + if n >= cache_len: + self._cache.extend(islice(self._it, n - cache_len)) + + return list(self._cache)[index] + + def __getitem__(self, index): + if isinstance(index, slice): + return self._get_slice(index) + + cache_len = len(self._cache) + if index < 0: + self._cache.extend(self._it) + elif index >= cache_len: + self._cache.extend(islice(self._it, index + 1 - cache_len)) + + return self._cache[index] + + +def consumer(func): + """Decorator that automatically advances a PEP-342-style "reverse iterator" + to its first yield point so you don't have to call ``next()`` on it + manually. + + >>> @consumer + ... def tally(): + ... i = 0 + ... while True: + ... print('Thing number %s is %s.' % (i, (yield))) + ... i += 1 + ... + >>> t = tally() + >>> t.send('red') + Thing number 0 is red. + >>> t.send('fish') + Thing number 1 is fish. + + Without the decorator, you would have to call ``next(t)`` before + ``t.send()`` could be used. + + """ + + @wraps(func) + def wrapper(*args, **kwargs): + gen = func(*args, **kwargs) + next(gen) + return gen + + return wrapper + + +def ilen(iterable): + """Return the number of items in *iterable*. + + >>> ilen(x for x in range(1000000) if x % 3 == 0) + 333334 + + This consumes the iterable, so handle with care. + + """ + # This approach was selected because benchmarks showed it's likely the + # fastest of the known implementations at the time of writing. + # See GitHub tracker: #236, #230. + counter = count() + deque(zip(iterable, counter), maxlen=0) + return next(counter) + + +def iterate(func, start): + """Return ``start``, ``func(start)``, ``func(func(start))``, ... + + >>> from itertools import islice + >>> list(islice(iterate(lambda x: 2*x, 1), 10)) + [1, 2, 4, 8, 16, 32, 64, 128, 256, 512] + + """ + while True: + yield start + start = func(start) + + +def with_iter(context_manager): + """Wrap an iterable in a ``with`` statement, so it closes once exhausted. + + For example, this will close the file when the iterator is exhausted:: + + upper_lines = (line.upper() for line in with_iter(open('foo'))) + + Any context manager which returns an iterable is a candidate for + ``with_iter``. + + """ + with context_manager as iterable: + yield from iterable + + +def one(iterable, too_short=None, too_long=None): + """Return the first item from *iterable*, which is expected to contain only + that item. Raise an exception if *iterable* is empty or has more than one + item. + + :func:`one` is useful for ensuring that an iterable contains only one item. + For example, it can be used to retrieve the result of a database query + that is expected to return a single row. + + If *iterable* is empty, ``ValueError`` will be raised. You may specify a + different exception with the *too_short* keyword: + + >>> it = [] + >>> one(it) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + ValueError: too many items in iterable (expected 1)' + >>> too_short = IndexError('too few items') + >>> one(it, too_short=too_short) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + IndexError: too few items + + Similarly, if *iterable* contains more than one item, ``ValueError`` will + be raised. You may specify a different exception with the *too_long* + keyword: + + >>> it = ['too', 'many'] + >>> one(it) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + ValueError: Expected exactly one item in iterable, but got 'too', + 'many', and perhaps more. + >>> too_long = RuntimeError + >>> one(it, too_long=too_long) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + RuntimeError + + Note that :func:`one` attempts to advance *iterable* twice to ensure there + is only one item. See :func:`spy` or :func:`peekable` to check iterable + contents less destructively. + + """ + it = iter(iterable) + + try: + first_value = next(it) + except StopIteration as e: + raise ( + too_short or ValueError('too few items in iterable (expected 1)') + ) from e + + try: + second_value = next(it) + except StopIteration: + pass + else: + msg = ( + 'Expected exactly one item in iterable, but got {!r}, {!r}, ' + 'and perhaps more.'.format(first_value, second_value) + ) + raise too_long or ValueError(msg) + + return first_value + + +def raise_(exception, *args): + raise exception(*args) + + +def strictly_n(iterable, n, too_short=None, too_long=None): + """Validate that *iterable* has exactly *n* items and return them if + it does. If it has fewer than *n* items, call function *too_short* + with those items. If it has more than *n* items, call function + *too_long* with the first ``n + 1`` items. + + >>> iterable = ['a', 'b', 'c', 'd'] + >>> n = 4 + >>> list(strictly_n(iterable, n)) + ['a', 'b', 'c', 'd'] + + By default, *too_short* and *too_long* are functions that raise + ``ValueError``. + + >>> list(strictly_n('ab', 3)) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + ValueError: too few items in iterable (got 2) + + >>> list(strictly_n('abc', 2)) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + ValueError: too many items in iterable (got at least 3) + + You can instead supply functions that do something else. + *too_short* will be called with the number of items in *iterable*. + *too_long* will be called with `n + 1`. + + >>> def too_short(item_count): + ... raise RuntimeError + >>> it = strictly_n('abcd', 6, too_short=too_short) + >>> list(it) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + RuntimeError + + >>> def too_long(item_count): + ... print('The boss is going to hear about this') + >>> it = strictly_n('abcdef', 4, too_long=too_long) + >>> list(it) + The boss is going to hear about this + ['a', 'b', 'c', 'd'] + + """ + if too_short is None: + too_short = lambda item_count: raise_( + ValueError, + 'Too few items in iterable (got {})'.format(item_count), + ) + + if too_long is None: + too_long = lambda item_count: raise_( + ValueError, + 'Too many items in iterable (got at least {})'.format(item_count), + ) + + it = iter(iterable) + for i in range(n): + try: + item = next(it) + except StopIteration: + too_short(i) + return + else: + yield item + + try: + next(it) + except StopIteration: + pass + else: + too_long(n + 1) + + +def distinct_permutations(iterable, r=None): + """Yield successive distinct permutations of the elements in *iterable*. + + >>> sorted(distinct_permutations([1, 0, 1])) + [(0, 1, 1), (1, 0, 1), (1, 1, 0)] + + Equivalent to ``set(permutations(iterable))``, except duplicates are not + generated and thrown away. For larger input sequences this is much more + efficient. + + Duplicate permutations arise when there are duplicated elements in the + input iterable. The number of items returned is + `n! / (x_1! * x_2! * ... * x_n!)`, where `n` is the total number of + items input, and each `x_i` is the count of a distinct item in the input + sequence. + + If *r* is given, only the *r*-length permutations are yielded. + + >>> sorted(distinct_permutations([1, 0, 1], r=2)) + [(0, 1), (1, 0), (1, 1)] + >>> sorted(distinct_permutations(range(3), r=2)) + [(0, 1), (0, 2), (1, 0), (1, 2), (2, 0), (2, 1)] + + """ + + # Algorithm: https://w.wiki/Qai + def _full(A): + while True: + # Yield the permutation we have + yield tuple(A) + + # Find the largest index i such that A[i] < A[i + 1] + for i in range(size - 2, -1, -1): + if A[i] < A[i + 1]: + break + # If no such index exists, this permutation is the last one + else: + return + + # Find the largest index j greater than j such that A[i] < A[j] + for j in range(size - 1, i, -1): + if A[i] < A[j]: + break + + # Swap the value of A[i] with that of A[j], then reverse the + # sequence from A[i + 1] to form the new permutation + A[i], A[j] = A[j], A[i] + A[i + 1 :] = A[: i - size : -1] # A[i + 1:][::-1] + + # Algorithm: modified from the above + def _partial(A, r): + # Split A into the first r items and the last r items + head, tail = A[:r], A[r:] + right_head_indexes = range(r - 1, -1, -1) + left_tail_indexes = range(len(tail)) + + while True: + # Yield the permutation we have + yield tuple(head) + + # Starting from the right, find the first index of the head with + # value smaller than the maximum value of the tail - call it i. + pivot = tail[-1] + for i in right_head_indexes: + if head[i] < pivot: + break + pivot = head[i] + else: + return + + # Starting from the left, find the first value of the tail + # with a value greater than head[i] and swap. + for j in left_tail_indexes: + if tail[j] > head[i]: + head[i], tail[j] = tail[j], head[i] + break + # If we didn't find one, start from the right and find the first + # index of the head with a value greater than head[i] and swap. + else: + for j in right_head_indexes: + if head[j] > head[i]: + head[i], head[j] = head[j], head[i] + break + + # Reverse head[i + 1:] and swap it with tail[:r - (i + 1)] + tail += head[: i - r : -1] # head[i + 1:][::-1] + i += 1 + head[i:], tail[:] = tail[: r - i], tail[r - i :] + + items = sorted(iterable) + + size = len(items) + if r is None: + r = size + + if 0 < r <= size: + return _full(items) if (r == size) else _partial(items, r) + + return iter(() if r else ((),)) + + +def intersperse(e, iterable, n=1): + """Intersperse filler element *e* among the items in *iterable*, leaving + *n* items between each filler element. + + >>> list(intersperse('!', [1, 2, 3, 4, 5])) + [1, '!', 2, '!', 3, '!', 4, '!', 5] + + >>> list(intersperse(None, [1, 2, 3, 4, 5], n=2)) + [1, 2, None, 3, 4, None, 5] + + """ + if n == 0: + raise ValueError('n must be > 0') + elif n == 1: + # interleave(repeat(e), iterable) -> e, x_0, e, x_1, e, x_2... + # islice(..., 1, None) -> x_0, e, x_1, e, x_2... + return islice(interleave(repeat(e), iterable), 1, None) + else: + # interleave(filler, chunks) -> [e], [x_0, x_1], [e], [x_2, x_3]... + # islice(..., 1, None) -> [x_0, x_1], [e], [x_2, x_3]... + # flatten(...) -> x_0, x_1, e, x_2, x_3... + filler = repeat([e]) + chunks = chunked(iterable, n) + return flatten(islice(interleave(filler, chunks), 1, None)) + + +def unique_to_each(*iterables): + """Return the elements from each of the input iterables that aren't in the + other input iterables. + + For example, suppose you have a set of packages, each with a set of + dependencies:: + + {'pkg_1': {'A', 'B'}, 'pkg_2': {'B', 'C'}, 'pkg_3': {'B', 'D'}} + + If you remove one package, which dependencies can also be removed? + + If ``pkg_1`` is removed, then ``A`` is no longer necessary - it is not + associated with ``pkg_2`` or ``pkg_3``. Similarly, ``C`` is only needed for + ``pkg_2``, and ``D`` is only needed for ``pkg_3``:: + + >>> unique_to_each({'A', 'B'}, {'B', 'C'}, {'B', 'D'}) + [['A'], ['C'], ['D']] + + If there are duplicates in one input iterable that aren't in the others + they will be duplicated in the output. Input order is preserved:: + + >>> unique_to_each("mississippi", "missouri") + [['p', 'p'], ['o', 'u', 'r']] + + It is assumed that the elements of each iterable are hashable. + + """ + pool = [list(it) for it in iterables] + counts = Counter(chain.from_iterable(map(set, pool))) + uniques = {element for element in counts if counts[element] == 1} + return [list(filter(uniques.__contains__, it)) for it in pool] + + +def windowed(seq, n, fillvalue=None, step=1): + """Return a sliding window of width *n* over the given iterable. + + >>> all_windows = windowed([1, 2, 3, 4, 5], 3) + >>> list(all_windows) + [(1, 2, 3), (2, 3, 4), (3, 4, 5)] + + When the window is larger than the iterable, *fillvalue* is used in place + of missing values: + + >>> list(windowed([1, 2, 3], 4)) + [(1, 2, 3, None)] + + Each window will advance in increments of *step*: + + >>> list(windowed([1, 2, 3, 4, 5, 6], 3, fillvalue='!', step=2)) + [(1, 2, 3), (3, 4, 5), (5, 6, '!')] + + To slide into the iterable's items, use :func:`chain` to add filler items + to the left: + + >>> iterable = [1, 2, 3, 4] + >>> n = 3 + >>> padding = [None] * (n - 1) + >>> list(windowed(chain(padding, iterable), 3)) + [(None, None, 1), (None, 1, 2), (1, 2, 3), (2, 3, 4)] + """ + if n < 0: + raise ValueError('n must be >= 0') + if n == 0: + yield tuple() + return + if step < 1: + raise ValueError('step must be >= 1') + + window = deque(maxlen=n) + i = n + for _ in map(window.append, seq): + i -= 1 + if not i: + i = step + yield tuple(window) + + size = len(window) + if size == 0: + return + elif size < n: + yield tuple(chain(window, repeat(fillvalue, n - size))) + elif 0 < i < min(step, n): + window += (fillvalue,) * i + yield tuple(window) + + +def substrings(iterable): + """Yield all of the substrings of *iterable*. + + >>> [''.join(s) for s in substrings('more')] + ['m', 'o', 'r', 'e', 'mo', 'or', 're', 'mor', 'ore', 'more'] + + Note that non-string iterables can also be subdivided. + + >>> list(substrings([0, 1, 2])) + [(0,), (1,), (2,), (0, 1), (1, 2), (0, 1, 2)] + + """ + # The length-1 substrings + seq = [] + for item in iter(iterable): + seq.append(item) + yield (item,) + seq = tuple(seq) + item_count = len(seq) + + # And the rest + for n in range(2, item_count + 1): + for i in range(item_count - n + 1): + yield seq[i : i + n] + + +def substrings_indexes(seq, reverse=False): + """Yield all substrings and their positions in *seq* + + The items yielded will be a tuple of the form ``(substr, i, j)``, where + ``substr == seq[i:j]``. + + This function only works for iterables that support slicing, such as + ``str`` objects. + + >>> for item in substrings_indexes('more'): + ... print(item) + ('m', 0, 1) + ('o', 1, 2) + ('r', 2, 3) + ('e', 3, 4) + ('mo', 0, 2) + ('or', 1, 3) + ('re', 2, 4) + ('mor', 0, 3) + ('ore', 1, 4) + ('more', 0, 4) + + Set *reverse* to ``True`` to yield the same items in the opposite order. + + + """ + r = range(1, len(seq) + 1) + if reverse: + r = reversed(r) + return ( + (seq[i : i + L], i, i + L) for L in r for i in range(len(seq) - L + 1) + ) + + +class bucket: + """Wrap *iterable* and return an object that buckets it iterable into + child iterables based on a *key* function. + + >>> iterable = ['a1', 'b1', 'c1', 'a2', 'b2', 'c2', 'b3'] + >>> s = bucket(iterable, key=lambda x: x[0]) # Bucket by 1st character + >>> sorted(list(s)) # Get the keys + ['a', 'b', 'c'] + >>> a_iterable = s['a'] + >>> next(a_iterable) + 'a1' + >>> next(a_iterable) + 'a2' + >>> list(s['b']) + ['b1', 'b2', 'b3'] + + The original iterable will be advanced and its items will be cached until + they are used by the child iterables. This may require significant storage. + + By default, attempting to select a bucket to which no items belong will + exhaust the iterable and cache all values. + If you specify a *validator* function, selected buckets will instead be + checked against it. + + >>> from itertools import count + >>> it = count(1, 2) # Infinite sequence of odd numbers + >>> key = lambda x: x % 10 # Bucket by last digit + >>> validator = lambda x: x in {1, 3, 5, 7, 9} # Odd digits only + >>> s = bucket(it, key=key, validator=validator) + >>> 2 in s + False + >>> list(s[2]) + [] + + """ + + def __init__(self, iterable, key, validator=None): + self._it = iter(iterable) + self._key = key + self._cache = defaultdict(deque) + self._validator = validator or (lambda x: True) + + def __contains__(self, value): + if not self._validator(value): + return False + + try: + item = next(self[value]) + except StopIteration: + return False + else: + self._cache[value].appendleft(item) + + return True + + def _get_values(self, value): + """ + Helper to yield items from the parent iterator that match *value*. + Items that don't match are stored in the local cache as they + are encountered. + """ + while True: + # If we've cached some items that match the target value, emit + # the first one and evict it from the cache. + if self._cache[value]: + yield self._cache[value].popleft() + # Otherwise we need to advance the parent iterator to search for + # a matching item, caching the rest. + else: + while True: + try: + item = next(self._it) + except StopIteration: + return + item_value = self._key(item) + if item_value == value: + yield item + break + elif self._validator(item_value): + self._cache[item_value].append(item) + + def __iter__(self): + for item in self._it: + item_value = self._key(item) + if self._validator(item_value): + self._cache[item_value].append(item) + + yield from self._cache.keys() + + def __getitem__(self, value): + if not self._validator(value): + return iter(()) + + return self._get_values(value) + + +def spy(iterable, n=1): + """Return a 2-tuple with a list containing the first *n* elements of + *iterable*, and an iterator with the same items as *iterable*. + This allows you to "look ahead" at the items in the iterable without + advancing it. + + There is one item in the list by default: + + >>> iterable = 'abcdefg' + >>> head, iterable = spy(iterable) + >>> head + ['a'] + >>> list(iterable) + ['a', 'b', 'c', 'd', 'e', 'f', 'g'] + + You may use unpacking to retrieve items instead of lists: + + >>> (head,), iterable = spy('abcdefg') + >>> head + 'a' + >>> (first, second), iterable = spy('abcdefg', 2) + >>> first + 'a' + >>> second + 'b' + + The number of items requested can be larger than the number of items in + the iterable: + + >>> iterable = [1, 2, 3, 4, 5] + >>> head, iterable = spy(iterable, 10) + >>> head + [1, 2, 3, 4, 5] + >>> list(iterable) + [1, 2, 3, 4, 5] + + """ + it = iter(iterable) + head = take(n, it) + + return head.copy(), chain(head, it) + + +def interleave(*iterables): + """Return a new iterable yielding from each iterable in turn, + until the shortest is exhausted. + + >>> list(interleave([1, 2, 3], [4, 5], [6, 7, 8])) + [1, 4, 6, 2, 5, 7] + + For a version that doesn't terminate after the shortest iterable is + exhausted, see :func:`interleave_longest`. + + """ + return chain.from_iterable(zip(*iterables)) + + +def interleave_longest(*iterables): + """Return a new iterable yielding from each iterable in turn, + skipping any that are exhausted. + + >>> list(interleave_longest([1, 2, 3], [4, 5], [6, 7, 8])) + [1, 4, 6, 2, 5, 7, 3, 8] + + This function produces the same output as :func:`roundrobin`, but may + perform better for some inputs (in particular when the number of iterables + is large). + + """ + i = chain.from_iterable(zip_longest(*iterables, fillvalue=_marker)) + return (x for x in i if x is not _marker) + + +def interleave_evenly(iterables, lengths=None): + """ + Interleave multiple iterables so that their elements are evenly distributed + throughout the output sequence. + + >>> iterables = [1, 2, 3, 4, 5], ['a', 'b'] + >>> list(interleave_evenly(iterables)) + [1, 2, 'a', 3, 4, 'b', 5] + + >>> iterables = [[1, 2, 3], [4, 5], [6, 7, 8]] + >>> list(interleave_evenly(iterables)) + [1, 6, 4, 2, 7, 3, 8, 5] + + This function requires iterables of known length. Iterables without + ``__len__()`` can be used by manually specifying lengths with *lengths*: + + >>> from itertools import combinations, repeat + >>> iterables = [combinations(range(4), 2), ['a', 'b', 'c']] + >>> lengths = [4 * (4 - 1) // 2, 3] + >>> list(interleave_evenly(iterables, lengths=lengths)) + [(0, 1), (0, 2), 'a', (0, 3), (1, 2), 'b', (1, 3), (2, 3), 'c'] + + Based on Bresenham's algorithm. + """ + if lengths is None: + try: + lengths = [len(it) for it in iterables] + except TypeError: + raise ValueError( + 'Iterable lengths could not be determined automatically. ' + 'Specify them with the lengths keyword.' + ) + elif len(iterables) != len(lengths): + raise ValueError('Mismatching number of iterables and lengths.') + + dims = len(lengths) + + # sort iterables by length, descending + lengths_permute = sorted( + range(dims), key=lambda i: lengths[i], reverse=True + ) + lengths_desc = [lengths[i] for i in lengths_permute] + iters_desc = [iter(iterables[i]) for i in lengths_permute] + + # the longest iterable is the primary one (Bresenham: the longest + # distance along an axis) + delta_primary, deltas_secondary = lengths_desc[0], lengths_desc[1:] + iter_primary, iters_secondary = iters_desc[0], iters_desc[1:] + errors = [delta_primary // dims] * len(deltas_secondary) + + to_yield = sum(lengths) + while to_yield: + yield next(iter_primary) + to_yield -= 1 + # update errors for each secondary iterable + errors = [e - delta for e, delta in zip(errors, deltas_secondary)] + + # those iterables for which the error is negative are yielded + # ("diagonal step" in Bresenham) + for i, e in enumerate(errors): + if e < 0: + yield next(iters_secondary[i]) + to_yield -= 1 + errors[i] += delta_primary + + +def collapse(iterable, base_type=None, levels=None): + """Flatten an iterable with multiple levels of nesting (e.g., a list of + lists of tuples) into non-iterable types. + + >>> iterable = [(1, 2), ([3, 4], [[5], [6]])] + >>> list(collapse(iterable)) + [1, 2, 3, 4, 5, 6] + + Binary and text strings are not considered iterable and + will not be collapsed. + + To avoid collapsing other types, specify *base_type*: + + >>> iterable = ['ab', ('cd', 'ef'), ['gh', 'ij']] + >>> list(collapse(iterable, base_type=tuple)) + ['ab', ('cd', 'ef'), 'gh', 'ij'] + + Specify *levels* to stop flattening after a certain level: + + >>> iterable = [('a', ['b']), ('c', ['d'])] + >>> list(collapse(iterable)) # Fully flattened + ['a', 'b', 'c', 'd'] + >>> list(collapse(iterable, levels=1)) # Only one level flattened + ['a', ['b'], 'c', ['d']] + + """ + + def walk(node, level): + if ( + ((levels is not None) and (level > levels)) + or isinstance(node, (str, bytes)) + or ((base_type is not None) and isinstance(node, base_type)) + ): + yield node + return + + try: + tree = iter(node) + except TypeError: + yield node + return + else: + for child in tree: + yield from walk(child, level + 1) + + yield from walk(iterable, 0) + + +def side_effect(func, iterable, chunk_size=None, before=None, after=None): + """Invoke *func* on each item in *iterable* (or on each *chunk_size* group + of items) before yielding the item. + + `func` must be a function that takes a single argument. Its return value + will be discarded. + + *before* and *after* are optional functions that take no arguments. They + will be executed before iteration starts and after it ends, respectively. + + `side_effect` can be used for logging, updating progress bars, or anything + that is not functionally "pure." + + Emitting a status message: + + >>> from more_itertools import consume + >>> func = lambda item: print('Received {}'.format(item)) + >>> consume(side_effect(func, range(2))) + Received 0 + Received 1 + + Operating on chunks of items: + + >>> pair_sums = [] + >>> func = lambda chunk: pair_sums.append(sum(chunk)) + >>> list(side_effect(func, [0, 1, 2, 3, 4, 5], 2)) + [0, 1, 2, 3, 4, 5] + >>> list(pair_sums) + [1, 5, 9] + + Writing to a file-like object: + + >>> from io import StringIO + >>> from more_itertools import consume + >>> f = StringIO() + >>> func = lambda x: print(x, file=f) + >>> before = lambda: print(u'HEADER', file=f) + >>> after = f.close + >>> it = [u'a', u'b', u'c'] + >>> consume(side_effect(func, it, before=before, after=after)) + >>> f.closed + True + + """ + try: + if before is not None: + before() + + if chunk_size is None: + for item in iterable: + func(item) + yield item + else: + for chunk in chunked(iterable, chunk_size): + func(chunk) + yield from chunk + finally: + if after is not None: + after() + + +def sliced(seq, n, strict=False): + """Yield slices of length *n* from the sequence *seq*. + + >>> list(sliced((1, 2, 3, 4, 5, 6), 3)) + [(1, 2, 3), (4, 5, 6)] + + By the default, the last yielded slice will have fewer than *n* elements + if the length of *seq* is not divisible by *n*: + + >>> list(sliced((1, 2, 3, 4, 5, 6, 7, 8), 3)) + [(1, 2, 3), (4, 5, 6), (7, 8)] + + If the length of *seq* is not divisible by *n* and *strict* is + ``True``, then ``ValueError`` will be raised before the last + slice is yielded. + + This function will only work for iterables that support slicing. + For non-sliceable iterables, see :func:`chunked`. + + """ + iterator = takewhile(len, (seq[i : i + n] for i in count(0, n))) + if strict: + + def ret(): + for _slice in iterator: + if len(_slice) != n: + raise ValueError("seq is not divisible by n.") + yield _slice + + return iter(ret()) + else: + return iterator + + +def split_at(iterable, pred, maxsplit=-1, keep_separator=False): + """Yield lists of items from *iterable*, where each list is delimited by + an item where callable *pred* returns ``True``. + + >>> list(split_at('abcdcba', lambda x: x == 'b')) + [['a'], ['c', 'd', 'c'], ['a']] + + >>> list(split_at(range(10), lambda n: n % 2 == 1)) + [[0], [2], [4], [6], [8], []] + + At most *maxsplit* splits are done. If *maxsplit* is not specified or -1, + then there is no limit on the number of splits: + + >>> list(split_at(range(10), lambda n: n % 2 == 1, maxsplit=2)) + [[0], [2], [4, 5, 6, 7, 8, 9]] + + By default, the delimiting items are not included in the output. + To include them, set *keep_separator* to ``True``. + + >>> list(split_at('abcdcba', lambda x: x == 'b', keep_separator=True)) + [['a'], ['b'], ['c', 'd', 'c'], ['b'], ['a']] + + """ + if maxsplit == 0: + yield list(iterable) + return + + buf = [] + it = iter(iterable) + for item in it: + if pred(item): + yield buf + if keep_separator: + yield [item] + if maxsplit == 1: + yield list(it) + return + buf = [] + maxsplit -= 1 + else: + buf.append(item) + yield buf + + +def split_before(iterable, pred, maxsplit=-1): + """Yield lists of items from *iterable*, where each list ends just before + an item for which callable *pred* returns ``True``: + + >>> list(split_before('OneTwo', lambda s: s.isupper())) + [['O', 'n', 'e'], ['T', 'w', 'o']] + + >>> list(split_before(range(10), lambda n: n % 3 == 0)) + [[0, 1, 2], [3, 4, 5], [6, 7, 8], [9]] + + At most *maxsplit* splits are done. If *maxsplit* is not specified or -1, + then there is no limit on the number of splits: + + >>> list(split_before(range(10), lambda n: n % 3 == 0, maxsplit=2)) + [[0, 1, 2], [3, 4, 5], [6, 7, 8, 9]] + """ + if maxsplit == 0: + yield list(iterable) + return + + buf = [] + it = iter(iterable) + for item in it: + if pred(item) and buf: + yield buf + if maxsplit == 1: + yield [item] + list(it) + return + buf = [] + maxsplit -= 1 + buf.append(item) + if buf: + yield buf + + +def split_after(iterable, pred, maxsplit=-1): + """Yield lists of items from *iterable*, where each list ends with an + item where callable *pred* returns ``True``: + + >>> list(split_after('one1two2', lambda s: s.isdigit())) + [['o', 'n', 'e', '1'], ['t', 'w', 'o', '2']] + + >>> list(split_after(range(10), lambda n: n % 3 == 0)) + [[0], [1, 2, 3], [4, 5, 6], [7, 8, 9]] + + At most *maxsplit* splits are done. If *maxsplit* is not specified or -1, + then there is no limit on the number of splits: + + >>> list(split_after(range(10), lambda n: n % 3 == 0, maxsplit=2)) + [[0], [1, 2, 3], [4, 5, 6, 7, 8, 9]] + + """ + if maxsplit == 0: + yield list(iterable) + return + + buf = [] + it = iter(iterable) + for item in it: + buf.append(item) + if pred(item) and buf: + yield buf + if maxsplit == 1: + buf = list(it) + if buf: + yield buf + return + buf = [] + maxsplit -= 1 + if buf: + yield buf + + +def split_when(iterable, pred, maxsplit=-1): + """Split *iterable* into pieces based on the output of *pred*. + *pred* should be a function that takes successive pairs of items and + returns ``True`` if the iterable should be split in between them. + + For example, to find runs of increasing numbers, split the iterable when + element ``i`` is larger than element ``i + 1``: + + >>> list(split_when([1, 2, 3, 3, 2, 5, 2, 4, 2], lambda x, y: x > y)) + [[1, 2, 3, 3], [2, 5], [2, 4], [2]] + + At most *maxsplit* splits are done. If *maxsplit* is not specified or -1, + then there is no limit on the number of splits: + + >>> list(split_when([1, 2, 3, 3, 2, 5, 2, 4, 2], + ... lambda x, y: x > y, maxsplit=2)) + [[1, 2, 3, 3], [2, 5], [2, 4, 2]] + + """ + if maxsplit == 0: + yield list(iterable) + return + + it = iter(iterable) + try: + cur_item = next(it) + except StopIteration: + return + + buf = [cur_item] + for next_item in it: + if pred(cur_item, next_item): + yield buf + if maxsplit == 1: + yield [next_item] + list(it) + return + buf = [] + maxsplit -= 1 + + buf.append(next_item) + cur_item = next_item + + yield buf + + +def split_into(iterable, sizes): + """Yield a list of sequential items from *iterable* of length 'n' for each + integer 'n' in *sizes*. + + >>> list(split_into([1,2,3,4,5,6], [1,2,3])) + [[1], [2, 3], [4, 5, 6]] + + If the sum of *sizes* is smaller than the length of *iterable*, then the + remaining items of *iterable* will not be returned. + + >>> list(split_into([1,2,3,4,5,6], [2,3])) + [[1, 2], [3, 4, 5]] + + If the sum of *sizes* is larger than the length of *iterable*, fewer items + will be returned in the iteration that overruns *iterable* and further + lists will be empty: + + >>> list(split_into([1,2,3,4], [1,2,3,4])) + [[1], [2, 3], [4], []] + + When a ``None`` object is encountered in *sizes*, the returned list will + contain items up to the end of *iterable* the same way that itertools.slice + does: + + >>> list(split_into([1,2,3,4,5,6,7,8,9,0], [2,3,None])) + [[1, 2], [3, 4, 5], [6, 7, 8, 9, 0]] + + :func:`split_into` can be useful for grouping a series of items where the + sizes of the groups are not uniform. An example would be where in a row + from a table, multiple columns represent elements of the same feature + (e.g. a point represented by x,y,z) but, the format is not the same for + all columns. + """ + # convert the iterable argument into an iterator so its contents can + # be consumed by islice in case it is a generator + it = iter(iterable) + + for size in sizes: + if size is None: + yield list(it) + return + else: + yield list(islice(it, size)) + + +def padded(iterable, fillvalue=None, n=None, next_multiple=False): + """Yield the elements from *iterable*, followed by *fillvalue*, such that + at least *n* items are emitted. + + >>> list(padded([1, 2, 3], '?', 5)) + [1, 2, 3, '?', '?'] + + If *next_multiple* is ``True``, *fillvalue* will be emitted until the + number of items emitted is a multiple of *n*:: + + >>> list(padded([1, 2, 3, 4], n=3, next_multiple=True)) + [1, 2, 3, 4, None, None] + + If *n* is ``None``, *fillvalue* will be emitted indefinitely. + + """ + it = iter(iterable) + if n is None: + yield from chain(it, repeat(fillvalue)) + elif n < 1: + raise ValueError('n must be at least 1') + else: + item_count = 0 + for item in it: + yield item + item_count += 1 + + remaining = (n - item_count) % n if next_multiple else n - item_count + for _ in range(remaining): + yield fillvalue + + +def repeat_each(iterable, n=2): + """Repeat each element in *iterable* *n* times. + + >>> list(repeat_each('ABC', 3)) + ['A', 'A', 'A', 'B', 'B', 'B', 'C', 'C', 'C'] + """ + return chain.from_iterable(map(repeat, iterable, repeat(n))) + + +def repeat_last(iterable, default=None): + """After the *iterable* is exhausted, keep yielding its last element. + + >>> list(islice(repeat_last(range(3)), 5)) + [0, 1, 2, 2, 2] + + If the iterable is empty, yield *default* forever:: + + >>> list(islice(repeat_last(range(0), 42), 5)) + [42, 42, 42, 42, 42] + + """ + item = _marker + for item in iterable: + yield item + final = default if item is _marker else item + yield from repeat(final) + + +def distribute(n, iterable): + """Distribute the items from *iterable* among *n* smaller iterables. + + >>> group_1, group_2 = distribute(2, [1, 2, 3, 4, 5, 6]) + >>> list(group_1) + [1, 3, 5] + >>> list(group_2) + [2, 4, 6] + + If the length of *iterable* is not evenly divisible by *n*, then the + length of the returned iterables will not be identical: + + >>> children = distribute(3, [1, 2, 3, 4, 5, 6, 7]) + >>> [list(c) for c in children] + [[1, 4, 7], [2, 5], [3, 6]] + + If the length of *iterable* is smaller than *n*, then the last returned + iterables will be empty: + + >>> children = distribute(5, [1, 2, 3]) + >>> [list(c) for c in children] + [[1], [2], [3], [], []] + + This function uses :func:`itertools.tee` and may require significant + storage. If you need the order items in the smaller iterables to match the + original iterable, see :func:`divide`. + + """ + if n < 1: + raise ValueError('n must be at least 1') + + children = tee(iterable, n) + return [islice(it, index, None, n) for index, it in enumerate(children)] + + +def stagger(iterable, offsets=(-1, 0, 1), longest=False, fillvalue=None): + """Yield tuples whose elements are offset from *iterable*. + The amount by which the `i`-th item in each tuple is offset is given by + the `i`-th item in *offsets*. + + >>> list(stagger([0, 1, 2, 3])) + [(None, 0, 1), (0, 1, 2), (1, 2, 3)] + >>> list(stagger(range(8), offsets=(0, 2, 4))) + [(0, 2, 4), (1, 3, 5), (2, 4, 6), (3, 5, 7)] + + By default, the sequence will end when the final element of a tuple is the + last item in the iterable. To continue until the first element of a tuple + is the last item in the iterable, set *longest* to ``True``:: + + >>> list(stagger([0, 1, 2, 3], longest=True)) + [(None, 0, 1), (0, 1, 2), (1, 2, 3), (2, 3, None), (3, None, None)] + + By default, ``None`` will be used to replace offsets beyond the end of the + sequence. Specify *fillvalue* to use some other value. + + """ + children = tee(iterable, len(offsets)) + + return zip_offset( + *children, offsets=offsets, longest=longest, fillvalue=fillvalue + ) + + +def zip_equal(*iterables): + """``zip`` the input *iterables* together, but raise + ``UnequalIterablesError`` if they aren't all the same length. + + >>> it_1 = range(3) + >>> it_2 = iter('abc') + >>> list(zip_equal(it_1, it_2)) + [(0, 'a'), (1, 'b'), (2, 'c')] + + >>> it_1 = range(3) + >>> it_2 = iter('abcd') + >>> list(zip_equal(it_1, it_2)) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + more_itertools.more.UnequalIterablesError: Iterables have different + lengths + + """ + if hexversion >= 0x30A00A6: + warnings.warn( + ( + 'zip_equal will be removed in a future version of ' + 'more-itertools. Use the builtin zip function with ' + 'strict=True instead.' + ), + DeprecationWarning, + ) + + return _zip_equal(*iterables) + + +def zip_offset(*iterables, offsets, longest=False, fillvalue=None): + """``zip`` the input *iterables* together, but offset the `i`-th iterable + by the `i`-th item in *offsets*. + + >>> list(zip_offset('0123', 'abcdef', offsets=(0, 1))) + [('0', 'b'), ('1', 'c'), ('2', 'd'), ('3', 'e')] + + This can be used as a lightweight alternative to SciPy or pandas to analyze + data sets in which some series have a lead or lag relationship. + + By default, the sequence will end when the shortest iterable is exhausted. + To continue until the longest iterable is exhausted, set *longest* to + ``True``. + + >>> list(zip_offset('0123', 'abcdef', offsets=(0, 1), longest=True)) + [('0', 'b'), ('1', 'c'), ('2', 'd'), ('3', 'e'), (None, 'f')] + + By default, ``None`` will be used to replace offsets beyond the end of the + sequence. Specify *fillvalue* to use some other value. + + """ + if len(iterables) != len(offsets): + raise ValueError("Number of iterables and offsets didn't match") + + staggered = [] + for it, n in zip(iterables, offsets): + if n < 0: + staggered.append(chain(repeat(fillvalue, -n), it)) + elif n > 0: + staggered.append(islice(it, n, None)) + else: + staggered.append(it) + + if longest: + return zip_longest(*staggered, fillvalue=fillvalue) + + return zip(*staggered) + + +def sort_together(iterables, key_list=(0,), key=None, reverse=False): + """Return the input iterables sorted together, with *key_list* as the + priority for sorting. All iterables are trimmed to the length of the + shortest one. + + This can be used like the sorting function in a spreadsheet. If each + iterable represents a column of data, the key list determines which + columns are used for sorting. + + By default, all iterables are sorted using the ``0``-th iterable:: + + >>> iterables = [(4, 3, 2, 1), ('a', 'b', 'c', 'd')] + >>> sort_together(iterables) + [(1, 2, 3, 4), ('d', 'c', 'b', 'a')] + + Set a different key list to sort according to another iterable. + Specifying multiple keys dictates how ties are broken:: + + >>> iterables = [(3, 1, 2), (0, 1, 0), ('c', 'b', 'a')] + >>> sort_together(iterables, key_list=(1, 2)) + [(2, 3, 1), (0, 0, 1), ('a', 'c', 'b')] + + To sort by a function of the elements of the iterable, pass a *key* + function. Its arguments are the elements of the iterables corresponding to + the key list:: + + >>> names = ('a', 'b', 'c') + >>> lengths = (1, 2, 3) + >>> widths = (5, 2, 1) + >>> def area(length, width): + ... return length * width + >>> sort_together([names, lengths, widths], key_list=(1, 2), key=area) + [('c', 'b', 'a'), (3, 2, 1), (1, 2, 5)] + + Set *reverse* to ``True`` to sort in descending order. + + >>> sort_together([(1, 2, 3), ('c', 'b', 'a')], reverse=True) + [(3, 2, 1), ('a', 'b', 'c')] + + """ + if key is None: + # if there is no key function, the key argument to sorted is an + # itemgetter + key_argument = itemgetter(*key_list) + else: + # if there is a key function, call it with the items at the offsets + # specified by the key function as arguments + key_list = list(key_list) + if len(key_list) == 1: + # if key_list contains a single item, pass the item at that offset + # as the only argument to the key function + key_offset = key_list[0] + key_argument = lambda zipped_items: key(zipped_items[key_offset]) + else: + # if key_list contains multiple items, use itemgetter to return a + # tuple of items, which we pass as *args to the key function + get_key_items = itemgetter(*key_list) + key_argument = lambda zipped_items: key( + *get_key_items(zipped_items) + ) + + return list( + zip(*sorted(zip(*iterables), key=key_argument, reverse=reverse)) + ) + + +def unzip(iterable): + """The inverse of :func:`zip`, this function disaggregates the elements + of the zipped *iterable*. + + The ``i``-th iterable contains the ``i``-th element from each element + of the zipped iterable. The first element is used to determine the + length of the remaining elements. + + >>> iterable = [('a', 1), ('b', 2), ('c', 3), ('d', 4)] + >>> letters, numbers = unzip(iterable) + >>> list(letters) + ['a', 'b', 'c', 'd'] + >>> list(numbers) + [1, 2, 3, 4] + + This is similar to using ``zip(*iterable)``, but it avoids reading + *iterable* into memory. Note, however, that this function uses + :func:`itertools.tee` and thus may require significant storage. + + """ + head, iterable = spy(iter(iterable)) + if not head: + # empty iterable, e.g. zip([], [], []) + return () + # spy returns a one-length iterable as head + head = head[0] + iterables = tee(iterable, len(head)) + + def itemgetter(i): + def getter(obj): + try: + return obj[i] + except IndexError: + # basically if we have an iterable like + # iter([(1, 2, 3), (4, 5), (6,)]) + # the second unzipped iterable would fail at the third tuple + # since it would try to access tup[1] + # same with the third unzipped iterable and the second tuple + # to support these "improperly zipped" iterables, + # we create a custom itemgetter + # which just stops the unzipped iterables + # at first length mismatch + raise StopIteration + + return getter + + return tuple(map(itemgetter(i), it) for i, it in enumerate(iterables)) + + +def divide(n, iterable): + """Divide the elements from *iterable* into *n* parts, maintaining + order. + + >>> group_1, group_2 = divide(2, [1, 2, 3, 4, 5, 6]) + >>> list(group_1) + [1, 2, 3] + >>> list(group_2) + [4, 5, 6] + + If the length of *iterable* is not evenly divisible by *n*, then the + length of the returned iterables will not be identical: + + >>> children = divide(3, [1, 2, 3, 4, 5, 6, 7]) + >>> [list(c) for c in children] + [[1, 2, 3], [4, 5], [6, 7]] + + If the length of the iterable is smaller than n, then the last returned + iterables will be empty: + + >>> children = divide(5, [1, 2, 3]) + >>> [list(c) for c in children] + [[1], [2], [3], [], []] + + This function will exhaust the iterable before returning and may require + significant storage. If order is not important, see :func:`distribute`, + which does not first pull the iterable into memory. + + """ + if n < 1: + raise ValueError('n must be at least 1') + + try: + iterable[:0] + except TypeError: + seq = tuple(iterable) + else: + seq = iterable + + q, r = divmod(len(seq), n) + + ret = [] + stop = 0 + for i in range(1, n + 1): + start = stop + stop += q + 1 if i <= r else q + ret.append(iter(seq[start:stop])) + + return ret + + +def always_iterable(obj, base_type=(str, bytes)): + """If *obj* is iterable, return an iterator over its items:: + + >>> obj = (1, 2, 3) + >>> list(always_iterable(obj)) + [1, 2, 3] + + If *obj* is not iterable, return a one-item iterable containing *obj*:: + + >>> obj = 1 + >>> list(always_iterable(obj)) + [1] + + If *obj* is ``None``, return an empty iterable: + + >>> obj = None + >>> list(always_iterable(None)) + [] + + By default, binary and text strings are not considered iterable:: + + >>> obj = 'foo' + >>> list(always_iterable(obj)) + ['foo'] + + If *base_type* is set, objects for which ``isinstance(obj, base_type)`` + returns ``True`` won't be considered iterable. + + >>> obj = {'a': 1} + >>> list(always_iterable(obj)) # Iterate over the dict's keys + ['a'] + >>> list(always_iterable(obj, base_type=dict)) # Treat dicts as a unit + [{'a': 1}] + + Set *base_type* to ``None`` to avoid any special handling and treat objects + Python considers iterable as iterable: + + >>> obj = 'foo' + >>> list(always_iterable(obj, base_type=None)) + ['f', 'o', 'o'] + """ + if obj is None: + return iter(()) + + if (base_type is not None) and isinstance(obj, base_type): + return iter((obj,)) + + try: + return iter(obj) + except TypeError: + return iter((obj,)) + + +def adjacent(predicate, iterable, distance=1): + """Return an iterable over `(bool, item)` tuples where the `item` is + drawn from *iterable* and the `bool` indicates whether + that item satisfies the *predicate* or is adjacent to an item that does. + + For example, to find whether items are adjacent to a ``3``:: + + >>> list(adjacent(lambda x: x == 3, range(6))) + [(False, 0), (False, 1), (True, 2), (True, 3), (True, 4), (False, 5)] + + Set *distance* to change what counts as adjacent. For example, to find + whether items are two places away from a ``3``: + + >>> list(adjacent(lambda x: x == 3, range(6), distance=2)) + [(False, 0), (True, 1), (True, 2), (True, 3), (True, 4), (True, 5)] + + This is useful for contextualizing the results of a search function. + For example, a code comparison tool might want to identify lines that + have changed, but also surrounding lines to give the viewer of the diff + context. + + The predicate function will only be called once for each item in the + iterable. + + See also :func:`groupby_transform`, which can be used with this function + to group ranges of items with the same `bool` value. + + """ + # Allow distance=0 mainly for testing that it reproduces results with map() + if distance < 0: + raise ValueError('distance must be at least 0') + + i1, i2 = tee(iterable) + padding = [False] * distance + selected = chain(padding, map(predicate, i1), padding) + adjacent_to_selected = map(any, windowed(selected, 2 * distance + 1)) + return zip(adjacent_to_selected, i2) + + +def groupby_transform(iterable, keyfunc=None, valuefunc=None, reducefunc=None): + """An extension of :func:`itertools.groupby` that can apply transformations + to the grouped data. + + * *keyfunc* is a function computing a key value for each item in *iterable* + * *valuefunc* is a function that transforms the individual items from + *iterable* after grouping + * *reducefunc* is a function that transforms each group of items + + >>> iterable = 'aAAbBBcCC' + >>> keyfunc = lambda k: k.upper() + >>> valuefunc = lambda v: v.lower() + >>> reducefunc = lambda g: ''.join(g) + >>> list(groupby_transform(iterable, keyfunc, valuefunc, reducefunc)) + [('A', 'aaa'), ('B', 'bbb'), ('C', 'ccc')] + + Each optional argument defaults to an identity function if not specified. + + :func:`groupby_transform` is useful when grouping elements of an iterable + using a separate iterable as the key. To do this, :func:`zip` the iterables + and pass a *keyfunc* that extracts the first element and a *valuefunc* + that extracts the second element:: + + >>> from operator import itemgetter + >>> keys = [0, 0, 1, 1, 1, 2, 2, 2, 3] + >>> values = 'abcdefghi' + >>> iterable = zip(keys, values) + >>> grouper = groupby_transform(iterable, itemgetter(0), itemgetter(1)) + >>> [(k, ''.join(g)) for k, g in grouper] + [(0, 'ab'), (1, 'cde'), (2, 'fgh'), (3, 'i')] + + Note that the order of items in the iterable is significant. + Only adjacent items are grouped together, so if you don't want any + duplicate groups, you should sort the iterable by the key function. + + """ + ret = groupby(iterable, keyfunc) + if valuefunc: + ret = ((k, map(valuefunc, g)) for k, g in ret) + if reducefunc: + ret = ((k, reducefunc(g)) for k, g in ret) + + return ret + + +class numeric_range(abc.Sequence, abc.Hashable): + """An extension of the built-in ``range()`` function whose arguments can + be any orderable numeric type. + + With only *stop* specified, *start* defaults to ``0`` and *step* + defaults to ``1``. The output items will match the type of *stop*: + + >>> list(numeric_range(3.5)) + [0.0, 1.0, 2.0, 3.0] + + With only *start* and *stop* specified, *step* defaults to ``1``. The + output items will match the type of *start*: + + >>> from decimal import Decimal + >>> start = Decimal('2.1') + >>> stop = Decimal('5.1') + >>> list(numeric_range(start, stop)) + [Decimal('2.1'), Decimal('3.1'), Decimal('4.1')] + + With *start*, *stop*, and *step* specified the output items will match + the type of ``start + step``: + + >>> from fractions import Fraction + >>> start = Fraction(1, 2) # Start at 1/2 + >>> stop = Fraction(5, 2) # End at 5/2 + >>> step = Fraction(1, 2) # Count by 1/2 + >>> list(numeric_range(start, stop, step)) + [Fraction(1, 2), Fraction(1, 1), Fraction(3, 2), Fraction(2, 1)] + + If *step* is zero, ``ValueError`` is raised. Negative steps are supported: + + >>> list(numeric_range(3, -1, -1.0)) + [3.0, 2.0, 1.0, 0.0] + + Be aware of the limitations of floating point numbers; the representation + of the yielded numbers may be surprising. + + ``datetime.datetime`` objects can be used for *start* and *stop*, if *step* + is a ``datetime.timedelta`` object: + + >>> import datetime + >>> start = datetime.datetime(2019, 1, 1) + >>> stop = datetime.datetime(2019, 1, 3) + >>> step = datetime.timedelta(days=1) + >>> items = iter(numeric_range(start, stop, step)) + >>> next(items) + datetime.datetime(2019, 1, 1, 0, 0) + >>> next(items) + datetime.datetime(2019, 1, 2, 0, 0) + + """ + + _EMPTY_HASH = hash(range(0, 0)) + + def __init__(self, *args): + argc = len(args) + if argc == 1: + (self._stop,) = args + self._start = type(self._stop)(0) + self._step = type(self._stop - self._start)(1) + elif argc == 2: + self._start, self._stop = args + self._step = type(self._stop - self._start)(1) + elif argc == 3: + self._start, self._stop, self._step = args + elif argc == 0: + raise TypeError( + 'numeric_range expected at least ' + '1 argument, got {}'.format(argc) + ) + else: + raise TypeError( + 'numeric_range expected at most ' + '3 arguments, got {}'.format(argc) + ) + + self._zero = type(self._step)(0) + if self._step == self._zero: + raise ValueError('numeric_range() arg 3 must not be zero') + self._growing = self._step > self._zero + self._init_len() + + def __bool__(self): + if self._growing: + return self._start < self._stop + else: + return self._start > self._stop + + def __contains__(self, elem): + if self._growing: + if self._start <= elem < self._stop: + return (elem - self._start) % self._step == self._zero + else: + if self._start >= elem > self._stop: + return (self._start - elem) % (-self._step) == self._zero + + return False + + def __eq__(self, other): + if isinstance(other, numeric_range): + empty_self = not bool(self) + empty_other = not bool(other) + if empty_self or empty_other: + return empty_self and empty_other # True if both empty + else: + return ( + self._start == other._start + and self._step == other._step + and self._get_by_index(-1) == other._get_by_index(-1) + ) + else: + return False + + def __getitem__(self, key): + if isinstance(key, int): + return self._get_by_index(key) + elif isinstance(key, slice): + step = self._step if key.step is None else key.step * self._step + + if key.start is None or key.start <= -self._len: + start = self._start + elif key.start >= self._len: + start = self._stop + else: # -self._len < key.start < self._len + start = self._get_by_index(key.start) + + if key.stop is None or key.stop >= self._len: + stop = self._stop + elif key.stop <= -self._len: + stop = self._start + else: # -self._len < key.stop < self._len + stop = self._get_by_index(key.stop) + + return numeric_range(start, stop, step) + else: + raise TypeError( + 'numeric range indices must be ' + 'integers or slices, not {}'.format(type(key).__name__) + ) + + def __hash__(self): + if self: + return hash((self._start, self._get_by_index(-1), self._step)) + else: + return self._EMPTY_HASH + + def __iter__(self): + values = (self._start + (n * self._step) for n in count()) + if self._growing: + return takewhile(partial(gt, self._stop), values) + else: + return takewhile(partial(lt, self._stop), values) + + def __len__(self): + return self._len + + def _init_len(self): + if self._growing: + start = self._start + stop = self._stop + step = self._step + else: + start = self._stop + stop = self._start + step = -self._step + distance = stop - start + if distance <= self._zero: + self._len = 0 + else: # distance > 0 and step > 0: regular euclidean division + q, r = divmod(distance, step) + self._len = int(q) + int(r != self._zero) + + def __reduce__(self): + return numeric_range, (self._start, self._stop, self._step) + + def __repr__(self): + if self._step == 1: + return "numeric_range({}, {})".format( + repr(self._start), repr(self._stop) + ) + else: + return "numeric_range({}, {}, {})".format( + repr(self._start), repr(self._stop), repr(self._step) + ) + + def __reversed__(self): + return iter( + numeric_range( + self._get_by_index(-1), self._start - self._step, -self._step + ) + ) + + def count(self, value): + return int(value in self) + + def index(self, value): + if self._growing: + if self._start <= value < self._stop: + q, r = divmod(value - self._start, self._step) + if r == self._zero: + return int(q) + else: + if self._start >= value > self._stop: + q, r = divmod(self._start - value, -self._step) + if r == self._zero: + return int(q) + + raise ValueError("{} is not in numeric range".format(value)) + + def _get_by_index(self, i): + if i < 0: + i += self._len + if i < 0 or i >= self._len: + raise IndexError("numeric range object index out of range") + return self._start + i * self._step + + +def count_cycle(iterable, n=None): + """Cycle through the items from *iterable* up to *n* times, yielding + the number of completed cycles along with each item. If *n* is omitted the + process repeats indefinitely. + + >>> list(count_cycle('AB', 3)) + [(0, 'A'), (0, 'B'), (1, 'A'), (1, 'B'), (2, 'A'), (2, 'B')] + + """ + iterable = tuple(iterable) + if not iterable: + return iter(()) + counter = count() if n is None else range(n) + return ((i, item) for i in counter for item in iterable) + + +def mark_ends(iterable): + """Yield 3-tuples of the form ``(is_first, is_last, item)``. + + >>> list(mark_ends('ABC')) + [(True, False, 'A'), (False, False, 'B'), (False, True, 'C')] + + Use this when looping over an iterable to take special action on its first + and/or last items: + + >>> iterable = ['Header', 100, 200, 'Footer'] + >>> total = 0 + >>> for is_first, is_last, item in mark_ends(iterable): + ... if is_first: + ... continue # Skip the header + ... if is_last: + ... continue # Skip the footer + ... total += item + >>> print(total) + 300 + """ + it = iter(iterable) + + try: + b = next(it) + except StopIteration: + return + + try: + for i in count(): + a = b + b = next(it) + yield i == 0, False, a + + except StopIteration: + yield i == 0, True, a + + +def locate(iterable, pred=bool, window_size=None): + """Yield the index of each item in *iterable* for which *pred* returns + ``True``. + + *pred* defaults to :func:`bool`, which will select truthy items: + + >>> list(locate([0, 1, 1, 0, 1, 0, 0])) + [1, 2, 4] + + Set *pred* to a custom function to, e.g., find the indexes for a particular + item. + + >>> list(locate(['a', 'b', 'c', 'b'], lambda x: x == 'b')) + [1, 3] + + If *window_size* is given, then the *pred* function will be called with + that many items. This enables searching for sub-sequences: + + >>> iterable = [0, 1, 2, 3, 0, 1, 2, 3, 0, 1, 2, 3] + >>> pred = lambda *args: args == (1, 2, 3) + >>> list(locate(iterable, pred=pred, window_size=3)) + [1, 5, 9] + + Use with :func:`seekable` to find indexes and then retrieve the associated + items: + + >>> from itertools import count + >>> from more_itertools import seekable + >>> source = (3 * n + 1 if (n % 2) else n // 2 for n in count()) + >>> it = seekable(source) + >>> pred = lambda x: x > 100 + >>> indexes = locate(it, pred=pred) + >>> i = next(indexes) + >>> it.seek(i) + >>> next(it) + 106 + + """ + if window_size is None: + return compress(count(), map(pred, iterable)) + + if window_size < 1: + raise ValueError('window size must be at least 1') + + it = windowed(iterable, window_size, fillvalue=_marker) + return compress(count(), starmap(pred, it)) + + +def longest_common_prefix(iterables): + """Yield elements of the longest common prefix amongst given *iterables*. + + >>> ''.join(longest_common_prefix(['abcd', 'abc', 'abf'])) + 'ab' + + """ + return (c[0] for c in takewhile(all_equal, zip(*iterables))) + + +def lstrip(iterable, pred): + """Yield the items from *iterable*, but strip any from the beginning + for which *pred* returns ``True``. + + For example, to remove a set of items from the start of an iterable: + + >>> iterable = (None, False, None, 1, 2, None, 3, False, None) + >>> pred = lambda x: x in {None, False, ''} + >>> list(lstrip(iterable, pred)) + [1, 2, None, 3, False, None] + + This function is analogous to to :func:`str.lstrip`, and is essentially + an wrapper for :func:`itertools.dropwhile`. + + """ + return dropwhile(pred, iterable) + + +def rstrip(iterable, pred): + """Yield the items from *iterable*, but strip any from the end + for which *pred* returns ``True``. + + For example, to remove a set of items from the end of an iterable: + + >>> iterable = (None, False, None, 1, 2, None, 3, False, None) + >>> pred = lambda x: x in {None, False, ''} + >>> list(rstrip(iterable, pred)) + [None, False, None, 1, 2, None, 3] + + This function is analogous to :func:`str.rstrip`. + + """ + cache = [] + cache_append = cache.append + cache_clear = cache.clear + for x in iterable: + if pred(x): + cache_append(x) + else: + yield from cache + cache_clear() + yield x + + +def strip(iterable, pred): + """Yield the items from *iterable*, but strip any from the + beginning and end for which *pred* returns ``True``. + + For example, to remove a set of items from both ends of an iterable: + + >>> iterable = (None, False, None, 1, 2, None, 3, False, None) + >>> pred = lambda x: x in {None, False, ''} + >>> list(strip(iterable, pred)) + [1, 2, None, 3] + + This function is analogous to :func:`str.strip`. + + """ + return rstrip(lstrip(iterable, pred), pred) + + +class islice_extended: + """An extension of :func:`itertools.islice` that supports negative values + for *stop*, *start*, and *step*. + + >>> iterable = iter('abcdefgh') + >>> list(islice_extended(iterable, -4, -1)) + ['e', 'f', 'g'] + + Slices with negative values require some caching of *iterable*, but this + function takes care to minimize the amount of memory required. + + For example, you can use a negative step with an infinite iterator: + + >>> from itertools import count + >>> list(islice_extended(count(), 110, 99, -2)) + [110, 108, 106, 104, 102, 100] + + You can also use slice notation directly: + + >>> iterable = map(str, count()) + >>> it = islice_extended(iterable)[10:20:2] + >>> list(it) + ['10', '12', '14', '16', '18'] + + """ + + def __init__(self, iterable, *args): + it = iter(iterable) + if args: + self._iterable = _islice_helper(it, slice(*args)) + else: + self._iterable = it + + def __iter__(self): + return self + + def __next__(self): + return next(self._iterable) + + def __getitem__(self, key): + if isinstance(key, slice): + return islice_extended(_islice_helper(self._iterable, key)) + + raise TypeError('islice_extended.__getitem__ argument must be a slice') + + +def _islice_helper(it, s): + start = s.start + stop = s.stop + if s.step == 0: + raise ValueError('step argument must be a non-zero integer or None.') + step = s.step or 1 + + if step > 0: + start = 0 if (start is None) else start + + if start < 0: + # Consume all but the last -start items + cache = deque(enumerate(it, 1), maxlen=-start) + len_iter = cache[-1][0] if cache else 0 + + # Adjust start to be positive + i = max(len_iter + start, 0) + + # Adjust stop to be positive + if stop is None: + j = len_iter + elif stop >= 0: + j = min(stop, len_iter) + else: + j = max(len_iter + stop, 0) + + # Slice the cache + n = j - i + if n <= 0: + return + + for index, item in islice(cache, 0, n, step): + yield item + elif (stop is not None) and (stop < 0): + # Advance to the start position + next(islice(it, start, start), None) + + # When stop is negative, we have to carry -stop items while + # iterating + cache = deque(islice(it, -stop), maxlen=-stop) + + for index, item in enumerate(it): + cached_item = cache.popleft() + if index % step == 0: + yield cached_item + cache.append(item) + else: + # When both start and stop are positive we have the normal case + yield from islice(it, start, stop, step) + else: + start = -1 if (start is None) else start + + if (stop is not None) and (stop < 0): + # Consume all but the last items + n = -stop - 1 + cache = deque(enumerate(it, 1), maxlen=n) + len_iter = cache[-1][0] if cache else 0 + + # If start and stop are both negative they are comparable and + # we can just slice. Otherwise we can adjust start to be negative + # and then slice. + if start < 0: + i, j = start, stop + else: + i, j = min(start - len_iter, -1), None + + for index, item in list(cache)[i:j:step]: + yield item + else: + # Advance to the stop position + if stop is not None: + m = stop + 1 + next(islice(it, m, m), None) + + # stop is positive, so if start is negative they are not comparable + # and we need the rest of the items. + if start < 0: + i = start + n = None + # stop is None and start is positive, so we just need items up to + # the start index. + elif stop is None: + i = None + n = start + 1 + # Both stop and start are positive, so they are comparable. + else: + i = None + n = start - stop + if n <= 0: + return + + cache = list(islice(it, n)) + + yield from cache[i::step] + + +def always_reversible(iterable): + """An extension of :func:`reversed` that supports all iterables, not + just those which implement the ``Reversible`` or ``Sequence`` protocols. + + >>> print(*always_reversible(x for x in range(3))) + 2 1 0 + + If the iterable is already reversible, this function returns the + result of :func:`reversed()`. If the iterable is not reversible, + this function will cache the remaining items in the iterable and + yield them in reverse order, which may require significant storage. + """ + try: + return reversed(iterable) + except TypeError: + return reversed(list(iterable)) + + +def consecutive_groups(iterable, ordering=lambda x: x): + """Yield groups of consecutive items using :func:`itertools.groupby`. + The *ordering* function determines whether two items are adjacent by + returning their position. + + By default, the ordering function is the identity function. This is + suitable for finding runs of numbers: + + >>> iterable = [1, 10, 11, 12, 20, 30, 31, 32, 33, 40] + >>> for group in consecutive_groups(iterable): + ... print(list(group)) + [1] + [10, 11, 12] + [20] + [30, 31, 32, 33] + [40] + + For finding runs of adjacent letters, try using the :meth:`index` method + of a string of letters: + + >>> from string import ascii_lowercase + >>> iterable = 'abcdfgilmnop' + >>> ordering = ascii_lowercase.index + >>> for group in consecutive_groups(iterable, ordering): + ... print(list(group)) + ['a', 'b', 'c', 'd'] + ['f', 'g'] + ['i'] + ['l', 'm', 'n', 'o', 'p'] + + Each group of consecutive items is an iterator that shares it source with + *iterable*. When an an output group is advanced, the previous group is + no longer available unless its elements are copied (e.g., into a ``list``). + + >>> iterable = [1, 2, 11, 12, 21, 22] + >>> saved_groups = [] + >>> for group in consecutive_groups(iterable): + ... saved_groups.append(list(group)) # Copy group elements + >>> saved_groups + [[1, 2], [11, 12], [21, 22]] + + """ + for k, g in groupby( + enumerate(iterable), key=lambda x: x[0] - ordering(x[1]) + ): + yield map(itemgetter(1), g) + + +def difference(iterable, func=sub, *, initial=None): + """This function is the inverse of :func:`itertools.accumulate`. By default + it will compute the first difference of *iterable* using + :func:`operator.sub`: + + >>> from itertools import accumulate + >>> iterable = accumulate([0, 1, 2, 3, 4]) # produces 0, 1, 3, 6, 10 + >>> list(difference(iterable)) + [0, 1, 2, 3, 4] + + *func* defaults to :func:`operator.sub`, but other functions can be + specified. They will be applied as follows:: + + A, B, C, D, ... --> A, func(B, A), func(C, B), func(D, C), ... + + For example, to do progressive division: + + >>> iterable = [1, 2, 6, 24, 120] + >>> func = lambda x, y: x // y + >>> list(difference(iterable, func)) + [1, 2, 3, 4, 5] + + If the *initial* keyword is set, the first element will be skipped when + computing successive differences. + + >>> it = [10, 11, 13, 16] # from accumulate([1, 2, 3], initial=10) + >>> list(difference(it, initial=10)) + [1, 2, 3] + + """ + a, b = tee(iterable) + try: + first = [next(b)] + except StopIteration: + return iter([]) + + if initial is not None: + first = [] + + return chain(first, map(func, b, a)) + + +class SequenceView(Sequence): + """Return a read-only view of the sequence object *target*. + + :class:`SequenceView` objects are analogous to Python's built-in + "dictionary view" types. They provide a dynamic view of a sequence's items, + meaning that when the sequence updates, so does the view. + + >>> seq = ['0', '1', '2'] + >>> view = SequenceView(seq) + >>> view + SequenceView(['0', '1', '2']) + >>> seq.append('3') + >>> view + SequenceView(['0', '1', '2', '3']) + + Sequence views support indexing, slicing, and length queries. They act + like the underlying sequence, except they don't allow assignment: + + >>> view[1] + '1' + >>> view[1:-1] + ['1', '2'] + >>> len(view) + 4 + + Sequence views are useful as an alternative to copying, as they don't + require (much) extra storage. + + """ + + def __init__(self, target): + if not isinstance(target, Sequence): + raise TypeError + self._target = target + + def __getitem__(self, index): + return self._target[index] + + def __len__(self): + return len(self._target) + + def __repr__(self): + return '{}({})'.format(self.__class__.__name__, repr(self._target)) + + +class seekable: + """Wrap an iterator to allow for seeking backward and forward. This + progressively caches the items in the source iterable so they can be + re-visited. + + Call :meth:`seek` with an index to seek to that position in the source + iterable. + + To "reset" an iterator, seek to ``0``: + + >>> from itertools import count + >>> it = seekable((str(n) for n in count())) + >>> next(it), next(it), next(it) + ('0', '1', '2') + >>> it.seek(0) + >>> next(it), next(it), next(it) + ('0', '1', '2') + >>> next(it) + '3' + + You can also seek forward: + + >>> it = seekable((str(n) for n in range(20))) + >>> it.seek(10) + >>> next(it) + '10' + >>> it.seek(20) # Seeking past the end of the source isn't a problem + >>> list(it) + [] + >>> it.seek(0) # Resetting works even after hitting the end + >>> next(it), next(it), next(it) + ('0', '1', '2') + + Call :meth:`peek` to look ahead one item without advancing the iterator: + + >>> it = seekable('1234') + >>> it.peek() + '1' + >>> list(it) + ['1', '2', '3', '4'] + >>> it.peek(default='empty') + 'empty' + + Before the iterator is at its end, calling :func:`bool` on it will return + ``True``. After it will return ``False``: + + >>> it = seekable('5678') + >>> bool(it) + True + >>> list(it) + ['5', '6', '7', '8'] + >>> bool(it) + False + + You may view the contents of the cache with the :meth:`elements` method. + That returns a :class:`SequenceView`, a view that updates automatically: + + >>> it = seekable((str(n) for n in range(10))) + >>> next(it), next(it), next(it) + ('0', '1', '2') + >>> elements = it.elements() + >>> elements + SequenceView(['0', '1', '2']) + >>> next(it) + '3' + >>> elements + SequenceView(['0', '1', '2', '3']) + + By default, the cache grows as the source iterable progresses, so beware of + wrapping very large or infinite iterables. Supply *maxlen* to limit the + size of the cache (this of course limits how far back you can seek). + + >>> from itertools import count + >>> it = seekable((str(n) for n in count()), maxlen=2) + >>> next(it), next(it), next(it), next(it) + ('0', '1', '2', '3') + >>> list(it.elements()) + ['2', '3'] + >>> it.seek(0) + >>> next(it), next(it), next(it), next(it) + ('2', '3', '4', '5') + >>> next(it) + '6' + + """ + + def __init__(self, iterable, maxlen=None): + self._source = iter(iterable) + if maxlen is None: + self._cache = [] + else: + self._cache = deque([], maxlen) + self._index = None + + def __iter__(self): + return self + + def __next__(self): + if self._index is not None: + try: + item = self._cache[self._index] + except IndexError: + self._index = None + else: + self._index += 1 + return item + + item = next(self._source) + self._cache.append(item) + return item + + def __bool__(self): + try: + self.peek() + except StopIteration: + return False + return True + + def peek(self, default=_marker): + try: + peeked = next(self) + except StopIteration: + if default is _marker: + raise + return default + if self._index is None: + self._index = len(self._cache) + self._index -= 1 + return peeked + + def elements(self): + return SequenceView(self._cache) + + def seek(self, index): + self._index = index + remainder = index - len(self._cache) + if remainder > 0: + consume(self, remainder) + + +class run_length: + """ + :func:`run_length.encode` compresses an iterable with run-length encoding. + It yields groups of repeated items with the count of how many times they + were repeated: + + >>> uncompressed = 'abbcccdddd' + >>> list(run_length.encode(uncompressed)) + [('a', 1), ('b', 2), ('c', 3), ('d', 4)] + + :func:`run_length.decode` decompresses an iterable that was previously + compressed with run-length encoding. It yields the items of the + decompressed iterable: + + >>> compressed = [('a', 1), ('b', 2), ('c', 3), ('d', 4)] + >>> list(run_length.decode(compressed)) + ['a', 'b', 'b', 'c', 'c', 'c', 'd', 'd', 'd', 'd'] + + """ + + @staticmethod + def encode(iterable): + return ((k, ilen(g)) for k, g in groupby(iterable)) + + @staticmethod + def decode(iterable): + return chain.from_iterable(repeat(k, n) for k, n in iterable) + + +def exactly_n(iterable, n, predicate=bool): + """Return ``True`` if exactly ``n`` items in the iterable are ``True`` + according to the *predicate* function. + + >>> exactly_n([True, True, False], 2) + True + >>> exactly_n([True, True, False], 1) + False + >>> exactly_n([0, 1, 2, 3, 4, 5], 3, lambda x: x < 3) + True + + The iterable will be advanced until ``n + 1`` truthy items are encountered, + so avoid calling it on infinite iterables. + + """ + return len(take(n + 1, filter(predicate, iterable))) == n + + +def circular_shifts(iterable): + """Return a list of circular shifts of *iterable*. + + >>> circular_shifts(range(4)) + [(0, 1, 2, 3), (1, 2, 3, 0), (2, 3, 0, 1), (3, 0, 1, 2)] + """ + lst = list(iterable) + return take(len(lst), windowed(cycle(lst), len(lst))) + + +def make_decorator(wrapping_func, result_index=0): + """Return a decorator version of *wrapping_func*, which is a function that + modifies an iterable. *result_index* is the position in that function's + signature where the iterable goes. + + This lets you use itertools on the "production end," i.e. at function + definition. This can augment what the function returns without changing the + function's code. + + For example, to produce a decorator version of :func:`chunked`: + + >>> from more_itertools import chunked + >>> chunker = make_decorator(chunked, result_index=0) + >>> @chunker(3) + ... def iter_range(n): + ... return iter(range(n)) + ... + >>> list(iter_range(9)) + [[0, 1, 2], [3, 4, 5], [6, 7, 8]] + + To only allow truthy items to be returned: + + >>> truth_serum = make_decorator(filter, result_index=1) + >>> @truth_serum(bool) + ... def boolean_test(): + ... return [0, 1, '', ' ', False, True] + ... + >>> list(boolean_test()) + [1, ' ', True] + + The :func:`peekable` and :func:`seekable` wrappers make for practical + decorators: + + >>> from more_itertools import peekable + >>> peekable_function = make_decorator(peekable) + >>> @peekable_function() + ... def str_range(*args): + ... return (str(x) for x in range(*args)) + ... + >>> it = str_range(1, 20, 2) + >>> next(it), next(it), next(it) + ('1', '3', '5') + >>> it.peek() + '7' + >>> next(it) + '7' + + """ + + # See https://sites.google.com/site/bbayles/index/decorator_factory for + # notes on how this works. + def decorator(*wrapping_args, **wrapping_kwargs): + def outer_wrapper(f): + def inner_wrapper(*args, **kwargs): + result = f(*args, **kwargs) + wrapping_args_ = list(wrapping_args) + wrapping_args_.insert(result_index, result) + return wrapping_func(*wrapping_args_, **wrapping_kwargs) + + return inner_wrapper + + return outer_wrapper + + return decorator + + +def map_reduce(iterable, keyfunc, valuefunc=None, reducefunc=None): + """Return a dictionary that maps the items in *iterable* to categories + defined by *keyfunc*, transforms them with *valuefunc*, and + then summarizes them by category with *reducefunc*. + + *valuefunc* defaults to the identity function if it is unspecified. + If *reducefunc* is unspecified, no summarization takes place: + + >>> keyfunc = lambda x: x.upper() + >>> result = map_reduce('abbccc', keyfunc) + >>> sorted(result.items()) + [('A', ['a']), ('B', ['b', 'b']), ('C', ['c', 'c', 'c'])] + + Specifying *valuefunc* transforms the categorized items: + + >>> keyfunc = lambda x: x.upper() + >>> valuefunc = lambda x: 1 + >>> result = map_reduce('abbccc', keyfunc, valuefunc) + >>> sorted(result.items()) + [('A', [1]), ('B', [1, 1]), ('C', [1, 1, 1])] + + Specifying *reducefunc* summarizes the categorized items: + + >>> keyfunc = lambda x: x.upper() + >>> valuefunc = lambda x: 1 + >>> reducefunc = sum + >>> result = map_reduce('abbccc', keyfunc, valuefunc, reducefunc) + >>> sorted(result.items()) + [('A', 1), ('B', 2), ('C', 3)] + + You may want to filter the input iterable before applying the map/reduce + procedure: + + >>> all_items = range(30) + >>> items = [x for x in all_items if 10 <= x <= 20] # Filter + >>> keyfunc = lambda x: x % 2 # Evens map to 0; odds to 1 + >>> categories = map_reduce(items, keyfunc=keyfunc) + >>> sorted(categories.items()) + [(0, [10, 12, 14, 16, 18, 20]), (1, [11, 13, 15, 17, 19])] + >>> summaries = map_reduce(items, keyfunc=keyfunc, reducefunc=sum) + >>> sorted(summaries.items()) + [(0, 90), (1, 75)] + + Note that all items in the iterable are gathered into a list before the + summarization step, which may require significant storage. + + The returned object is a :obj:`collections.defaultdict` with the + ``default_factory`` set to ``None``, such that it behaves like a normal + dictionary. + + """ + valuefunc = (lambda x: x) if (valuefunc is None) else valuefunc + + ret = defaultdict(list) + for item in iterable: + key = keyfunc(item) + value = valuefunc(item) + ret[key].append(value) + + if reducefunc is not None: + for key, value_list in ret.items(): + ret[key] = reducefunc(value_list) + + ret.default_factory = None + return ret + + +def rlocate(iterable, pred=bool, window_size=None): + """Yield the index of each item in *iterable* for which *pred* returns + ``True``, starting from the right and moving left. + + *pred* defaults to :func:`bool`, which will select truthy items: + + >>> list(rlocate([0, 1, 1, 0, 1, 0, 0])) # Truthy at 1, 2, and 4 + [4, 2, 1] + + Set *pred* to a custom function to, e.g., find the indexes for a particular + item: + + >>> iterable = iter('abcb') + >>> pred = lambda x: x == 'b' + >>> list(rlocate(iterable, pred)) + [3, 1] + + If *window_size* is given, then the *pred* function will be called with + that many items. This enables searching for sub-sequences: + + >>> iterable = [0, 1, 2, 3, 0, 1, 2, 3, 0, 1, 2, 3] + >>> pred = lambda *args: args == (1, 2, 3) + >>> list(rlocate(iterable, pred=pred, window_size=3)) + [9, 5, 1] + + Beware, this function won't return anything for infinite iterables. + If *iterable* is reversible, ``rlocate`` will reverse it and search from + the right. Otherwise, it will search from the left and return the results + in reverse order. + + See :func:`locate` to for other example applications. + + """ + if window_size is None: + try: + len_iter = len(iterable) + return (len_iter - i - 1 for i in locate(reversed(iterable), pred)) + except TypeError: + pass + + return reversed(list(locate(iterable, pred, window_size))) + + +def replace(iterable, pred, substitutes, count=None, window_size=1): + """Yield the items from *iterable*, replacing the items for which *pred* + returns ``True`` with the items from the iterable *substitutes*. + + >>> iterable = [1, 1, 0, 1, 1, 0, 1, 1] + >>> pred = lambda x: x == 0 + >>> substitutes = (2, 3) + >>> list(replace(iterable, pred, substitutes)) + [1, 1, 2, 3, 1, 1, 2, 3, 1, 1] + + If *count* is given, the number of replacements will be limited: + + >>> iterable = [1, 1, 0, 1, 1, 0, 1, 1, 0] + >>> pred = lambda x: x == 0 + >>> substitutes = [None] + >>> list(replace(iterable, pred, substitutes, count=2)) + [1, 1, None, 1, 1, None, 1, 1, 0] + + Use *window_size* to control the number of items passed as arguments to + *pred*. This allows for locating and replacing subsequences. + + >>> iterable = [0, 1, 2, 5, 0, 1, 2, 5] + >>> window_size = 3 + >>> pred = lambda *args: args == (0, 1, 2) # 3 items passed to pred + >>> substitutes = [3, 4] # Splice in these items + >>> list(replace(iterable, pred, substitutes, window_size=window_size)) + [3, 4, 5, 3, 4, 5] + + """ + if window_size < 1: + raise ValueError('window_size must be at least 1') + + # Save the substitutes iterable, since it's used more than once + substitutes = tuple(substitutes) + + # Add padding such that the number of windows matches the length of the + # iterable + it = chain(iterable, [_marker] * (window_size - 1)) + windows = windowed(it, window_size) + + n = 0 + for w in windows: + # If the current window matches our predicate (and we haven't hit + # our maximum number of replacements), splice in the substitutes + # and then consume the following windows that overlap with this one. + # For example, if the iterable is (0, 1, 2, 3, 4...) + # and the window size is 2, we have (0, 1), (1, 2), (2, 3)... + # If the predicate matches on (0, 1), we need to zap (0, 1) and (1, 2) + if pred(*w): + if (count is None) or (n < count): + n += 1 + yield from substitutes + consume(windows, window_size - 1) + continue + + # If there was no match (or we've reached the replacement limit), + # yield the first item from the window. + if w and (w[0] is not _marker): + yield w[0] + + +def partitions(iterable): + """Yield all possible order-preserving partitions of *iterable*. + + >>> iterable = 'abc' + >>> for part in partitions(iterable): + ... print([''.join(p) for p in part]) + ['abc'] + ['a', 'bc'] + ['ab', 'c'] + ['a', 'b', 'c'] + + This is unrelated to :func:`partition`. + + """ + sequence = list(iterable) + n = len(sequence) + for i in powerset(range(1, n)): + yield [sequence[i:j] for i, j in zip((0,) + i, i + (n,))] + + +def set_partitions(iterable, k=None): + """ + Yield the set partitions of *iterable* into *k* parts. Set partitions are + not order-preserving. + + >>> iterable = 'abc' + >>> for part in set_partitions(iterable, 2): + ... print([''.join(p) for p in part]) + ['a', 'bc'] + ['ab', 'c'] + ['b', 'ac'] + + + If *k* is not given, every set partition is generated. + + >>> iterable = 'abc' + >>> for part in set_partitions(iterable): + ... print([''.join(p) for p in part]) + ['abc'] + ['a', 'bc'] + ['ab', 'c'] + ['b', 'ac'] + ['a', 'b', 'c'] + + """ + L = list(iterable) + n = len(L) + if k is not None: + if k < 1: + raise ValueError( + "Can't partition in a negative or zero number of groups" + ) + elif k > n: + return + + def set_partitions_helper(L, k): + n = len(L) + if k == 1: + yield [L] + elif n == k: + yield [[s] for s in L] + else: + e, *M = L + for p in set_partitions_helper(M, k - 1): + yield [[e], *p] + for p in set_partitions_helper(M, k): + for i in range(len(p)): + yield p[:i] + [[e] + p[i]] + p[i + 1 :] + + if k is None: + for k in range(1, n + 1): + yield from set_partitions_helper(L, k) + else: + yield from set_partitions_helper(L, k) + + +class time_limited: + """ + Yield items from *iterable* until *limit_seconds* have passed. + If the time limit expires before all items have been yielded, the + ``timed_out`` parameter will be set to ``True``. + + >>> from time import sleep + >>> def generator(): + ... yield 1 + ... yield 2 + ... sleep(0.2) + ... yield 3 + >>> iterable = time_limited(0.1, generator()) + >>> list(iterable) + [1, 2] + >>> iterable.timed_out + True + + Note that the time is checked before each item is yielded, and iteration + stops if the time elapsed is greater than *limit_seconds*. If your time + limit is 1 second, but it takes 2 seconds to generate the first item from + the iterable, the function will run for 2 seconds and not yield anything. + + """ + + def __init__(self, limit_seconds, iterable): + if limit_seconds < 0: + raise ValueError('limit_seconds must be positive') + self.limit_seconds = limit_seconds + self._iterable = iter(iterable) + self._start_time = monotonic() + self.timed_out = False + + def __iter__(self): + return self + + def __next__(self): + item = next(self._iterable) + if monotonic() - self._start_time > self.limit_seconds: + self.timed_out = True + raise StopIteration + + return item + + +def only(iterable, default=None, too_long=None): + """If *iterable* has only one item, return it. + If it has zero items, return *default*. + If it has more than one item, raise the exception given by *too_long*, + which is ``ValueError`` by default. + + >>> only([], default='missing') + 'missing' + >>> only([1]) + 1 + >>> only([1, 2]) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + ValueError: Expected exactly one item in iterable, but got 1, 2, + and perhaps more.' + >>> only([1, 2], too_long=TypeError) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + TypeError + + Note that :func:`only` attempts to advance *iterable* twice to ensure there + is only one item. See :func:`spy` or :func:`peekable` to check + iterable contents less destructively. + """ + it = iter(iterable) + first_value = next(it, default) + + try: + second_value = next(it) + except StopIteration: + pass + else: + msg = ( + 'Expected exactly one item in iterable, but got {!r}, {!r}, ' + 'and perhaps more.'.format(first_value, second_value) + ) + raise too_long or ValueError(msg) + + return first_value + + +class _IChunk: + def __init__(self, iterable, n): + self._it = islice(iterable, n) + self._cache = deque() + + def fill_cache(self): + self._cache.extend(self._it) + + def __iter__(self): + return self + + def __next__(self): + try: + return next(self._it) + except StopIteration: + if self._cache: + return self._cache.popleft() + else: + raise + + +def ichunked(iterable, n): + """Break *iterable* into sub-iterables with *n* elements each. + :func:`ichunked` is like :func:`chunked`, but it yields iterables + instead of lists. + + If the sub-iterables are read in order, the elements of *iterable* + won't be stored in memory. + If they are read out of order, :func:`itertools.tee` is used to cache + elements as necessary. + + >>> from itertools import count + >>> all_chunks = ichunked(count(), 4) + >>> c_1, c_2, c_3 = next(all_chunks), next(all_chunks), next(all_chunks) + >>> list(c_2) # c_1's elements have been cached; c_3's haven't been + [4, 5, 6, 7] + >>> list(c_1) + [0, 1, 2, 3] + >>> list(c_3) + [8, 9, 10, 11] + + """ + source = peekable(iter(iterable)) + ichunk_marker = object() + while True: + # Check to see whether we're at the end of the source iterable + item = source.peek(ichunk_marker) + if item is ichunk_marker: + return + + chunk = _IChunk(source, n) + yield chunk + + # Advance the source iterable and fill previous chunk's cache + chunk.fill_cache() + + +def iequals(*iterables): + """Return ``True`` if all given *iterables* are equal to each other, + which means that they contain the same elements in the same order. + + The function is useful for comparing iterables of different data types + or iterables that do not support equality checks. + + >>> iequals("abc", ['a', 'b', 'c'], ('a', 'b', 'c'), iter("abc")) + True + + >>> iequals("abc", "acb") + False + + Not to be confused with :func:`all_equals`, which checks whether all + elements of iterable are equal to each other. + + """ + return all(map(all_equal, zip_longest(*iterables, fillvalue=object()))) + + +def distinct_combinations(iterable, r): + """Yield the distinct combinations of *r* items taken from *iterable*. + + >>> list(distinct_combinations([0, 0, 1], 2)) + [(0, 0), (0, 1)] + + Equivalent to ``set(combinations(iterable))``, except duplicates are not + generated and thrown away. For larger input sequences this is much more + efficient. + + """ + if r < 0: + raise ValueError('r must be non-negative') + elif r == 0: + yield () + return + pool = tuple(iterable) + generators = [unique_everseen(enumerate(pool), key=itemgetter(1))] + current_combo = [None] * r + level = 0 + while generators: + try: + cur_idx, p = next(generators[-1]) + except StopIteration: + generators.pop() + level -= 1 + continue + current_combo[level] = p + if level + 1 == r: + yield tuple(current_combo) + else: + generators.append( + unique_everseen( + enumerate(pool[cur_idx + 1 :], cur_idx + 1), + key=itemgetter(1), + ) + ) + level += 1 + + +def filter_except(validator, iterable, *exceptions): + """Yield the items from *iterable* for which the *validator* function does + not raise one of the specified *exceptions*. + + *validator* is called for each item in *iterable*. + It should be a function that accepts one argument and raises an exception + if that item is not valid. + + >>> iterable = ['1', '2', 'three', '4', None] + >>> list(filter_except(int, iterable, ValueError, TypeError)) + ['1', '2', '4'] + + If an exception other than one given by *exceptions* is raised by + *validator*, it is raised like normal. + """ + for item in iterable: + try: + validator(item) + except exceptions: + pass + else: + yield item + + +def map_except(function, iterable, *exceptions): + """Transform each item from *iterable* with *function* and yield the + result, unless *function* raises one of the specified *exceptions*. + + *function* is called to transform each item in *iterable*. + It should accept one argument. + + >>> iterable = ['1', '2', 'three', '4', None] + >>> list(map_except(int, iterable, ValueError, TypeError)) + [1, 2, 4] + + If an exception other than one given by *exceptions* is raised by + *function*, it is raised like normal. + """ + for item in iterable: + try: + yield function(item) + except exceptions: + pass + + +def map_if(iterable, pred, func, func_else=lambda x: x): + """Evaluate each item from *iterable* using *pred*. If the result is + equivalent to ``True``, transform the item with *func* and yield it. + Otherwise, transform the item with *func_else* and yield it. + + *pred*, *func*, and *func_else* should each be functions that accept + one argument. By default, *func_else* is the identity function. + + >>> from math import sqrt + >>> iterable = list(range(-5, 5)) + >>> iterable + [-5, -4, -3, -2, -1, 0, 1, 2, 3, 4] + >>> list(map_if(iterable, lambda x: x > 3, lambda x: 'toobig')) + [-5, -4, -3, -2, -1, 0, 1, 2, 3, 'toobig'] + >>> list(map_if(iterable, lambda x: x >= 0, + ... lambda x: f'{sqrt(x):.2f}', lambda x: None)) + [None, None, None, None, None, '0.00', '1.00', '1.41', '1.73', '2.00'] + """ + for item in iterable: + yield func(item) if pred(item) else func_else(item) + + +def _sample_unweighted(iterable, k): + # Implementation of "Algorithm L" from the 1994 paper by Kim-Hung Li: + # "Reservoir-Sampling Algorithms of Time Complexity O(n(1+log(N/n)))". + + # Fill up the reservoir (collection of samples) with the first `k` samples + reservoir = take(k, iterable) + + # Generate random number that's the largest in a sample of k U(0,1) numbers + # Largest order statistic: https://en.wikipedia.org/wiki/Order_statistic + W = exp(log(random()) / k) + + # The number of elements to skip before changing the reservoir is a random + # number with a geometric distribution. Sample it using random() and logs. + next_index = k + floor(log(random()) / log(1 - W)) + + for index, element in enumerate(iterable, k): + if index == next_index: + reservoir[randrange(k)] = element + # The new W is the largest in a sample of k U(0, `old_W`) numbers + W *= exp(log(random()) / k) + next_index += floor(log(random()) / log(1 - W)) + 1 + + return reservoir + + +def _sample_weighted(iterable, k, weights): + # Implementation of "A-ExpJ" from the 2006 paper by Efraimidis et al. : + # "Weighted random sampling with a reservoir". + + # Log-transform for numerical stability for weights that are small/large + weight_keys = (log(random()) / weight for weight in weights) + + # Fill up the reservoir (collection of samples) with the first `k` + # weight-keys and elements, then heapify the list. + reservoir = take(k, zip(weight_keys, iterable)) + heapify(reservoir) + + # The number of jumps before changing the reservoir is a random variable + # with an exponential distribution. Sample it using random() and logs. + smallest_weight_key, _ = reservoir[0] + weights_to_skip = log(random()) / smallest_weight_key + + for weight, element in zip(weights, iterable): + if weight >= weights_to_skip: + # The notation here is consistent with the paper, but we store + # the weight-keys in log-space for better numerical stability. + smallest_weight_key, _ = reservoir[0] + t_w = exp(weight * smallest_weight_key) + r_2 = uniform(t_w, 1) # generate U(t_w, 1) + weight_key = log(r_2) / weight + heapreplace(reservoir, (weight_key, element)) + smallest_weight_key, _ = reservoir[0] + weights_to_skip = log(random()) / smallest_weight_key + else: + weights_to_skip -= weight + + # Equivalent to [element for weight_key, element in sorted(reservoir)] + return [heappop(reservoir)[1] for _ in range(k)] + + +def sample(iterable, k, weights=None): + """Return a *k*-length list of elements chosen (without replacement) + from the *iterable*. Like :func:`random.sample`, but works on iterables + of unknown length. + + >>> iterable = range(100) + >>> sample(iterable, 5) # doctest: +SKIP + [81, 60, 96, 16, 4] + + An iterable with *weights* may also be given: + + >>> iterable = range(100) + >>> weights = (i * i + 1 for i in range(100)) + >>> sampled = sample(iterable, 5, weights=weights) # doctest: +SKIP + [79, 67, 74, 66, 78] + + The algorithm can also be used to generate weighted random permutations. + The relative weight of each item determines the probability that it + appears late in the permutation. + + >>> data = "abcdefgh" + >>> weights = range(1, len(data) + 1) + >>> sample(data, k=len(data), weights=weights) # doctest: +SKIP + ['c', 'a', 'b', 'e', 'g', 'd', 'h', 'f'] + """ + if k == 0: + return [] + + iterable = iter(iterable) + if weights is None: + return _sample_unweighted(iterable, k) + else: + weights = iter(weights) + return _sample_weighted(iterable, k, weights) + + +def is_sorted(iterable, key=None, reverse=False, strict=False): + """Returns ``True`` if the items of iterable are in sorted order, and + ``False`` otherwise. *key* and *reverse* have the same meaning that they do + in the built-in :func:`sorted` function. + + >>> is_sorted(['1', '2', '3', '4', '5'], key=int) + True + >>> is_sorted([5, 4, 3, 1, 2], reverse=True) + False + + If *strict*, tests for strict sorting, that is, returns ``False`` if equal + elements are found: + + >>> is_sorted([1, 2, 2]) + True + >>> is_sorted([1, 2, 2], strict=True) + False + + The function returns ``False`` after encountering the first out-of-order + item. If there are no out-of-order items, the iterable is exhausted. + """ + + compare = (le if reverse else ge) if strict else (lt if reverse else gt) + it = iterable if key is None else map(key, iterable) + return not any(starmap(compare, pairwise(it))) + + +class AbortThread(BaseException): + pass + + +class callback_iter: + """Convert a function that uses callbacks to an iterator. + + Let *func* be a function that takes a `callback` keyword argument. + For example: + + >>> def func(callback=None): + ... for i, c in [(1, 'a'), (2, 'b'), (3, 'c')]: + ... if callback: + ... callback(i, c) + ... return 4 + + + Use ``with callback_iter(func)`` to get an iterator over the parameters + that are delivered to the callback. + + >>> with callback_iter(func) as it: + ... for args, kwargs in it: + ... print(args) + (1, 'a') + (2, 'b') + (3, 'c') + + The function will be called in a background thread. The ``done`` property + indicates whether it has completed execution. + + >>> it.done + True + + If it completes successfully, its return value will be available + in the ``result`` property. + + >>> it.result + 4 + + Notes: + + * If the function uses some keyword argument besides ``callback``, supply + *callback_kwd*. + * If it finished executing, but raised an exception, accessing the + ``result`` property will raise the same exception. + * If it hasn't finished executing, accessing the ``result`` + property from within the ``with`` block will raise ``RuntimeError``. + * If it hasn't finished executing, accessing the ``result`` property from + outside the ``with`` block will raise a + ``more_itertools.AbortThread`` exception. + * Provide *wait_seconds* to adjust how frequently the it is polled for + output. + + """ + + def __init__(self, func, callback_kwd='callback', wait_seconds=0.1): + self._func = func + self._callback_kwd = callback_kwd + self._aborted = False + self._future = None + self._wait_seconds = wait_seconds + # Lazily import concurrent.future + self._executor = __import__( + ).futures.__import__("concurrent.futures").futures.ThreadPoolExecutor(max_workers=1) + self._iterator = self._reader() + + def __enter__(self): + return self + + def __exit__(self, exc_type, exc_value, traceback): + self._aborted = True + self._executor.shutdown() + + def __iter__(self): + return self + + def __next__(self): + return next(self._iterator) + + @property + def done(self): + if self._future is None: + return False + return self._future.done() + + @property + def result(self): + if not self.done: + raise RuntimeError('Function has not yet completed') + + return self._future.result() + + def _reader(self): + q = Queue() + + def callback(*args, **kwargs): + if self._aborted: + raise AbortThread('canceled by user') + + q.put((args, kwargs)) + + self._future = self._executor.submit( + self._func, **{self._callback_kwd: callback} + ) + + while True: + try: + item = q.get(timeout=self._wait_seconds) + except Empty: + pass + else: + q.task_done() + yield item + + if self._future.done(): + break + + remaining = [] + while True: + try: + item = q.get_nowait() + except Empty: + break + else: + q.task_done() + remaining.append(item) + q.join() + yield from remaining + + +def windowed_complete(iterable, n): + """ + Yield ``(beginning, middle, end)`` tuples, where: + + * Each ``middle`` has *n* items from *iterable* + * Each ``beginning`` has the items before the ones in ``middle`` + * Each ``end`` has the items after the ones in ``middle`` + + >>> iterable = range(7) + >>> n = 3 + >>> for beginning, middle, end in windowed_complete(iterable, n): + ... print(beginning, middle, end) + () (0, 1, 2) (3, 4, 5, 6) + (0,) (1, 2, 3) (4, 5, 6) + (0, 1) (2, 3, 4) (5, 6) + (0, 1, 2) (3, 4, 5) (6,) + (0, 1, 2, 3) (4, 5, 6) () + + Note that *n* must be at least 0 and most equal to the length of + *iterable*. + + This function will exhaust the iterable and may require significant + storage. + """ + if n < 0: + raise ValueError('n must be >= 0') + + seq = tuple(iterable) + size = len(seq) + + if n > size: + raise ValueError('n must be <= len(seq)') + + for i in range(size - n + 1): + beginning = seq[:i] + middle = seq[i : i + n] + end = seq[i + n :] + yield beginning, middle, end + + +def all_unique(iterable, key=None): + """ + Returns ``True`` if all the elements of *iterable* are unique (no two + elements are equal). + + >>> all_unique('ABCB') + False + + If a *key* function is specified, it will be used to make comparisons. + + >>> all_unique('ABCb') + True + >>> all_unique('ABCb', str.lower) + False + + The function returns as soon as the first non-unique element is + encountered. Iterables with a mix of hashable and unhashable items can + be used, but the function will be slower for unhashable items. + """ + seenset = set() + seenset_add = seenset.add + seenlist = [] + seenlist_add = seenlist.append + for element in map(key, iterable) if key else iterable: + try: + if element in seenset: + return False + seenset_add(element) + except TypeError: + if element in seenlist: + return False + seenlist_add(element) + return True + + +def nth_product(index, *args): + """Equivalent to ``list(product(*args))[index]``. + + The products of *args* can be ordered lexicographically. + :func:`nth_product` computes the product at sort position *index* without + computing the previous products. + + >>> nth_product(8, range(2), range(2), range(2), range(2)) + (1, 0, 0, 0) + + ``IndexError`` will be raised if the given *index* is invalid. + """ + pools = list(map(tuple, reversed(args))) + ns = list(map(len, pools)) + + c = reduce(mul, ns) + + if index < 0: + index += c + + if not 0 <= index < c: + raise IndexError + + result = [] + for pool, n in zip(pools, ns): + result.append(pool[index % n]) + index //= n + + return tuple(reversed(result)) + + +def nth_permutation(iterable, r, index): + """Equivalent to ``list(permutations(iterable, r))[index]``` + + The subsequences of *iterable* that are of length *r* where order is + important can be ordered lexicographically. :func:`nth_permutation` + computes the subsequence at sort position *index* directly, without + computing the previous subsequences. + + >>> nth_permutation('ghijk', 2, 5) + ('h', 'i') + + ``ValueError`` will be raised If *r* is negative or greater than the length + of *iterable*. + ``IndexError`` will be raised if the given *index* is invalid. + """ + pool = list(iterable) + n = len(pool) + + if r is None or r == n: + r, c = n, factorial(n) + elif not 0 <= r < n: + raise ValueError + else: + c = factorial(n) // factorial(n - r) + + if index < 0: + index += c + + if not 0 <= index < c: + raise IndexError + + if c == 0: + return tuple() + + result = [0] * r + q = index * factorial(n) // c if r < n else index + for d in range(1, n + 1): + q, i = divmod(q, d) + if 0 <= n - d < r: + result[n - d] = i + if q == 0: + break + + return tuple(map(pool.pop, result)) + + +def value_chain(*args): + """Yield all arguments passed to the function in the same order in which + they were passed. If an argument itself is iterable then iterate over its + values. + + >>> list(value_chain(1, 2, 3, [4, 5, 6])) + [1, 2, 3, 4, 5, 6] + + Binary and text strings are not considered iterable and are emitted + as-is: + + >>> list(value_chain('12', '34', ['56', '78'])) + ['12', '34', '56', '78'] + + + Multiple levels of nesting are not flattened. + + """ + for value in args: + if isinstance(value, (str, bytes)): + yield value + continue + try: + yield from value + except TypeError: + yield value + + +def product_index(element, *args): + """Equivalent to ``list(product(*args)).index(element)`` + + The products of *args* can be ordered lexicographically. + :func:`product_index` computes the first index of *element* without + computing the previous products. + + >>> product_index([8, 2], range(10), range(5)) + 42 + + ``ValueError`` will be raised if the given *element* isn't in the product + of *args*. + """ + index = 0 + + for x, pool in zip_longest(element, args, fillvalue=_marker): + if x is _marker or pool is _marker: + raise ValueError('element is not a product of args') + + pool = tuple(pool) + index = index * len(pool) + pool.index(x) + + return index + + +def combination_index(element, iterable): + """Equivalent to ``list(combinations(iterable, r)).index(element)`` + + The subsequences of *iterable* that are of length *r* can be ordered + lexicographically. :func:`combination_index` computes the index of the + first *element*, without computing the previous combinations. + + >>> combination_index('adf', 'abcdefg') + 10 + + ``ValueError`` will be raised if the given *element* isn't one of the + combinations of *iterable*. + """ + element = enumerate(element) + k, y = next(element, (None, None)) + if k is None: + return 0 + + indexes = [] + pool = enumerate(iterable) + for n, x in pool: + if x == y: + indexes.append(n) + tmp, y = next(element, (None, None)) + if tmp is None: + break + else: + k = tmp + else: + raise ValueError('element is not a combination of iterable') + + n, _ = last(pool, default=(n, None)) + + # Python versions below 3.8 don't have math.comb + index = 1 + for i, j in enumerate(reversed(indexes), start=1): + j = n - j + if i <= j: + index += factorial(j) // (factorial(i) * factorial(j - i)) + + return factorial(n + 1) // (factorial(k + 1) * factorial(n - k)) - index + + +def permutation_index(element, iterable): + """Equivalent to ``list(permutations(iterable, r)).index(element)``` + + The subsequences of *iterable* that are of length *r* where order is + important can be ordered lexicographically. :func:`permutation_index` + computes the index of the first *element* directly, without computing + the previous permutations. + + >>> permutation_index([1, 3, 2], range(5)) + 19 + + ``ValueError`` will be raised if the given *element* isn't one of the + permutations of *iterable*. + """ + index = 0 + pool = list(iterable) + for i, x in zip(range(len(pool), -1, -1), element): + r = pool.index(x) + index = index * i + r + del pool[r] + + return index + + +class countable: + """Wrap *iterable* and keep a count of how many items have been consumed. + + The ``items_seen`` attribute starts at ``0`` and increments as the iterable + is consumed: + + >>> iterable = map(str, range(10)) + >>> it = countable(iterable) + >>> it.items_seen + 0 + >>> next(it), next(it) + ('0', '1') + >>> list(it) + ['2', '3', '4', '5', '6', '7', '8', '9'] + >>> it.items_seen + 10 + """ + + def __init__(self, iterable): + self._it = iter(iterable) + self.items_seen = 0 + + def __iter__(self): + return self + + def __next__(self): + item = next(self._it) + self.items_seen += 1 + + return item + + +def chunked_even(iterable, n): + """Break *iterable* into lists of approximately length *n*. + Items are distributed such the lengths of the lists differ by at most + 1 item. + + >>> iterable = [1, 2, 3, 4, 5, 6, 7] + >>> n = 3 + >>> list(chunked_even(iterable, n)) # List lengths: 3, 2, 2 + [[1, 2, 3], [4, 5], [6, 7]] + >>> list(chunked(iterable, n)) # List lengths: 3, 3, 1 + [[1, 2, 3], [4, 5, 6], [7]] + + """ + + len_method = getattr(iterable, '__len__', None) + + if len_method is None: + return _chunked_even_online(iterable, n) + else: + return _chunked_even_finite(iterable, len_method(), n) + + +def _chunked_even_online(iterable, n): + buffer = [] + maxbuf = n + (n - 2) * (n - 1) + for x in iterable: + buffer.append(x) + if len(buffer) == maxbuf: + yield buffer[:n] + buffer = buffer[n:] + yield from _chunked_even_finite(buffer, len(buffer), n) + + +def _chunked_even_finite(iterable, N, n): + if N < 1: + return + + # Lists are either size `full_size <= n` or `partial_size = full_size - 1` + q, r = divmod(N, n) + num_lists = q + (1 if r > 0 else 0) + q, r = divmod(N, num_lists) + full_size = q + (1 if r > 0 else 0) + partial_size = full_size - 1 + num_full = N - partial_size * num_lists + num_partial = num_lists - num_full + + buffer = [] + iterator = iter(iterable) + + # Yield num_full lists of full_size + for x in iterator: + buffer.append(x) + if len(buffer) == full_size: + yield buffer + buffer = [] + num_full -= 1 + if num_full <= 0: + break + + # Yield num_partial lists of partial_size + for x in iterator: + buffer.append(x) + if len(buffer) == partial_size: + yield buffer + buffer = [] + num_partial -= 1 + + +def zip_broadcast(*objects, scalar_types=(str, bytes), strict=False): + """A version of :func:`zip` that "broadcasts" any scalar + (i.e., non-iterable) items into output tuples. + + >>> iterable_1 = [1, 2, 3] + >>> iterable_2 = ['a', 'b', 'c'] + >>> scalar = '_' + >>> list(zip_broadcast(iterable_1, iterable_2, scalar)) + [(1, 'a', '_'), (2, 'b', '_'), (3, 'c', '_')] + + The *scalar_types* keyword argument determines what types are considered + scalar. It is set to ``(str, bytes)`` by default. Set it to ``None`` to + treat strings and byte strings as iterable: + + >>> list(zip_broadcast('abc', 0, 'xyz', scalar_types=None)) + [('a', 0, 'x'), ('b', 0, 'y'), ('c', 0, 'z')] + + If the *strict* keyword argument is ``True``, then + ``UnequalIterablesError`` will be raised if any of the iterables have + different lengths. + """ + + def is_scalar(obj): + if scalar_types and isinstance(obj, scalar_types): + return True + try: + iter(obj) + except TypeError: + return True + else: + return False + + size = len(objects) + if not size: + return + + iterables, iterable_positions = [], [] + scalars, scalar_positions = [], [] + for i, obj in enumerate(objects): + if is_scalar(obj): + scalars.append(obj) + scalar_positions.append(i) + else: + iterables.append(iter(obj)) + iterable_positions.append(i) + + if len(scalars) == size: + yield tuple(objects) + return + + zipper = _zip_equal if strict else zip + for item in zipper(*iterables): + new_item = [None] * size + + for i, elem in zip(iterable_positions, item): + new_item[i] = elem + + for i, elem in zip(scalar_positions, scalars): + new_item[i] = elem + + yield tuple(new_item) + + +def unique_in_window(iterable, n, key=None): + """Yield the items from *iterable* that haven't been seen recently. + *n* is the size of the lookback window. + + >>> iterable = [0, 1, 0, 2, 3, 0] + >>> n = 3 + >>> list(unique_in_window(iterable, n)) + [0, 1, 2, 3, 0] + + The *key* function, if provided, will be used to determine uniqueness: + + >>> list(unique_in_window('abAcda', 3, key=lambda x: x.lower())) + ['a', 'b', 'c', 'd', 'a'] + + The items in *iterable* must be hashable. + + """ + if n <= 0: + raise ValueError('n must be greater than 0') + + window = deque(maxlen=n) + uniques = set() + use_key = key is not None + + for item in iterable: + k = key(item) if use_key else item + if k in uniques: + continue + + if len(uniques) == n: + uniques.discard(window[0]) + + uniques.add(k) + window.append(k) + + yield item + + +def duplicates_everseen(iterable, key=None): + """Yield duplicate elements after their first appearance. + + >>> list(duplicates_everseen('mississippi')) + ['s', 'i', 's', 's', 'i', 'p', 'i'] + >>> list(duplicates_everseen('AaaBbbCccAaa', str.lower)) + ['a', 'a', 'b', 'b', 'c', 'c', 'A', 'a', 'a'] + + This function is analagous to :func:`unique_everseen` and is subject to + the same performance considerations. + + """ + seen_set = set() + seen_list = [] + use_key = key is not None + + for element in iterable: + k = key(element) if use_key else element + try: + if k not in seen_set: + seen_set.add(k) + else: + yield element + except TypeError: + if k not in seen_list: + seen_list.append(k) + else: + yield element + + +def duplicates_justseen(iterable, key=None): + """Yields serially-duplicate elements after their first appearance. + + >>> list(duplicates_justseen('mississippi')) + ['s', 's', 'p'] + >>> list(duplicates_justseen('AaaBbbCccAaa', str.lower)) + ['a', 'a', 'b', 'b', 'c', 'c', 'a', 'a'] + + This function is analagous to :func:`unique_justseen`. + + """ + return flatten( + map( + lambda group_tuple: islice_extended(group_tuple[1])[1:], + groupby(iterable, key), + ) + ) + + +def minmax(iterable_or_value, *others, key=None, default=_marker): + """Returns both the smallest and largest items in an iterable + or the largest of two or more arguments. + + >>> minmax([3, 1, 5]) + (1, 5) + + >>> minmax(4, 2, 6) + (2, 6) + + If a *key* function is provided, it will be used to transform the input + items for comparison. + + >>> minmax([5, 30], key=str) # '30' sorts before '5' + (30, 5) + + If a *default* value is provided, it will be returned if there are no + input items. + + >>> minmax([], default=(0, 0)) + (0, 0) + + Otherwise ``ValueError`` is raised. + + This function is based on the + `recipe <http://code.activestate.com/recipes/577916/>`__ by + Raymond Hettinger and takes care to minimize the number of comparisons + performed. + """ + iterable = (iterable_or_value, *others) if others else iterable_or_value + + it = iter(iterable) + + try: + lo = hi = next(it) + except StopIteration as e: + if default is _marker: + raise ValueError( + '`minmax()` argument is an empty iterable. ' + 'Provide a `default` value to suppress this error.' + ) from e + return default + + # Different branches depending on the presence of key. This saves a lot + # of unimportant copies which would slow the "key=None" branch + # significantly down. + if key is None: + for x, y in zip_longest(it, it, fillvalue=lo): + if y < x: + x, y = y, x + if x < lo: + lo = x + if hi < y: + hi = y + + else: + lo_key = hi_key = key(lo) + + for x, y in zip_longest(it, it, fillvalue=lo): + x_key, y_key = key(x), key(y) + + if y_key < x_key: + x, y, x_key, y_key = y, x, y_key, x_key + if x_key < lo_key: + lo, lo_key = x, x_key + if hi_key < y_key: + hi, hi_key = y, y_key + + return lo, hi + + +def constrained_batches( + iterable, max_size, max_count=None, get_len=len, strict=True +): + """Yield batches of items from *iterable* with a combined size limited by + *max_size*. + + >>> iterable = [b'12345', b'123', b'12345678', b'1', b'1', b'12', b'1'] + >>> list(constrained_batches(iterable, 10)) + [(b'12345', b'123'), (b'12345678', b'1', b'1'), (b'12', b'1')] + + If a *max_count* is supplied, the number of items per batch is also + limited: + + >>> iterable = [b'12345', b'123', b'12345678', b'1', b'1', b'12', b'1'] + >>> list(constrained_batches(iterable, 10, max_count = 2)) + [(b'12345', b'123'), (b'12345678', b'1'), (b'1', b'12'), (b'1',)] + + If a *get_len* function is supplied, use that instead of :func:`len` to + determine item size. + + If *strict* is ``True``, raise ``ValueError`` if any single item is bigger + than *max_size*. Otherwise, allow single items to exceed *max_size*. + """ + if max_size <= 0: + raise ValueError('maximum size must be greater than zero') + + batch = [] + batch_size = 0 + batch_count = 0 + for item in iterable: + item_len = get_len(item) + if strict and item_len > max_size: + raise ValueError('item size exceeds maximum size') + + reached_count = batch_count == max_count + reached_size = item_len + batch_size > max_size + if batch_count and (reached_size or reached_count): + yield tuple(batch) + batch.clear() + batch_size = 0 + batch_count = 0 + + batch.append(item) + batch_size += item_len + batch_count += 1 + + if batch: + yield tuple(batch) + + +def gray_product(*iterables): + """Like :func:`itertools.product`, but return tuples in an order such + that only one element in the generated tuple changes from one iteration + to the next. + + >>> list(gray_product('AB','CD')) + [('A', 'C'), ('B', 'C'), ('B', 'D'), ('A', 'D')] + + This function consumes all of the input iterables before producing output. + If any of the input iterables have fewer than two items, ``ValueError`` + is raised. + + For information on the algorithm, see + `this section <https://www-cs-faculty.stanford.edu/~knuth/fasc2a.ps.gz>`__ + of Donald Knuth's *The Art of Computer Programming*. + """ + all_iterables = tuple(tuple(x) for x in iterables) + iterable_count = len(all_iterables) + for iterable in all_iterables: + if len(iterable) < 2: + raise ValueError("each iterable must have two or more items") + + # This is based on "Algorithm H" from section 7.2.1.1, page 20. + # a holds the indexes of the source iterables for the n-tuple to be yielded + # f is the array of "focus pointers" + # o is the array of "directions" + a = [0] * iterable_count + f = list(range(iterable_count + 1)) + o = [1] * iterable_count + while True: + yield tuple(all_iterables[i][a[i]] for i in range(iterable_count)) + j = f[0] + f[0] = 0 + if j == iterable_count: + break + a[j] = a[j] + o[j] + if a[j] == 0 or a[j] == len(all_iterables[j]) - 1: + o[j] = -o[j] + f[j] = f[j + 1] + f[j + 1] = j + 1 diff --git a/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/recipes.py b/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/recipes.py new file mode 100644 index 0000000000..3facc2e3a6 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/more_itertools/recipes.py @@ -0,0 +1,930 @@ +"""Imported from the recipes section of the itertools documentation. + +All functions taken from the recipes section of the itertools library docs +[1]_. +Some backward-compatible usability improvements have been made. + +.. [1] http://docs.python.org/library/itertools.html#recipes + +""" +import math +import operator +import warnings + +from collections import deque +from collections.abc import Sized +from functools import reduce +from itertools import ( + chain, + combinations, + compress, + count, + cycle, + groupby, + islice, + product, + repeat, + starmap, + tee, + zip_longest, +) +from random import randrange, sample, choice +from sys import hexversion + +__all__ = [ + 'all_equal', + 'batched', + 'before_and_after', + 'consume', + 'convolve', + 'dotproduct', + 'first_true', + 'factor', + 'flatten', + 'grouper', + 'iter_except', + 'iter_index', + 'matmul', + 'ncycles', + 'nth', + 'nth_combination', + 'padnone', + 'pad_none', + 'pairwise', + 'partition', + 'polynomial_from_roots', + 'powerset', + 'prepend', + 'quantify', + 'random_combination_with_replacement', + 'random_combination', + 'random_permutation', + 'random_product', + 'repeatfunc', + 'roundrobin', + 'sieve', + 'sliding_window', + 'subslices', + 'tabulate', + 'tail', + 'take', + 'transpose', + 'triplewise', + 'unique_everseen', + 'unique_justseen', +] + +_marker = object() + + +def take(n, iterable): + """Return first *n* items of the iterable as a list. + + >>> take(3, range(10)) + [0, 1, 2] + + If there are fewer than *n* items in the iterable, all of them are + returned. + + >>> take(10, range(3)) + [0, 1, 2] + + """ + return list(islice(iterable, n)) + + +def tabulate(function, start=0): + """Return an iterator over the results of ``func(start)``, + ``func(start + 1)``, ``func(start + 2)``... + + *func* should be a function that accepts one integer argument. + + If *start* is not specified it defaults to 0. It will be incremented each + time the iterator is advanced. + + >>> square = lambda x: x ** 2 + >>> iterator = tabulate(square, -3) + >>> take(4, iterator) + [9, 4, 1, 0] + + """ + return map(function, count(start)) + + +def tail(n, iterable): + """Return an iterator over the last *n* items of *iterable*. + + >>> t = tail(3, 'ABCDEFG') + >>> list(t) + ['E', 'F', 'G'] + + """ + # If the given iterable has a length, then we can use islice to get its + # final elements. Note that if the iterable is not actually Iterable, + # either islice or deque will throw a TypeError. This is why we don't + # check if it is Iterable. + if isinstance(iterable, Sized): + yield from islice(iterable, max(0, len(iterable) - n), None) + else: + yield from iter(deque(iterable, maxlen=n)) + + +def consume(iterator, n=None): + """Advance *iterable* by *n* steps. If *n* is ``None``, consume it + entirely. + + Efficiently exhausts an iterator without returning values. Defaults to + consuming the whole iterator, but an optional second argument may be + provided to limit consumption. + + >>> i = (x for x in range(10)) + >>> next(i) + 0 + >>> consume(i, 3) + >>> next(i) + 4 + >>> consume(i) + >>> next(i) + Traceback (most recent call last): + File "<stdin>", line 1, in <module> + StopIteration + + If the iterator has fewer items remaining than the provided limit, the + whole iterator will be consumed. + + >>> i = (x for x in range(3)) + >>> consume(i, 5) + >>> next(i) + Traceback (most recent call last): + File "<stdin>", line 1, in <module> + StopIteration + + """ + # Use functions that consume iterators at C speed. + if n is None: + # feed the entire iterator into a zero-length deque + deque(iterator, maxlen=0) + else: + # advance to the empty slice starting at position n + next(islice(iterator, n, n), None) + + +def nth(iterable, n, default=None): + """Returns the nth item or a default value. + + >>> l = range(10) + >>> nth(l, 3) + 3 + >>> nth(l, 20, "zebra") + 'zebra' + + """ + return next(islice(iterable, n, None), default) + + +def all_equal(iterable): + """ + Returns ``True`` if all the elements are equal to each other. + + >>> all_equal('aaaa') + True + >>> all_equal('aaab') + False + + """ + g = groupby(iterable) + return next(g, True) and not next(g, False) + + +def quantify(iterable, pred=bool): + """Return the how many times the predicate is true. + + >>> quantify([True, False, True]) + 2 + + """ + return sum(map(pred, iterable)) + + +def pad_none(iterable): + """Returns the sequence of elements and then returns ``None`` indefinitely. + + >>> take(5, pad_none(range(3))) + [0, 1, 2, None, None] + + Useful for emulating the behavior of the built-in :func:`map` function. + + See also :func:`padded`. + + """ + return chain(iterable, repeat(None)) + + +padnone = pad_none + + +def ncycles(iterable, n): + """Returns the sequence elements *n* times + + >>> list(ncycles(["a", "b"], 3)) + ['a', 'b', 'a', 'b', 'a', 'b'] + + """ + return chain.from_iterable(repeat(tuple(iterable), n)) + + +def dotproduct(vec1, vec2): + """Returns the dot product of the two iterables. + + >>> dotproduct([10, 10], [20, 20]) + 400 + + """ + return sum(map(operator.mul, vec1, vec2)) + + +def flatten(listOfLists): + """Return an iterator flattening one level of nesting in a list of lists. + + >>> list(flatten([[0, 1], [2, 3]])) + [0, 1, 2, 3] + + See also :func:`collapse`, which can flatten multiple levels of nesting. + + """ + return chain.from_iterable(listOfLists) + + +def repeatfunc(func, times=None, *args): + """Call *func* with *args* repeatedly, returning an iterable over the + results. + + If *times* is specified, the iterable will terminate after that many + repetitions: + + >>> from operator import add + >>> times = 4 + >>> args = 3, 5 + >>> list(repeatfunc(add, times, *args)) + [8, 8, 8, 8] + + If *times* is ``None`` the iterable will not terminate: + + >>> from random import randrange + >>> times = None + >>> args = 1, 11 + >>> take(6, repeatfunc(randrange, times, *args)) # doctest:+SKIP + [2, 4, 8, 1, 8, 4] + + """ + if times is None: + return starmap(func, repeat(args)) + return starmap(func, repeat(args, times)) + + +def _pairwise(iterable): + """Returns an iterator of paired items, overlapping, from the original + + >>> take(4, pairwise(count())) + [(0, 1), (1, 2), (2, 3), (3, 4)] + + On Python 3.10 and above, this is an alias for :func:`itertools.pairwise`. + + """ + a, b = tee(iterable) + next(b, None) + yield from zip(a, b) + + +try: + from itertools import pairwise as itertools_pairwise +except ImportError: + pairwise = _pairwise +else: + + def pairwise(iterable): + yield from itertools_pairwise(iterable) + + pairwise.__doc__ = _pairwise.__doc__ + + +class UnequalIterablesError(ValueError): + def __init__(self, details=None): + msg = 'Iterables have different lengths' + if details is not None: + msg += (': index 0 has length {}; index {} has length {}').format( + *details + ) + + super().__init__(msg) + + +def _zip_equal_generator(iterables): + for combo in zip_longest(*iterables, fillvalue=_marker): + for val in combo: + if val is _marker: + raise UnequalIterablesError() + yield combo + + +def _zip_equal(*iterables): + # Check whether the iterables are all the same size. + try: + first_size = len(iterables[0]) + for i, it in enumerate(iterables[1:], 1): + size = len(it) + if size != first_size: + break + else: + # If we didn't break out, we can use the built-in zip. + return zip(*iterables) + + # If we did break out, there was a mismatch. + raise UnequalIterablesError(details=(first_size, i, size)) + # If any one of the iterables didn't have a length, start reading + # them until one runs out. + except TypeError: + return _zip_equal_generator(iterables) + + +def grouper(iterable, n, incomplete='fill', fillvalue=None): + """Group elements from *iterable* into fixed-length groups of length *n*. + + >>> list(grouper('ABCDEF', 3)) + [('A', 'B', 'C'), ('D', 'E', 'F')] + + The keyword arguments *incomplete* and *fillvalue* control what happens for + iterables whose length is not a multiple of *n*. + + When *incomplete* is `'fill'`, the last group will contain instances of + *fillvalue*. + + >>> list(grouper('ABCDEFG', 3, incomplete='fill', fillvalue='x')) + [('A', 'B', 'C'), ('D', 'E', 'F'), ('G', 'x', 'x')] + + When *incomplete* is `'ignore'`, the last group will not be emitted. + + >>> list(grouper('ABCDEFG', 3, incomplete='ignore', fillvalue='x')) + [('A', 'B', 'C'), ('D', 'E', 'F')] + + When *incomplete* is `'strict'`, a subclass of `ValueError` will be raised. + + >>> it = grouper('ABCDEFG', 3, incomplete='strict') + >>> list(it) # doctest: +IGNORE_EXCEPTION_DETAIL + Traceback (most recent call last): + ... + UnequalIterablesError + + """ + args = [iter(iterable)] * n + if incomplete == 'fill': + return zip_longest(*args, fillvalue=fillvalue) + if incomplete == 'strict': + return _zip_equal(*args) + if incomplete == 'ignore': + return zip(*args) + else: + raise ValueError('Expected fill, strict, or ignore') + + +def roundrobin(*iterables): + """Yields an item from each iterable, alternating between them. + + >>> list(roundrobin('ABC', 'D', 'EF')) + ['A', 'D', 'E', 'B', 'F', 'C'] + + This function produces the same output as :func:`interleave_longest`, but + may perform better for some inputs (in particular when the number of + iterables is small). + + """ + # Recipe credited to George Sakkis + pending = len(iterables) + nexts = cycle(iter(it).__next__ for it in iterables) + while pending: + try: + for next in nexts: + yield next() + except StopIteration: + pending -= 1 + nexts = cycle(islice(nexts, pending)) + + +def partition(pred, iterable): + """ + Returns a 2-tuple of iterables derived from the input iterable. + The first yields the items that have ``pred(item) == False``. + The second yields the items that have ``pred(item) == True``. + + >>> is_odd = lambda x: x % 2 != 0 + >>> iterable = range(10) + >>> even_items, odd_items = partition(is_odd, iterable) + >>> list(even_items), list(odd_items) + ([0, 2, 4, 6, 8], [1, 3, 5, 7, 9]) + + If *pred* is None, :func:`bool` is used. + + >>> iterable = [0, 1, False, True, '', ' '] + >>> false_items, true_items = partition(None, iterable) + >>> list(false_items), list(true_items) + ([0, False, ''], [1, True, ' ']) + + """ + if pred is None: + pred = bool + + evaluations = ((pred(x), x) for x in iterable) + t1, t2 = tee(evaluations) + return ( + (x for (cond, x) in t1 if not cond), + (x for (cond, x) in t2 if cond), + ) + + +def powerset(iterable): + """Yields all possible subsets of the iterable. + + >>> list(powerset([1, 2, 3])) + [(), (1,), (2,), (3,), (1, 2), (1, 3), (2, 3), (1, 2, 3)] + + :func:`powerset` will operate on iterables that aren't :class:`set` + instances, so repeated elements in the input will produce repeated elements + in the output. Use :func:`unique_everseen` on the input to avoid generating + duplicates: + + >>> seq = [1, 1, 0] + >>> list(powerset(seq)) + [(), (1,), (1,), (0,), (1, 1), (1, 0), (1, 0), (1, 1, 0)] + >>> from more_itertools import unique_everseen + >>> list(powerset(unique_everseen(seq))) + [(), (1,), (0,), (1, 0)] + + """ + s = list(iterable) + return chain.from_iterable(combinations(s, r) for r in range(len(s) + 1)) + + +def unique_everseen(iterable, key=None): + """ + Yield unique elements, preserving order. + + >>> list(unique_everseen('AAAABBBCCDAABBB')) + ['A', 'B', 'C', 'D'] + >>> list(unique_everseen('ABBCcAD', str.lower)) + ['A', 'B', 'C', 'D'] + + Sequences with a mix of hashable and unhashable items can be used. + The function will be slower (i.e., `O(n^2)`) for unhashable items. + + Remember that ``list`` objects are unhashable - you can use the *key* + parameter to transform the list to a tuple (which is hashable) to + avoid a slowdown. + + >>> iterable = ([1, 2], [2, 3], [1, 2]) + >>> list(unique_everseen(iterable)) # Slow + [[1, 2], [2, 3]] + >>> list(unique_everseen(iterable, key=tuple)) # Faster + [[1, 2], [2, 3]] + + Similary, you may want to convert unhashable ``set`` objects with + ``key=frozenset``. For ``dict`` objects, + ``key=lambda x: frozenset(x.items())`` can be used. + + """ + seenset = set() + seenset_add = seenset.add + seenlist = [] + seenlist_add = seenlist.append + use_key = key is not None + + for element in iterable: + k = key(element) if use_key else element + try: + if k not in seenset: + seenset_add(k) + yield element + except TypeError: + if k not in seenlist: + seenlist_add(k) + yield element + + +def unique_justseen(iterable, key=None): + """Yields elements in order, ignoring serial duplicates + + >>> list(unique_justseen('AAAABBBCCDAABBB')) + ['A', 'B', 'C', 'D', 'A', 'B'] + >>> list(unique_justseen('ABBCcAD', str.lower)) + ['A', 'B', 'C', 'A', 'D'] + + """ + return map(next, map(operator.itemgetter(1), groupby(iterable, key))) + + +def iter_except(func, exception, first=None): + """Yields results from a function repeatedly until an exception is raised. + + Converts a call-until-exception interface to an iterator interface. + Like ``iter(func, sentinel)``, but uses an exception instead of a sentinel + to end the loop. + + >>> l = [0, 1, 2] + >>> list(iter_except(l.pop, IndexError)) + [2, 1, 0] + + Multiple exceptions can be specified as a stopping condition: + + >>> l = [1, 2, 3, '...', 4, 5, 6] + >>> list(iter_except(lambda: 1 + l.pop(), (IndexError, TypeError))) + [7, 6, 5] + >>> list(iter_except(lambda: 1 + l.pop(), (IndexError, TypeError))) + [4, 3, 2] + >>> list(iter_except(lambda: 1 + l.pop(), (IndexError, TypeError))) + [] + + """ + try: + if first is not None: + yield first() + while 1: + yield func() + except exception: + pass + + +def first_true(iterable, default=None, pred=None): + """ + Returns the first true value in the iterable. + + If no true value is found, returns *default* + + If *pred* is not None, returns the first item for which + ``pred(item) == True`` . + + >>> first_true(range(10)) + 1 + >>> first_true(range(10), pred=lambda x: x > 5) + 6 + >>> first_true(range(10), default='missing', pred=lambda x: x > 9) + 'missing' + + """ + return next(filter(pred, iterable), default) + + +def random_product(*args, repeat=1): + """Draw an item at random from each of the input iterables. + + >>> random_product('abc', range(4), 'XYZ') # doctest:+SKIP + ('c', 3, 'Z') + + If *repeat* is provided as a keyword argument, that many items will be + drawn from each iterable. + + >>> random_product('abcd', range(4), repeat=2) # doctest:+SKIP + ('a', 2, 'd', 3) + + This equivalent to taking a random selection from + ``itertools.product(*args, **kwarg)``. + + """ + pools = [tuple(pool) for pool in args] * repeat + return tuple(choice(pool) for pool in pools) + + +def random_permutation(iterable, r=None): + """Return a random *r* length permutation of the elements in *iterable*. + + If *r* is not specified or is ``None``, then *r* defaults to the length of + *iterable*. + + >>> random_permutation(range(5)) # doctest:+SKIP + (3, 4, 0, 1, 2) + + This equivalent to taking a random selection from + ``itertools.permutations(iterable, r)``. + + """ + pool = tuple(iterable) + r = len(pool) if r is None else r + return tuple(sample(pool, r)) + + +def random_combination(iterable, r): + """Return a random *r* length subsequence of the elements in *iterable*. + + >>> random_combination(range(5), 3) # doctest:+SKIP + (2, 3, 4) + + This equivalent to taking a random selection from + ``itertools.combinations(iterable, r)``. + + """ + pool = tuple(iterable) + n = len(pool) + indices = sorted(sample(range(n), r)) + return tuple(pool[i] for i in indices) + + +def random_combination_with_replacement(iterable, r): + """Return a random *r* length subsequence of elements in *iterable*, + allowing individual elements to be repeated. + + >>> random_combination_with_replacement(range(3), 5) # doctest:+SKIP + (0, 0, 1, 2, 2) + + This equivalent to taking a random selection from + ``itertools.combinations_with_replacement(iterable, r)``. + + """ + pool = tuple(iterable) + n = len(pool) + indices = sorted(randrange(n) for i in range(r)) + return tuple(pool[i] for i in indices) + + +def nth_combination(iterable, r, index): + """Equivalent to ``list(combinations(iterable, r))[index]``. + + The subsequences of *iterable* that are of length *r* can be ordered + lexicographically. :func:`nth_combination` computes the subsequence at + sort position *index* directly, without computing the previous + subsequences. + + >>> nth_combination(range(5), 3, 5) + (0, 3, 4) + + ``ValueError`` will be raised If *r* is negative or greater than the length + of *iterable*. + ``IndexError`` will be raised if the given *index* is invalid. + """ + pool = tuple(iterable) + n = len(pool) + if (r < 0) or (r > n): + raise ValueError + + c = 1 + k = min(r, n - r) + for i in range(1, k + 1): + c = c * (n - k + i) // i + + if index < 0: + index += c + + if (index < 0) or (index >= c): + raise IndexError + + result = [] + while r: + c, n, r = c * r // n, n - 1, r - 1 + while index >= c: + index -= c + c, n = c * (n - r) // n, n - 1 + result.append(pool[-1 - n]) + + return tuple(result) + + +def prepend(value, iterator): + """Yield *value*, followed by the elements in *iterator*. + + >>> value = '0' + >>> iterator = ['1', '2', '3'] + >>> list(prepend(value, iterator)) + ['0', '1', '2', '3'] + + To prepend multiple values, see :func:`itertools.chain` + or :func:`value_chain`. + + """ + return chain([value], iterator) + + +def convolve(signal, kernel): + """Convolve the iterable *signal* with the iterable *kernel*. + + >>> signal = (1, 2, 3, 4, 5) + >>> kernel = [3, 2, 1] + >>> list(convolve(signal, kernel)) + [3, 8, 14, 20, 26, 14, 5] + + Note: the input arguments are not interchangeable, as the *kernel* + is immediately consumed and stored. + + """ + kernel = tuple(kernel)[::-1] + n = len(kernel) + window = deque([0], maxlen=n) * n + for x in chain(signal, repeat(0, n - 1)): + window.append(x) + yield sum(map(operator.mul, kernel, window)) + + +def before_and_after(predicate, it): + """A variant of :func:`takewhile` that allows complete access to the + remainder of the iterator. + + >>> it = iter('ABCdEfGhI') + >>> all_upper, remainder = before_and_after(str.isupper, it) + >>> ''.join(all_upper) + 'ABC' + >>> ''.join(remainder) # takewhile() would lose the 'd' + 'dEfGhI' + + Note that the first iterator must be fully consumed before the second + iterator can generate valid results. + """ + it = iter(it) + transition = [] + + def true_iterator(): + for elem in it: + if predicate(elem): + yield elem + else: + transition.append(elem) + return + + # Note: this is different from itertools recipes to allow nesting + # before_and_after remainders into before_and_after again. See tests + # for an example. + remainder_iterator = chain(transition, it) + + return true_iterator(), remainder_iterator + + +def triplewise(iterable): + """Return overlapping triplets from *iterable*. + + >>> list(triplewise('ABCDE')) + [('A', 'B', 'C'), ('B', 'C', 'D'), ('C', 'D', 'E')] + + """ + for (a, _), (b, c) in pairwise(pairwise(iterable)): + yield a, b, c + + +def sliding_window(iterable, n): + """Return a sliding window of width *n* over *iterable*. + + >>> list(sliding_window(range(6), 4)) + [(0, 1, 2, 3), (1, 2, 3, 4), (2, 3, 4, 5)] + + If *iterable* has fewer than *n* items, then nothing is yielded: + + >>> list(sliding_window(range(3), 4)) + [] + + For a variant with more features, see :func:`windowed`. + """ + it = iter(iterable) + window = deque(islice(it, n), maxlen=n) + if len(window) == n: + yield tuple(window) + for x in it: + window.append(x) + yield tuple(window) + + +def subslices(iterable): + """Return all contiguous non-empty subslices of *iterable*. + + >>> list(subslices('ABC')) + [['A'], ['A', 'B'], ['A', 'B', 'C'], ['B'], ['B', 'C'], ['C']] + + This is similar to :func:`substrings`, but emits items in a different + order. + """ + seq = list(iterable) + slices = starmap(slice, combinations(range(len(seq) + 1), 2)) + return map(operator.getitem, repeat(seq), slices) + + +def polynomial_from_roots(roots): + """Compute a polynomial's coefficients from its roots. + + >>> roots = [5, -4, 3] # (x - 5) * (x + 4) * (x - 3) + >>> polynomial_from_roots(roots) # x^3 - 4 * x^2 - 17 * x + 60 + [1, -4, -17, 60] + """ + # Use math.prod for Python 3.8+, + prod = getattr(math, 'prod', lambda x: reduce(operator.mul, x, 1)) + roots = list(map(operator.neg, roots)) + return [ + sum(map(prod, combinations(roots, k))) for k in range(len(roots) + 1) + ] + + +def iter_index(iterable, value, start=0): + """Yield the index of each place in *iterable* that *value* occurs, + beginning with index *start*. + + See :func:`locate` for a more general means of finding the indexes + associated with particular values. + + >>> list(iter_index('AABCADEAF', 'A')) + [0, 1, 4, 7] + """ + try: + seq_index = iterable.index + except AttributeError: + # Slow path for general iterables + it = islice(iterable, start, None) + for i, element in enumerate(it, start): + if element is value or element == value: + yield i + else: + # Fast path for sequences + i = start - 1 + try: + while True: + i = seq_index(value, i + 1) + yield i + except ValueError: + pass + + +def sieve(n): + """Yield the primes less than n. + + >>> list(sieve(30)) + [2, 3, 5, 7, 11, 13, 17, 19, 23, 29] + """ + isqrt = getattr(math, 'isqrt', lambda x: int(math.sqrt(x))) + data = bytearray((0, 1)) * (n // 2) + data[:3] = 0, 0, 0 + limit = isqrt(n) + 1 + for p in compress(range(limit), data): + data[p * p : n : p + p] = bytes(len(range(p * p, n, p + p))) + data[2] = 1 + return iter_index(data, 1) if n > 2 else iter([]) + + +def batched(iterable, n): + """Batch data into lists of length *n*. The last batch may be shorter. + + >>> list(batched('ABCDEFG', 3)) + [['A', 'B', 'C'], ['D', 'E', 'F'], ['G']] + + This recipe is from the ``itertools`` docs. This library also provides + :func:`chunked`, which has a different implementation. + """ + if hexversion >= 0x30C00A0: # Python 3.12.0a0 + warnings.warn( + ( + 'batched will be removed in a future version of ' + 'more-itertools. Use the standard library ' + 'itertools.batched function instead' + ), + DeprecationWarning, + ) + + it = iter(iterable) + while True: + batch = list(islice(it, n)) + if not batch: + break + yield batch + + +def transpose(it): + """Swap the rows and columns of the input. + + >>> list(transpose([(1, 2, 3), (11, 22, 33)])) + [(1, 11), (2, 22), (3, 33)] + + The caller should ensure that the dimensions of the input are compatible. + """ + # TODO: when 3.9 goes end-of-life, add stric=True to this. + return zip(*it) + + +def matmul(m1, m2): + """Multiply two matrices. + >>> list(matmul([(7, 5), (3, 5)], [(2, 5), (7, 9)])) + [[49, 80], [41, 60]] + + The caller should ensure that the dimensions of the input matrices are + compatible with each other. + """ + n = len(m2[0]) + return batched(starmap(dotproduct, product(m1, transpose(m2))), n) + + +def factor(n): + """Yield the prime factors of n. + >>> list(factor(360)) + [2, 2, 2, 3, 3, 5] + """ + isqrt = getattr(math, 'isqrt', lambda x: int(math.sqrt(x))) + for prime in sieve(isqrt(n) + 1): + while True: + quotient, remainder = divmod(n, prime) + if remainder: + break + yield prime + n = quotient + if n == 1: + return + if n >= 2: + yield n diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/__init__.py new file mode 100644 index 0000000000..13cadc7f04 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/__init__.py @@ -0,0 +1,15 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. + +__title__ = "packaging" +__summary__ = "Core utilities for Python packages" +__uri__ = "https://github.com/pypa/packaging" + +__version__ = "23.1" + +__author__ = "Donald Stufft and individual contributors" +__email__ = "donald@stufft.io" + +__license__ = "BSD-2-Clause or Apache-2.0" +__copyright__ = "2014-2019 %s" % __author__ diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/_elffile.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_elffile.py new file mode 100644 index 0000000000..6fb19b30bb --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_elffile.py @@ -0,0 +1,108 @@ +""" +ELF file parser. + +This provides a class ``ELFFile`` that parses an ELF executable in a similar +interface to ``ZipFile``. Only the read interface is implemented. + +Based on: https://gist.github.com/lyssdod/f51579ae8d93c8657a5564aefc2ffbca +ELF header: https://refspecs.linuxfoundation.org/elf/gabi4+/ch4.eheader.html +""" + +import enum +import os +import struct +from typing import IO, Optional, Tuple + + +class ELFInvalid(ValueError): + pass + + +class EIClass(enum.IntEnum): + C32 = 1 + C64 = 2 + + +class EIData(enum.IntEnum): + Lsb = 1 + Msb = 2 + + +class EMachine(enum.IntEnum): + I386 = 3 + S390 = 22 + Arm = 40 + X8664 = 62 + AArc64 = 183 + + +class ELFFile: + """ + Representation of an ELF executable. + """ + + def __init__(self, f: IO[bytes]) -> None: + self._f = f + + try: + ident = self._read("16B") + except struct.error: + raise ELFInvalid("unable to parse identification") + magic = bytes(ident[:4]) + if magic != b"\x7fELF": + raise ELFInvalid(f"invalid magic: {magic!r}") + + self.capacity = ident[4] # Format for program header (bitness). + self.encoding = ident[5] # Data structure encoding (endianness). + + try: + # e_fmt: Format for program header. + # p_fmt: Format for section header. + # p_idx: Indexes to find p_type, p_offset, and p_filesz. + e_fmt, self._p_fmt, self._p_idx = { + (1, 1): ("<HHIIIIIHHH", "<IIIIIIII", (0, 1, 4)), # 32-bit LSB. + (1, 2): (">HHIIIIIHHH", ">IIIIIIII", (0, 1, 4)), # 32-bit MSB. + (2, 1): ("<HHIQQQIHHH", "<IIQQQQQQ", (0, 2, 5)), # 64-bit LSB. + (2, 2): (">HHIQQQIHHH", ">IIQQQQQQ", (0, 2, 5)), # 64-bit MSB. + }[(self.capacity, self.encoding)] + except KeyError: + raise ELFInvalid( + f"unrecognized capacity ({self.capacity}) or " + f"encoding ({self.encoding})" + ) + + try: + ( + _, + self.machine, # Architecture type. + _, + _, + self._e_phoff, # Offset of program header. + _, + self.flags, # Processor-specific flags. + _, + self._e_phentsize, # Size of section. + self._e_phnum, # Number of sections. + ) = self._read(e_fmt) + except struct.error as e: + raise ELFInvalid("unable to parse machine and section information") from e + + def _read(self, fmt: str) -> Tuple[int, ...]: + return struct.unpack(fmt, self._f.read(struct.calcsize(fmt))) + + @property + def interpreter(self) -> Optional[str]: + """ + The path recorded in the ``PT_INTERP`` section header. + """ + for index in range(self._e_phnum): + self._f.seek(self._e_phoff + self._e_phentsize * index) + try: + data = self._read(self._p_fmt) + except struct.error: + continue + if data[self._p_idx[0]] != 3: # Not PT_INTERP. + continue + self._f.seek(data[self._p_idx[1]]) + return os.fsdecode(self._f.read(data[self._p_idx[2]])).strip("\0") + return None diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/_manylinux.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_manylinux.py new file mode 100644 index 0000000000..449c655be6 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_manylinux.py @@ -0,0 +1,240 @@ +import collections +import contextlib +import functools +import os +import re +import sys +import warnings +from typing import Dict, Generator, Iterator, NamedTuple, Optional, Tuple + +from ._elffile import EIClass, EIData, ELFFile, EMachine + +EF_ARM_ABIMASK = 0xFF000000 +EF_ARM_ABI_VER5 = 0x05000000 +EF_ARM_ABI_FLOAT_HARD = 0x00000400 + + +# `os.PathLike` not a generic type until Python 3.9, so sticking with `str` +# as the type for `path` until then. +@contextlib.contextmanager +def _parse_elf(path: str) -> Generator[Optional[ELFFile], None, None]: + try: + with open(path, "rb") as f: + yield ELFFile(f) + except (OSError, TypeError, ValueError): + yield None + + +def _is_linux_armhf(executable: str) -> bool: + # hard-float ABI can be detected from the ELF header of the running + # process + # https://static.docs.arm.com/ihi0044/g/aaelf32.pdf + with _parse_elf(executable) as f: + return ( + f is not None + and f.capacity == EIClass.C32 + and f.encoding == EIData.Lsb + and f.machine == EMachine.Arm + and f.flags & EF_ARM_ABIMASK == EF_ARM_ABI_VER5 + and f.flags & EF_ARM_ABI_FLOAT_HARD == EF_ARM_ABI_FLOAT_HARD + ) + + +def _is_linux_i686(executable: str) -> bool: + with _parse_elf(executable) as f: + return ( + f is not None + and f.capacity == EIClass.C32 + and f.encoding == EIData.Lsb + and f.machine == EMachine.I386 + ) + + +def _have_compatible_abi(executable: str, arch: str) -> bool: + if arch == "armv7l": + return _is_linux_armhf(executable) + if arch == "i686": + return _is_linux_i686(executable) + return arch in {"x86_64", "aarch64", "ppc64", "ppc64le", "s390x"} + + +# If glibc ever changes its major version, we need to know what the last +# minor version was, so we can build the complete list of all versions. +# For now, guess what the highest minor version might be, assume it will +# be 50 for testing. Once this actually happens, update the dictionary +# with the actual value. +_LAST_GLIBC_MINOR: Dict[int, int] = collections.defaultdict(lambda: 50) + + +class _GLibCVersion(NamedTuple): + major: int + minor: int + + +def _glibc_version_string_confstr() -> Optional[str]: + """ + Primary implementation of glibc_version_string using os.confstr. + """ + # os.confstr is quite a bit faster than ctypes.DLL. It's also less likely + # to be broken or missing. This strategy is used in the standard library + # platform module. + # https://github.com/python/cpython/blob/fcf1d003bf4f0100c/Lib/platform.py#L175-L183 + try: + # Should be a string like "glibc 2.17". + version_string: str = getattr(os, "confstr")("CS_GNU_LIBC_VERSION") + assert version_string is not None + _, version = version_string.rsplit() + except (AssertionError, AttributeError, OSError, ValueError): + # os.confstr() or CS_GNU_LIBC_VERSION not available (or a bad value)... + return None + return version + + +def _glibc_version_string_ctypes() -> Optional[str]: + """ + Fallback implementation of glibc_version_string using ctypes. + """ + try: + import ctypes + except ImportError: + return None + + # ctypes.CDLL(None) internally calls dlopen(NULL), and as the dlopen + # manpage says, "If filename is NULL, then the returned handle is for the + # main program". This way we can let the linker do the work to figure out + # which libc our process is actually using. + # + # We must also handle the special case where the executable is not a + # dynamically linked executable. This can occur when using musl libc, + # for example. In this situation, dlopen() will error, leading to an + # OSError. Interestingly, at least in the case of musl, there is no + # errno set on the OSError. The single string argument used to construct + # OSError comes from libc itself and is therefore not portable to + # hard code here. In any case, failure to call dlopen() means we + # can proceed, so we bail on our attempt. + try: + process_namespace = ctypes.CDLL(None) + except OSError: + return None + + try: + gnu_get_libc_version = process_namespace.gnu_get_libc_version + except AttributeError: + # Symbol doesn't exist -> therefore, we are not linked to + # glibc. + return None + + # Call gnu_get_libc_version, which returns a string like "2.5" + gnu_get_libc_version.restype = ctypes.c_char_p + version_str: str = gnu_get_libc_version() + # py2 / py3 compatibility: + if not isinstance(version_str, str): + version_str = version_str.decode("ascii") + + return version_str + + +def _glibc_version_string() -> Optional[str]: + """Returns glibc version string, or None if not using glibc.""" + return _glibc_version_string_confstr() or _glibc_version_string_ctypes() + + +def _parse_glibc_version(version_str: str) -> Tuple[int, int]: + """Parse glibc version. + + We use a regexp instead of str.split because we want to discard any + random junk that might come after the minor version -- this might happen + in patched/forked versions of glibc (e.g. Linaro's version of glibc + uses version strings like "2.20-2014.11"). See gh-3588. + """ + m = re.match(r"(?P<major>[0-9]+)\.(?P<minor>[0-9]+)", version_str) + if not m: + warnings.warn( + f"Expected glibc version with 2 components major.minor," + f" got: {version_str}", + RuntimeWarning, + ) + return -1, -1 + return int(m.group("major")), int(m.group("minor")) + + +@functools.lru_cache() +def _get_glibc_version() -> Tuple[int, int]: + version_str = _glibc_version_string() + if version_str is None: + return (-1, -1) + return _parse_glibc_version(version_str) + + +# From PEP 513, PEP 600 +def _is_compatible(name: str, arch: str, version: _GLibCVersion) -> bool: + sys_glibc = _get_glibc_version() + if sys_glibc < version: + return False + # Check for presence of _manylinux module. + try: + import _manylinux # noqa + except ImportError: + return True + if hasattr(_manylinux, "manylinux_compatible"): + result = _manylinux.manylinux_compatible(version[0], version[1], arch) + if result is not None: + return bool(result) + return True + if version == _GLibCVersion(2, 5): + if hasattr(_manylinux, "manylinux1_compatible"): + return bool(_manylinux.manylinux1_compatible) + if version == _GLibCVersion(2, 12): + if hasattr(_manylinux, "manylinux2010_compatible"): + return bool(_manylinux.manylinux2010_compatible) + if version == _GLibCVersion(2, 17): + if hasattr(_manylinux, "manylinux2014_compatible"): + return bool(_manylinux.manylinux2014_compatible) + return True + + +_LEGACY_MANYLINUX_MAP = { + # CentOS 7 w/ glibc 2.17 (PEP 599) + (2, 17): "manylinux2014", + # CentOS 6 w/ glibc 2.12 (PEP 571) + (2, 12): "manylinux2010", + # CentOS 5 w/ glibc 2.5 (PEP 513) + (2, 5): "manylinux1", +} + + +def platform_tags(linux: str, arch: str) -> Iterator[str]: + if not _have_compatible_abi(sys.executable, arch): + return + # Oldest glibc to be supported regardless of architecture is (2, 17). + too_old_glibc2 = _GLibCVersion(2, 16) + if arch in {"x86_64", "i686"}: + # On x86/i686 also oldest glibc to be supported is (2, 5). + too_old_glibc2 = _GLibCVersion(2, 4) + current_glibc = _GLibCVersion(*_get_glibc_version()) + glibc_max_list = [current_glibc] + # We can assume compatibility across glibc major versions. + # https://sourceware.org/bugzilla/show_bug.cgi?id=24636 + # + # Build a list of maximum glibc versions so that we can + # output the canonical list of all glibc from current_glibc + # down to too_old_glibc2, including all intermediary versions. + for glibc_major in range(current_glibc.major - 1, 1, -1): + glibc_minor = _LAST_GLIBC_MINOR[glibc_major] + glibc_max_list.append(_GLibCVersion(glibc_major, glibc_minor)) + for glibc_max in glibc_max_list: + if glibc_max.major == too_old_glibc2.major: + min_minor = too_old_glibc2.minor + else: + # For other glibc major versions oldest supported is (x, 0). + min_minor = -1 + for glibc_minor in range(glibc_max.minor, min_minor, -1): + glibc_version = _GLibCVersion(glibc_max.major, glibc_minor) + tag = "manylinux_{}_{}".format(*glibc_version) + if _is_compatible(tag, arch, glibc_version): + yield linux.replace("linux", tag) + # Handle the legacy manylinux1, manylinux2010, manylinux2014 tags. + if glibc_version in _LEGACY_MANYLINUX_MAP: + legacy_tag = _LEGACY_MANYLINUX_MAP[glibc_version] + if _is_compatible(legacy_tag, arch, glibc_version): + yield linux.replace("linux", legacy_tag) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/_musllinux.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_musllinux.py new file mode 100644 index 0000000000..706ba600a9 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_musllinux.py @@ -0,0 +1,80 @@ +"""PEP 656 support. + +This module implements logic to detect if the currently running Python is +linked against musl, and what musl version is used. +""" + +import functools +import re +import subprocess +import sys +from typing import Iterator, NamedTuple, Optional + +from ._elffile import ELFFile + + +class _MuslVersion(NamedTuple): + major: int + minor: int + + +def _parse_musl_version(output: str) -> Optional[_MuslVersion]: + lines = [n for n in (n.strip() for n in output.splitlines()) if n] + if len(lines) < 2 or lines[0][:4] != "musl": + return None + m = re.match(r"Version (\d+)\.(\d+)", lines[1]) + if not m: + return None + return _MuslVersion(major=int(m.group(1)), minor=int(m.group(2))) + + +@functools.lru_cache() +def _get_musl_version(executable: str) -> Optional[_MuslVersion]: + """Detect currently-running musl runtime version. + + This is done by checking the specified executable's dynamic linking + information, and invoking the loader to parse its output for a version + string. If the loader is musl, the output would be something like:: + + musl libc (x86_64) + Version 1.2.2 + Dynamic Program Loader + """ + try: + with open(executable, "rb") as f: + ld = ELFFile(f).interpreter + except (OSError, TypeError, ValueError): + return None + if ld is None or "musl" not in ld: + return None + proc = subprocess.run([ld], stderr=subprocess.PIPE, universal_newlines=True) + return _parse_musl_version(proc.stderr) + + +def platform_tags(arch: str) -> Iterator[str]: + """Generate musllinux tags compatible to the current platform. + + :param arch: Should be the part of platform tag after the ``linux_`` + prefix, e.g. ``x86_64``. The ``linux_`` prefix is assumed as a + prerequisite for the current platform to be musllinux-compatible. + + :returns: An iterator of compatible musllinux tags. + """ + sys_musl = _get_musl_version(sys.executable) + if sys_musl is None: # Python not dynamically linked against musl. + return + for minor in range(sys_musl.minor, -1, -1): + yield f"musllinux_{sys_musl.major}_{minor}_{arch}" + + +if __name__ == "__main__": # pragma: no cover + import sysconfig + + plat = sysconfig.get_platform() + assert plat.startswith("linux-"), "not linux" + + print("plat:", plat) + print("musl:", _get_musl_version(sys.executable)) + print("tags:", end=" ") + for t in platform_tags(re.sub(r"[.-]", "_", plat.split("-", 1)[-1])): + print(t, end="\n ") diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/_parser.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_parser.py new file mode 100644 index 0000000000..5a18b758fe --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_parser.py @@ -0,0 +1,353 @@ +"""Handwritten parser of dependency specifiers. + +The docstring for each __parse_* function contains ENBF-inspired grammar representing +the implementation. +""" + +import ast +from typing import Any, List, NamedTuple, Optional, Tuple, Union + +from ._tokenizer import DEFAULT_RULES, Tokenizer + + +class Node: + def __init__(self, value: str) -> None: + self.value = value + + def __str__(self) -> str: + return self.value + + def __repr__(self) -> str: + return f"<{self.__class__.__name__}('{self}')>" + + def serialize(self) -> str: + raise NotImplementedError + + +class Variable(Node): + def serialize(self) -> str: + return str(self) + + +class Value(Node): + def serialize(self) -> str: + return f'"{self}"' + + +class Op(Node): + def serialize(self) -> str: + return str(self) + + +MarkerVar = Union[Variable, Value] +MarkerItem = Tuple[MarkerVar, Op, MarkerVar] +# MarkerAtom = Union[MarkerItem, List["MarkerAtom"]] +# MarkerList = List[Union["MarkerList", MarkerAtom, str]] +# mypy does not support recursive type definition +# https://github.com/python/mypy/issues/731 +MarkerAtom = Any +MarkerList = List[Any] + + +class ParsedRequirement(NamedTuple): + name: str + url: str + extras: List[str] + specifier: str + marker: Optional[MarkerList] + + +# -------------------------------------------------------------------------------------- +# Recursive descent parser for dependency specifier +# -------------------------------------------------------------------------------------- +def parse_requirement(source: str) -> ParsedRequirement: + return _parse_requirement(Tokenizer(source, rules=DEFAULT_RULES)) + + +def _parse_requirement(tokenizer: Tokenizer) -> ParsedRequirement: + """ + requirement = WS? IDENTIFIER WS? extras WS? requirement_details + """ + tokenizer.consume("WS") + + name_token = tokenizer.expect( + "IDENTIFIER", expected="package name at the start of dependency specifier" + ) + name = name_token.text + tokenizer.consume("WS") + + extras = _parse_extras(tokenizer) + tokenizer.consume("WS") + + url, specifier, marker = _parse_requirement_details(tokenizer) + tokenizer.expect("END", expected="end of dependency specifier") + + return ParsedRequirement(name, url, extras, specifier, marker) + + +def _parse_requirement_details( + tokenizer: Tokenizer, +) -> Tuple[str, str, Optional[MarkerList]]: + """ + requirement_details = AT URL (WS requirement_marker?)? + | specifier WS? (requirement_marker)? + """ + + specifier = "" + url = "" + marker = None + + if tokenizer.check("AT"): + tokenizer.read() + tokenizer.consume("WS") + + url_start = tokenizer.position + url = tokenizer.expect("URL", expected="URL after @").text + if tokenizer.check("END", peek=True): + return (url, specifier, marker) + + tokenizer.expect("WS", expected="whitespace after URL") + + # The input might end after whitespace. + if tokenizer.check("END", peek=True): + return (url, specifier, marker) + + marker = _parse_requirement_marker( + tokenizer, span_start=url_start, after="URL and whitespace" + ) + else: + specifier_start = tokenizer.position + specifier = _parse_specifier(tokenizer) + tokenizer.consume("WS") + + if tokenizer.check("END", peek=True): + return (url, specifier, marker) + + marker = _parse_requirement_marker( + tokenizer, + span_start=specifier_start, + after=( + "version specifier" + if specifier + else "name and no valid version specifier" + ), + ) + + return (url, specifier, marker) + + +def _parse_requirement_marker( + tokenizer: Tokenizer, *, span_start: int, after: str +) -> MarkerList: + """ + requirement_marker = SEMICOLON marker WS? + """ + + if not tokenizer.check("SEMICOLON"): + tokenizer.raise_syntax_error( + f"Expected end or semicolon (after {after})", + span_start=span_start, + ) + tokenizer.read() + + marker = _parse_marker(tokenizer) + tokenizer.consume("WS") + + return marker + + +def _parse_extras(tokenizer: Tokenizer) -> List[str]: + """ + extras = (LEFT_BRACKET wsp* extras_list? wsp* RIGHT_BRACKET)? + """ + if not tokenizer.check("LEFT_BRACKET", peek=True): + return [] + + with tokenizer.enclosing_tokens( + "LEFT_BRACKET", + "RIGHT_BRACKET", + around="extras", + ): + tokenizer.consume("WS") + extras = _parse_extras_list(tokenizer) + tokenizer.consume("WS") + + return extras + + +def _parse_extras_list(tokenizer: Tokenizer) -> List[str]: + """ + extras_list = identifier (wsp* ',' wsp* identifier)* + """ + extras: List[str] = [] + + if not tokenizer.check("IDENTIFIER"): + return extras + + extras.append(tokenizer.read().text) + + while True: + tokenizer.consume("WS") + if tokenizer.check("IDENTIFIER", peek=True): + tokenizer.raise_syntax_error("Expected comma between extra names") + elif not tokenizer.check("COMMA"): + break + + tokenizer.read() + tokenizer.consume("WS") + + extra_token = tokenizer.expect("IDENTIFIER", expected="extra name after comma") + extras.append(extra_token.text) + + return extras + + +def _parse_specifier(tokenizer: Tokenizer) -> str: + """ + specifier = LEFT_PARENTHESIS WS? version_many WS? RIGHT_PARENTHESIS + | WS? version_many WS? + """ + with tokenizer.enclosing_tokens( + "LEFT_PARENTHESIS", + "RIGHT_PARENTHESIS", + around="version specifier", + ): + tokenizer.consume("WS") + parsed_specifiers = _parse_version_many(tokenizer) + tokenizer.consume("WS") + + return parsed_specifiers + + +def _parse_version_many(tokenizer: Tokenizer) -> str: + """ + version_many = (SPECIFIER (WS? COMMA WS? SPECIFIER)*)? + """ + parsed_specifiers = "" + while tokenizer.check("SPECIFIER"): + span_start = tokenizer.position + parsed_specifiers += tokenizer.read().text + if tokenizer.check("VERSION_PREFIX_TRAIL", peek=True): + tokenizer.raise_syntax_error( + ".* suffix can only be used with `==` or `!=` operators", + span_start=span_start, + span_end=tokenizer.position + 1, + ) + if tokenizer.check("VERSION_LOCAL_LABEL_TRAIL", peek=True): + tokenizer.raise_syntax_error( + "Local version label can only be used with `==` or `!=` operators", + span_start=span_start, + span_end=tokenizer.position, + ) + tokenizer.consume("WS") + if not tokenizer.check("COMMA"): + break + parsed_specifiers += tokenizer.read().text + tokenizer.consume("WS") + + return parsed_specifiers + + +# -------------------------------------------------------------------------------------- +# Recursive descent parser for marker expression +# -------------------------------------------------------------------------------------- +def parse_marker(source: str) -> MarkerList: + return _parse_marker(Tokenizer(source, rules=DEFAULT_RULES)) + + +def _parse_marker(tokenizer: Tokenizer) -> MarkerList: + """ + marker = marker_atom (BOOLOP marker_atom)+ + """ + expression = [_parse_marker_atom(tokenizer)] + while tokenizer.check("BOOLOP"): + token = tokenizer.read() + expr_right = _parse_marker_atom(tokenizer) + expression.extend((token.text, expr_right)) + return expression + + +def _parse_marker_atom(tokenizer: Tokenizer) -> MarkerAtom: + """ + marker_atom = WS? LEFT_PARENTHESIS WS? marker WS? RIGHT_PARENTHESIS WS? + | WS? marker_item WS? + """ + + tokenizer.consume("WS") + if tokenizer.check("LEFT_PARENTHESIS", peek=True): + with tokenizer.enclosing_tokens( + "LEFT_PARENTHESIS", + "RIGHT_PARENTHESIS", + around="marker expression", + ): + tokenizer.consume("WS") + marker: MarkerAtom = _parse_marker(tokenizer) + tokenizer.consume("WS") + else: + marker = _parse_marker_item(tokenizer) + tokenizer.consume("WS") + return marker + + +def _parse_marker_item(tokenizer: Tokenizer) -> MarkerItem: + """ + marker_item = WS? marker_var WS? marker_op WS? marker_var WS? + """ + tokenizer.consume("WS") + marker_var_left = _parse_marker_var(tokenizer) + tokenizer.consume("WS") + marker_op = _parse_marker_op(tokenizer) + tokenizer.consume("WS") + marker_var_right = _parse_marker_var(tokenizer) + tokenizer.consume("WS") + return (marker_var_left, marker_op, marker_var_right) + + +def _parse_marker_var(tokenizer: Tokenizer) -> MarkerVar: + """ + marker_var = VARIABLE | QUOTED_STRING + """ + if tokenizer.check("VARIABLE"): + return process_env_var(tokenizer.read().text.replace(".", "_")) + elif tokenizer.check("QUOTED_STRING"): + return process_python_str(tokenizer.read().text) + else: + tokenizer.raise_syntax_error( + message="Expected a marker variable or quoted string" + ) + + +def process_env_var(env_var: str) -> Variable: + if ( + env_var == "platform_python_implementation" + or env_var == "python_implementation" + ): + return Variable("platform_python_implementation") + else: + return Variable(env_var) + + +def process_python_str(python_str: str) -> Value: + value = ast.literal_eval(python_str) + return Value(str(value)) + + +def _parse_marker_op(tokenizer: Tokenizer) -> Op: + """ + marker_op = IN | NOT IN | OP + """ + if tokenizer.check("IN"): + tokenizer.read() + return Op("in") + elif tokenizer.check("NOT"): + tokenizer.read() + tokenizer.expect("WS", expected="whitespace after 'not'") + tokenizer.expect("IN", expected="'in' after 'not'") + return Op("not in") + elif tokenizer.check("OP"): + return Op(tokenizer.read().text) + else: + return tokenizer.raise_syntax_error( + "Expected marker operator, one of " + "<=, <, !=, ==, >=, >, ~=, ===, in, not in" + ) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/_structures.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_structures.py new file mode 100644 index 0000000000..90a6465f96 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_structures.py @@ -0,0 +1,61 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. + + +class InfinityType: + def __repr__(self) -> str: + return "Infinity" + + def __hash__(self) -> int: + return hash(repr(self)) + + def __lt__(self, other: object) -> bool: + return False + + def __le__(self, other: object) -> bool: + return False + + def __eq__(self, other: object) -> bool: + return isinstance(other, self.__class__) + + def __gt__(self, other: object) -> bool: + return True + + def __ge__(self, other: object) -> bool: + return True + + def __neg__(self: object) -> "NegativeInfinityType": + return NegativeInfinity + + +Infinity = InfinityType() + + +class NegativeInfinityType: + def __repr__(self) -> str: + return "-Infinity" + + def __hash__(self) -> int: + return hash(repr(self)) + + def __lt__(self, other: object) -> bool: + return True + + def __le__(self, other: object) -> bool: + return True + + def __eq__(self, other: object) -> bool: + return isinstance(other, self.__class__) + + def __gt__(self, other: object) -> bool: + return False + + def __ge__(self, other: object) -> bool: + return False + + def __neg__(self: object) -> InfinityType: + return Infinity + + +NegativeInfinity = NegativeInfinityType() diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/_tokenizer.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_tokenizer.py new file mode 100644 index 0000000000..dd0d648d49 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/_tokenizer.py @@ -0,0 +1,192 @@ +import contextlib +import re +from dataclasses import dataclass +from typing import Dict, Iterator, NoReturn, Optional, Tuple, Union + +from .specifiers import Specifier + + +@dataclass +class Token: + name: str + text: str + position: int + + +class ParserSyntaxError(Exception): + """The provided source text could not be parsed correctly.""" + + def __init__( + self, + message: str, + *, + source: str, + span: Tuple[int, int], + ) -> None: + self.span = span + self.message = message + self.source = source + + super().__init__() + + def __str__(self) -> str: + marker = " " * self.span[0] + "~" * (self.span[1] - self.span[0]) + "^" + return "\n ".join([self.message, self.source, marker]) + + +DEFAULT_RULES: "Dict[str, Union[str, re.Pattern[str]]]" = { + "LEFT_PARENTHESIS": r"\(", + "RIGHT_PARENTHESIS": r"\)", + "LEFT_BRACKET": r"\[", + "RIGHT_BRACKET": r"\]", + "SEMICOLON": r";", + "COMMA": r",", + "QUOTED_STRING": re.compile( + r""" + ( + ('[^']*') + | + ("[^"]*") + ) + """, + re.VERBOSE, + ), + "OP": r"(===|==|~=|!=|<=|>=|<|>)", + "BOOLOP": r"\b(or|and)\b", + "IN": r"\bin\b", + "NOT": r"\bnot\b", + "VARIABLE": re.compile( + r""" + \b( + python_version + |python_full_version + |os[._]name + |sys[._]platform + |platform_(release|system) + |platform[._](version|machine|python_implementation) + |python_implementation + |implementation_(name|version) + |extra + )\b + """, + re.VERBOSE, + ), + "SPECIFIER": re.compile( + Specifier._operator_regex_str + Specifier._version_regex_str, + re.VERBOSE | re.IGNORECASE, + ), + "AT": r"\@", + "URL": r"[^ \t]+", + "IDENTIFIER": r"\b[a-zA-Z0-9][a-zA-Z0-9._-]*\b", + "VERSION_PREFIX_TRAIL": r"\.\*", + "VERSION_LOCAL_LABEL_TRAIL": r"\+[a-z0-9]+(?:[-_\.][a-z0-9]+)*", + "WS": r"[ \t]+", + "END": r"$", +} + + +class Tokenizer: + """Context-sensitive token parsing. + + Provides methods to examine the input stream to check whether the next token + matches. + """ + + def __init__( + self, + source: str, + *, + rules: "Dict[str, Union[str, re.Pattern[str]]]", + ) -> None: + self.source = source + self.rules: Dict[str, re.Pattern[str]] = { + name: re.compile(pattern) for name, pattern in rules.items() + } + self.next_token: Optional[Token] = None + self.position = 0 + + def consume(self, name: str) -> None: + """Move beyond provided token name, if at current position.""" + if self.check(name): + self.read() + + def check(self, name: str, *, peek: bool = False) -> bool: + """Check whether the next token has the provided name. + + By default, if the check succeeds, the token *must* be read before + another check. If `peek` is set to `True`, the token is not loaded and + would need to be checked again. + """ + assert ( + self.next_token is None + ), f"Cannot check for {name!r}, already have {self.next_token!r}" + assert name in self.rules, f"Unknown token name: {name!r}" + + expression = self.rules[name] + + match = expression.match(self.source, self.position) + if match is None: + return False + if not peek: + self.next_token = Token(name, match[0], self.position) + return True + + def expect(self, name: str, *, expected: str) -> Token: + """Expect a certain token name next, failing with a syntax error otherwise. + + The token is *not* read. + """ + if not self.check(name): + raise self.raise_syntax_error(f"Expected {expected}") + return self.read() + + def read(self) -> Token: + """Consume the next token and return it.""" + token = self.next_token + assert token is not None + + self.position += len(token.text) + self.next_token = None + + return token + + def raise_syntax_error( + self, + message: str, + *, + span_start: Optional[int] = None, + span_end: Optional[int] = None, + ) -> NoReturn: + """Raise ParserSyntaxError at the given position.""" + span = ( + self.position if span_start is None else span_start, + self.position if span_end is None else span_end, + ) + raise ParserSyntaxError( + message, + source=self.source, + span=span, + ) + + @contextlib.contextmanager + def enclosing_tokens( + self, open_token: str, close_token: str, *, around: str + ) -> Iterator[None]: + if self.check(open_token): + open_position = self.position + self.read() + else: + open_position = None + + yield + + if open_position is None: + return + + if not self.check(close_token): + self.raise_syntax_error( + f"Expected matching {close_token} for {open_token}, after {around}", + span_start=open_position, + ) + + self.read() diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/markers.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/markers.py new file mode 100644 index 0000000000..8b98fca723 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/markers.py @@ -0,0 +1,252 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. + +import operator +import os +import platform +import sys +from typing import Any, Callable, Dict, List, Optional, Tuple, Union + +from ._parser import ( + MarkerAtom, + MarkerList, + Op, + Value, + Variable, + parse_marker as _parse_marker, +) +from ._tokenizer import ParserSyntaxError +from .specifiers import InvalidSpecifier, Specifier +from .utils import canonicalize_name + +__all__ = [ + "InvalidMarker", + "UndefinedComparison", + "UndefinedEnvironmentName", + "Marker", + "default_environment", +] + +Operator = Callable[[str, str], bool] + + +class InvalidMarker(ValueError): + """ + An invalid marker was found, users should refer to PEP 508. + """ + + +class UndefinedComparison(ValueError): + """ + An invalid operation was attempted on a value that doesn't support it. + """ + + +class UndefinedEnvironmentName(ValueError): + """ + A name was attempted to be used that does not exist inside of the + environment. + """ + + +def _normalize_extra_values(results: Any) -> Any: + """ + Normalize extra values. + """ + if isinstance(results[0], tuple): + lhs, op, rhs = results[0] + if isinstance(lhs, Variable) and lhs.value == "extra": + normalized_extra = canonicalize_name(rhs.value) + rhs = Value(normalized_extra) + elif isinstance(rhs, Variable) and rhs.value == "extra": + normalized_extra = canonicalize_name(lhs.value) + lhs = Value(normalized_extra) + results[0] = lhs, op, rhs + return results + + +def _format_marker( + marker: Union[List[str], MarkerAtom, str], first: Optional[bool] = True +) -> str: + + assert isinstance(marker, (list, tuple, str)) + + # Sometimes we have a structure like [[...]] which is a single item list + # where the single item is itself it's own list. In that case we want skip + # the rest of this function so that we don't get extraneous () on the + # outside. + if ( + isinstance(marker, list) + and len(marker) == 1 + and isinstance(marker[0], (list, tuple)) + ): + return _format_marker(marker[0]) + + if isinstance(marker, list): + inner = (_format_marker(m, first=False) for m in marker) + if first: + return " ".join(inner) + else: + return "(" + " ".join(inner) + ")" + elif isinstance(marker, tuple): + return " ".join([m.serialize() for m in marker]) + else: + return marker + + +_operators: Dict[str, Operator] = { + "in": lambda lhs, rhs: lhs in rhs, + "not in": lambda lhs, rhs: lhs not in rhs, + "<": operator.lt, + "<=": operator.le, + "==": operator.eq, + "!=": operator.ne, + ">=": operator.ge, + ">": operator.gt, +} + + +def _eval_op(lhs: str, op: Op, rhs: str) -> bool: + try: + spec = Specifier("".join([op.serialize(), rhs])) + except InvalidSpecifier: + pass + else: + return spec.contains(lhs, prereleases=True) + + oper: Optional[Operator] = _operators.get(op.serialize()) + if oper is None: + raise UndefinedComparison(f"Undefined {op!r} on {lhs!r} and {rhs!r}.") + + return oper(lhs, rhs) + + +def _normalize(*values: str, key: str) -> Tuple[str, ...]: + # PEP 685 – Comparison of extra names for optional distribution dependencies + # https://peps.python.org/pep-0685/ + # > When comparing extra names, tools MUST normalize the names being + # > compared using the semantics outlined in PEP 503 for names + if key == "extra": + return tuple(canonicalize_name(v) for v in values) + + # other environment markers don't have such standards + return values + + +def _evaluate_markers(markers: MarkerList, environment: Dict[str, str]) -> bool: + groups: List[List[bool]] = [[]] + + for marker in markers: + assert isinstance(marker, (list, tuple, str)) + + if isinstance(marker, list): + groups[-1].append(_evaluate_markers(marker, environment)) + elif isinstance(marker, tuple): + lhs, op, rhs = marker + + if isinstance(lhs, Variable): + environment_key = lhs.value + lhs_value = environment[environment_key] + rhs_value = rhs.value + else: + lhs_value = lhs.value + environment_key = rhs.value + rhs_value = environment[environment_key] + + lhs_value, rhs_value = _normalize(lhs_value, rhs_value, key=environment_key) + groups[-1].append(_eval_op(lhs_value, op, rhs_value)) + else: + assert marker in ["and", "or"] + if marker == "or": + groups.append([]) + + return any(all(item) for item in groups) + + +def format_full_version(info: "sys._version_info") -> str: + version = "{0.major}.{0.minor}.{0.micro}".format(info) + kind = info.releaselevel + if kind != "final": + version += kind[0] + str(info.serial) + return version + + +def default_environment() -> Dict[str, str]: + iver = format_full_version(sys.implementation.version) + implementation_name = sys.implementation.name + return { + "implementation_name": implementation_name, + "implementation_version": iver, + "os_name": os.name, + "platform_machine": platform.machine(), + "platform_release": platform.release(), + "platform_system": platform.system(), + "platform_version": platform.version(), + "python_full_version": platform.python_version(), + "platform_python_implementation": platform.python_implementation(), + "python_version": ".".join(platform.python_version_tuple()[:2]), + "sys_platform": sys.platform, + } + + +class Marker: + def __init__(self, marker: str) -> None: + # Note: We create a Marker object without calling this constructor in + # packaging.requirements.Requirement. If any additional logic is + # added here, make sure to mirror/adapt Requirement. + try: + self._markers = _normalize_extra_values(_parse_marker(marker)) + # The attribute `_markers` can be described in terms of a recursive type: + # MarkerList = List[Union[Tuple[Node, ...], str, MarkerList]] + # + # For example, the following expression: + # python_version > "3.6" or (python_version == "3.6" and os_name == "unix") + # + # is parsed into: + # [ + # (<Variable('python_version')>, <Op('>')>, <Value('3.6')>), + # 'and', + # [ + # (<Variable('python_version')>, <Op('==')>, <Value('3.6')>), + # 'or', + # (<Variable('os_name')>, <Op('==')>, <Value('unix')>) + # ] + # ] + except ParserSyntaxError as e: + raise InvalidMarker(str(e)) from e + + def __str__(self) -> str: + return _format_marker(self._markers) + + def __repr__(self) -> str: + return f"<Marker('{self}')>" + + def __hash__(self) -> int: + return hash((self.__class__.__name__, str(self))) + + def __eq__(self, other: Any) -> bool: + if not isinstance(other, Marker): + return NotImplemented + + return str(self) == str(other) + + def evaluate(self, environment: Optional[Dict[str, str]] = None) -> bool: + """Evaluate a marker. + + Return the boolean from evaluating the given marker against the + environment. environment is an optional argument to override all or + part of the determined environment. + + The environment is determined from the current Python process. + """ + current_environment = default_environment() + current_environment["extra"] = "" + if environment is not None: + current_environment.update(environment) + # The API used to allow setting extra to None. We need to handle this + # case for backwards compatibility. + if current_environment["extra"] is None: + current_environment["extra"] = "" + + return _evaluate_markers(self._markers, current_environment) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/metadata.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/metadata.py new file mode 100644 index 0000000000..e76a60c395 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/metadata.py @@ -0,0 +1,408 @@ +import email.feedparser +import email.header +import email.message +import email.parser +import email.policy +import sys +import typing +from typing import Dict, List, Optional, Tuple, Union, cast + +if sys.version_info >= (3, 8): # pragma: no cover + from typing import TypedDict +else: # pragma: no cover + if typing.TYPE_CHECKING: + from typing_extensions import TypedDict + else: + try: + from typing_extensions import TypedDict + except ImportError: + + class TypedDict: + def __init_subclass__(*_args, **_kwargs): + pass + + +# The RawMetadata class attempts to make as few assumptions about the underlying +# serialization formats as possible. The idea is that as long as a serialization +# formats offer some very basic primitives in *some* way then we can support +# serializing to and from that format. +class RawMetadata(TypedDict, total=False): + """A dictionary of raw core metadata. + + Each field in core metadata maps to a key of this dictionary (when data is + provided). The key is lower-case and underscores are used instead of dashes + compared to the equivalent core metadata field. Any core metadata field that + can be specified multiple times or can hold multiple values in a single + field have a key with a plural name. + + Core metadata fields that can be specified multiple times are stored as a + list or dict depending on which is appropriate for the field. Any fields + which hold multiple values in a single field are stored as a list. + + """ + + # Metadata 1.0 - PEP 241 + metadata_version: str + name: str + version: str + platforms: List[str] + summary: str + description: str + keywords: List[str] + home_page: str + author: str + author_email: str + license: str + + # Metadata 1.1 - PEP 314 + supported_platforms: List[str] + download_url: str + classifiers: List[str] + requires: List[str] + provides: List[str] + obsoletes: List[str] + + # Metadata 1.2 - PEP 345 + maintainer: str + maintainer_email: str + requires_dist: List[str] + provides_dist: List[str] + obsoletes_dist: List[str] + requires_python: str + requires_external: List[str] + project_urls: Dict[str, str] + + # Metadata 2.0 + # PEP 426 attempted to completely revamp the metadata format + # but got stuck without ever being able to build consensus on + # it and ultimately ended up withdrawn. + # + # However, a number of tools had started emiting METADATA with + # `2.0` Metadata-Version, so for historical reasons, this version + # was skipped. + + # Metadata 2.1 - PEP 566 + description_content_type: str + provides_extra: List[str] + + # Metadata 2.2 - PEP 643 + dynamic: List[str] + + # Metadata 2.3 - PEP 685 + # No new fields were added in PEP 685, just some edge case were + # tightened up to provide better interoptability. + + +_STRING_FIELDS = { + "author", + "author_email", + "description", + "description_content_type", + "download_url", + "home_page", + "license", + "maintainer", + "maintainer_email", + "metadata_version", + "name", + "requires_python", + "summary", + "version", +} + +_LIST_STRING_FIELDS = { + "classifiers", + "dynamic", + "obsoletes", + "obsoletes_dist", + "platforms", + "provides", + "provides_dist", + "provides_extra", + "requires", + "requires_dist", + "requires_external", + "supported_platforms", +} + + +def _parse_keywords(data: str) -> List[str]: + """Split a string of comma-separate keyboards into a list of keywords.""" + return [k.strip() for k in data.split(",")] + + +def _parse_project_urls(data: List[str]) -> Dict[str, str]: + """Parse a list of label/URL string pairings separated by a comma.""" + urls = {} + for pair in data: + # Our logic is slightly tricky here as we want to try and do + # *something* reasonable with malformed data. + # + # The main thing that we have to worry about, is data that does + # not have a ',' at all to split the label from the Value. There + # isn't a singular right answer here, and we will fail validation + # later on (if the caller is validating) so it doesn't *really* + # matter, but since the missing value has to be an empty str + # and our return value is dict[str, str], if we let the key + # be the missing value, then they'd have multiple '' values that + # overwrite each other in a accumulating dict. + # + # The other potentional issue is that it's possible to have the + # same label multiple times in the metadata, with no solid "right" + # answer with what to do in that case. As such, we'll do the only + # thing we can, which is treat the field as unparseable and add it + # to our list of unparsed fields. + parts = [p.strip() for p in pair.split(",", 1)] + parts.extend([""] * (max(0, 2 - len(parts)))) # Ensure 2 items + + # TODO: The spec doesn't say anything about if the keys should be + # considered case sensitive or not... logically they should + # be case-preserving and case-insensitive, but doing that + # would open up more cases where we might have duplicate + # entries. + label, url = parts + if label in urls: + # The label already exists in our set of urls, so this field + # is unparseable, and we can just add the whole thing to our + # unparseable data and stop processing it. + raise KeyError("duplicate labels in project urls") + urls[label] = url + + return urls + + +def _get_payload(msg: email.message.Message, source: Union[bytes, str]) -> str: + """Get the body of the message.""" + # If our source is a str, then our caller has managed encodings for us, + # and we don't need to deal with it. + if isinstance(source, str): + payload: str = msg.get_payload() + return payload + # If our source is a bytes, then we're managing the encoding and we need + # to deal with it. + else: + bpayload: bytes = msg.get_payload(decode=True) + try: + return bpayload.decode("utf8", "strict") + except UnicodeDecodeError: + raise ValueError("payload in an invalid encoding") + + +# The various parse_FORMAT functions here are intended to be as lenient as +# possible in their parsing, while still returning a correctly typed +# RawMetadata. +# +# To aid in this, we also generally want to do as little touching of the +# data as possible, except where there are possibly some historic holdovers +# that make valid data awkward to work with. +# +# While this is a lower level, intermediate format than our ``Metadata`` +# class, some light touch ups can make a massive difference in usability. + +# Map METADATA fields to RawMetadata. +_EMAIL_TO_RAW_MAPPING = { + "author": "author", + "author-email": "author_email", + "classifier": "classifiers", + "description": "description", + "description-content-type": "description_content_type", + "download-url": "download_url", + "dynamic": "dynamic", + "home-page": "home_page", + "keywords": "keywords", + "license": "license", + "maintainer": "maintainer", + "maintainer-email": "maintainer_email", + "metadata-version": "metadata_version", + "name": "name", + "obsoletes": "obsoletes", + "obsoletes-dist": "obsoletes_dist", + "platform": "platforms", + "project-url": "project_urls", + "provides": "provides", + "provides-dist": "provides_dist", + "provides-extra": "provides_extra", + "requires": "requires", + "requires-dist": "requires_dist", + "requires-external": "requires_external", + "requires-python": "requires_python", + "summary": "summary", + "supported-platform": "supported_platforms", + "version": "version", +} + + +def parse_email(data: Union[bytes, str]) -> Tuple[RawMetadata, Dict[str, List[str]]]: + """Parse a distribution's metadata. + + This function returns a two-item tuple of dicts. The first dict is of + recognized fields from the core metadata specification. Fields that can be + parsed and translated into Python's built-in types are converted + appropriately. All other fields are left as-is. Fields that are allowed to + appear multiple times are stored as lists. + + The second dict contains all other fields from the metadata. This includes + any unrecognized fields. It also includes any fields which are expected to + be parsed into a built-in type but were not formatted appropriately. Finally, + any fields that are expected to appear only once but are repeated are + included in this dict. + + """ + raw: Dict[str, Union[str, List[str], Dict[str, str]]] = {} + unparsed: Dict[str, List[str]] = {} + + if isinstance(data, str): + parsed = email.parser.Parser(policy=email.policy.compat32).parsestr(data) + else: + parsed = email.parser.BytesParser(policy=email.policy.compat32).parsebytes(data) + + # We have to wrap parsed.keys() in a set, because in the case of multiple + # values for a key (a list), the key will appear multiple times in the + # list of keys, but we're avoiding that by using get_all(). + for name in frozenset(parsed.keys()): + # Header names in RFC are case insensitive, so we'll normalize to all + # lower case to make comparisons easier. + name = name.lower() + + # We use get_all() here, even for fields that aren't multiple use, + # because otherwise someone could have e.g. two Name fields, and we + # would just silently ignore it rather than doing something about it. + headers = parsed.get_all(name) + + # The way the email module works when parsing bytes is that it + # unconditionally decodes the bytes as ascii using the surrogateescape + # handler. When you pull that data back out (such as with get_all() ), + # it looks to see if the str has any surrogate escapes, and if it does + # it wraps it in a Header object instead of returning the string. + # + # As such, we'll look for those Header objects, and fix up the encoding. + value = [] + # Flag if we have run into any issues processing the headers, thus + # signalling that the data belongs in 'unparsed'. + valid_encoding = True + for h in headers: + # It's unclear if this can return more types than just a Header or + # a str, so we'll just assert here to make sure. + assert isinstance(h, (email.header.Header, str)) + + # If it's a header object, we need to do our little dance to get + # the real data out of it. In cases where there is invalid data + # we're going to end up with mojibake, but there's no obvious, good + # way around that without reimplementing parts of the Header object + # ourselves. + # + # That should be fine since, if mojibacked happens, this key is + # going into the unparsed dict anyways. + if isinstance(h, email.header.Header): + # The Header object stores it's data as chunks, and each chunk + # can be independently encoded, so we'll need to check each + # of them. + chunks: List[Tuple[bytes, Optional[str]]] = [] + for bin, encoding in email.header.decode_header(h): + try: + bin.decode("utf8", "strict") + except UnicodeDecodeError: + # Enable mojibake. + encoding = "latin1" + valid_encoding = False + else: + encoding = "utf8" + chunks.append((bin, encoding)) + + # Turn our chunks back into a Header object, then let that + # Header object do the right thing to turn them into a + # string for us. + value.append(str(email.header.make_header(chunks))) + # This is already a string, so just add it. + else: + value.append(h) + + # We've processed all of our values to get them into a list of str, + # but we may have mojibake data, in which case this is an unparsed + # field. + if not valid_encoding: + unparsed[name] = value + continue + + raw_name = _EMAIL_TO_RAW_MAPPING.get(name) + if raw_name is None: + # This is a bit of a weird situation, we've encountered a key that + # we don't know what it means, so we don't know whether it's meant + # to be a list or not. + # + # Since we can't really tell one way or another, we'll just leave it + # as a list, even though it may be a single item list, because that's + # what makes the most sense for email headers. + unparsed[name] = value + continue + + # If this is one of our string fields, then we'll check to see if our + # value is a list of a single item. If it is then we'll assume that + # it was emitted as a single string, and unwrap the str from inside + # the list. + # + # If it's any other kind of data, then we haven't the faintest clue + # what we should parse it as, and we have to just add it to our list + # of unparsed stuff. + if raw_name in _STRING_FIELDS and len(value) == 1: + raw[raw_name] = value[0] + # If this is one of our list of string fields, then we can just assign + # the value, since email *only* has strings, and our get_all() call + # above ensures that this is a list. + elif raw_name in _LIST_STRING_FIELDS: + raw[raw_name] = value + # Special Case: Keywords + # The keywords field is implemented in the metadata spec as a str, + # but it conceptually is a list of strings, and is serialized using + # ", ".join(keywords), so we'll do some light data massaging to turn + # this into what it logically is. + elif raw_name == "keywords" and len(value) == 1: + raw[raw_name] = _parse_keywords(value[0]) + # Special Case: Project-URL + # The project urls is implemented in the metadata spec as a list of + # specially-formatted strings that represent a key and a value, which + # is fundamentally a mapping, however the email format doesn't support + # mappings in a sane way, so it was crammed into a list of strings + # instead. + # + # We will do a little light data massaging to turn this into a map as + # it logically should be. + elif raw_name == "project_urls": + try: + raw[raw_name] = _parse_project_urls(value) + except KeyError: + unparsed[name] = value + # Nothing that we've done has managed to parse this, so it'll just + # throw it in our unparseable data and move on. + else: + unparsed[name] = value + + # We need to support getting the Description from the message payload in + # addition to getting it from the the headers. This does mean, though, there + # is the possibility of it being set both ways, in which case we put both + # in 'unparsed' since we don't know which is right. + try: + payload = _get_payload(parsed, data) + except ValueError: + unparsed.setdefault("description", []).append( + parsed.get_payload(decode=isinstance(data, bytes)) + ) + else: + if payload: + # Check to see if we've already got a description, if so then both + # it, and this body move to unparseable. + if "description" in raw: + description_header = cast(str, raw.pop("description")) + unparsed.setdefault("description", []).extend( + [description_header, payload] + ) + elif "description" in unparsed: + unparsed["description"].append(payload) + else: + raw["description"] = payload + + # We need to cast our `raw` to a metadata, because a TypedDict only support + # literal key names, but we're computing our key names on purpose, but the + # way this function is implemented, our `TypedDict` can only have valid key + # names. + return cast(RawMetadata, raw), unparsed diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/requirements.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/requirements.py new file mode 100644 index 0000000000..f34bfa85c8 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/requirements.py @@ -0,0 +1,95 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. + +import urllib.parse +from typing import Any, List, Optional, Set + +from ._parser import parse_requirement as _parse_requirement +from ._tokenizer import ParserSyntaxError +from .markers import Marker, _normalize_extra_values +from .specifiers import SpecifierSet + + +class InvalidRequirement(ValueError): + """ + An invalid requirement was found, users should refer to PEP 508. + """ + + +class Requirement: + """Parse a requirement. + + Parse a given requirement string into its parts, such as name, specifier, + URL, and extras. Raises InvalidRequirement on a badly-formed requirement + string. + """ + + # TODO: Can we test whether something is contained within a requirement? + # If so how do we do that? Do we need to test against the _name_ of + # the thing as well as the version? What about the markers? + # TODO: Can we normalize the name and extra name? + + def __init__(self, requirement_string: str) -> None: + try: + parsed = _parse_requirement(requirement_string) + except ParserSyntaxError as e: + raise InvalidRequirement(str(e)) from e + + self.name: str = parsed.name + if parsed.url: + parsed_url = urllib.parse.urlparse(parsed.url) + if parsed_url.scheme == "file": + if urllib.parse.urlunparse(parsed_url) != parsed.url: + raise InvalidRequirement("Invalid URL given") + elif not (parsed_url.scheme and parsed_url.netloc) or ( + not parsed_url.scheme and not parsed_url.netloc + ): + raise InvalidRequirement(f"Invalid URL: {parsed.url}") + self.url: Optional[str] = parsed.url + else: + self.url = None + self.extras: Set[str] = set(parsed.extras if parsed.extras else []) + self.specifier: SpecifierSet = SpecifierSet(parsed.specifier) + self.marker: Optional[Marker] = None + if parsed.marker is not None: + self.marker = Marker.__new__(Marker) + self.marker._markers = _normalize_extra_values(parsed.marker) + + def __str__(self) -> str: + parts: List[str] = [self.name] + + if self.extras: + formatted_extras = ",".join(sorted(self.extras)) + parts.append(f"[{formatted_extras}]") + + if self.specifier: + parts.append(str(self.specifier)) + + if self.url: + parts.append(f"@ {self.url}") + if self.marker: + parts.append(" ") + + if self.marker: + parts.append(f"; {self.marker}") + + return "".join(parts) + + def __repr__(self) -> str: + return f"<Requirement('{self}')>" + + def __hash__(self) -> int: + return hash((self.__class__.__name__, str(self))) + + def __eq__(self, other: Any) -> bool: + if not isinstance(other, Requirement): + return NotImplemented + + return ( + self.name == other.name + and self.extras == other.extras + and self.specifier == other.specifier + and self.url == other.url + and self.marker == other.marker + ) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/specifiers.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/specifiers.py new file mode 100644 index 0000000000..ba8fe37b7f --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/specifiers.py @@ -0,0 +1,1008 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. +""" +.. testsetup:: + + from packaging.specifiers import Specifier, SpecifierSet, InvalidSpecifier + from packaging.version import Version +""" + +import abc +import itertools +import re +from typing import ( + Callable, + Iterable, + Iterator, + List, + Optional, + Set, + Tuple, + TypeVar, + Union, +) + +from .utils import canonicalize_version +from .version import Version + +UnparsedVersion = Union[Version, str] +UnparsedVersionVar = TypeVar("UnparsedVersionVar", bound=UnparsedVersion) +CallableOperator = Callable[[Version, str], bool] + + +def _coerce_version(version: UnparsedVersion) -> Version: + if not isinstance(version, Version): + version = Version(version) + return version + + +class InvalidSpecifier(ValueError): + """ + Raised when attempting to create a :class:`Specifier` with a specifier + string that is invalid. + + >>> Specifier("lolwat") + Traceback (most recent call last): + ... + packaging.specifiers.InvalidSpecifier: Invalid specifier: 'lolwat' + """ + + +class BaseSpecifier(metaclass=abc.ABCMeta): + @abc.abstractmethod + def __str__(self) -> str: + """ + Returns the str representation of this Specifier-like object. This + should be representative of the Specifier itself. + """ + + @abc.abstractmethod + def __hash__(self) -> int: + """ + Returns a hash value for this Specifier-like object. + """ + + @abc.abstractmethod + def __eq__(self, other: object) -> bool: + """ + Returns a boolean representing whether or not the two Specifier-like + objects are equal. + + :param other: The other object to check against. + """ + + @property + @abc.abstractmethod + def prereleases(self) -> Optional[bool]: + """Whether or not pre-releases as a whole are allowed. + + This can be set to either ``True`` or ``False`` to explicitly enable or disable + prereleases or it can be set to ``None`` (the default) to use default semantics. + """ + + @prereleases.setter + def prereleases(self, value: bool) -> None: + """Setter for :attr:`prereleases`. + + :param value: The value to set. + """ + + @abc.abstractmethod + def contains(self, item: str, prereleases: Optional[bool] = None) -> bool: + """ + Determines if the given item is contained within this specifier. + """ + + @abc.abstractmethod + def filter( + self, iterable: Iterable[UnparsedVersionVar], prereleases: Optional[bool] = None + ) -> Iterator[UnparsedVersionVar]: + """ + Takes an iterable of items and filters them so that only items which + are contained within this specifier are allowed in it. + """ + + +class Specifier(BaseSpecifier): + """This class abstracts handling of version specifiers. + + .. tip:: + + It is generally not required to instantiate this manually. You should instead + prefer to work with :class:`SpecifierSet` instead, which can parse + comma-separated version specifiers (which is what package metadata contains). + """ + + _operator_regex_str = r""" + (?P<operator>(~=|==|!=|<=|>=|<|>|===)) + """ + _version_regex_str = r""" + (?P<version> + (?: + # The identity operators allow for an escape hatch that will + # do an exact string match of the version you wish to install. + # This will not be parsed by PEP 440 and we cannot determine + # any semantic meaning from it. This operator is discouraged + # but included entirely as an escape hatch. + (?<====) # Only match for the identity operator + \s* + [^\s;)]* # The arbitrary version can be just about anything, + # we match everything except for whitespace, a + # semi-colon for marker support, and a closing paren + # since versions can be enclosed in them. + ) + | + (?: + # The (non)equality operators allow for wild card and local + # versions to be specified so we have to define these two + # operators separately to enable that. + (?<===|!=) # Only match for equals and not equals + + \s* + v? + (?:[0-9]+!)? # epoch + [0-9]+(?:\.[0-9]+)* # release + + # You cannot use a wild card and a pre-release, post-release, a dev or + # local version together so group them with a | and make them optional. + (?: + \.\* # Wild card syntax of .* + | + (?: # pre release + [-_\.]? + (alpha|beta|preview|pre|a|b|c|rc) + [-_\.]? + [0-9]* + )? + (?: # post release + (?:-[0-9]+)|(?:[-_\.]?(post|rev|r)[-_\.]?[0-9]*) + )? + (?:[-_\.]?dev[-_\.]?[0-9]*)? # dev release + (?:\+[a-z0-9]+(?:[-_\.][a-z0-9]+)*)? # local + )? + ) + | + (?: + # The compatible operator requires at least two digits in the + # release segment. + (?<=~=) # Only match for the compatible operator + + \s* + v? + (?:[0-9]+!)? # epoch + [0-9]+(?:\.[0-9]+)+ # release (We have a + instead of a *) + (?: # pre release + [-_\.]? + (alpha|beta|preview|pre|a|b|c|rc) + [-_\.]? + [0-9]* + )? + (?: # post release + (?:-[0-9]+)|(?:[-_\.]?(post|rev|r)[-_\.]?[0-9]*) + )? + (?:[-_\.]?dev[-_\.]?[0-9]*)? # dev release + ) + | + (?: + # All other operators only allow a sub set of what the + # (non)equality operators do. Specifically they do not allow + # local versions to be specified nor do they allow the prefix + # matching wild cards. + (?<!==|!=|~=) # We have special cases for these + # operators so we want to make sure they + # don't match here. + + \s* + v? + (?:[0-9]+!)? # epoch + [0-9]+(?:\.[0-9]+)* # release + (?: # pre release + [-_\.]? + (alpha|beta|preview|pre|a|b|c|rc) + [-_\.]? + [0-9]* + )? + (?: # post release + (?:-[0-9]+)|(?:[-_\.]?(post|rev|r)[-_\.]?[0-9]*) + )? + (?:[-_\.]?dev[-_\.]?[0-9]*)? # dev release + ) + ) + """ + + _regex = re.compile( + r"^\s*" + _operator_regex_str + _version_regex_str + r"\s*$", + re.VERBOSE | re.IGNORECASE, + ) + + _operators = { + "~=": "compatible", + "==": "equal", + "!=": "not_equal", + "<=": "less_than_equal", + ">=": "greater_than_equal", + "<": "less_than", + ">": "greater_than", + "===": "arbitrary", + } + + def __init__(self, spec: str = "", prereleases: Optional[bool] = None) -> None: + """Initialize a Specifier instance. + + :param spec: + The string representation of a specifier which will be parsed and + normalized before use. + :param prereleases: + This tells the specifier if it should accept prerelease versions if + applicable or not. The default of ``None`` will autodetect it from the + given specifiers. + :raises InvalidSpecifier: + If the given specifier is invalid (i.e. bad syntax). + """ + match = self._regex.search(spec) + if not match: + raise InvalidSpecifier(f"Invalid specifier: '{spec}'") + + self._spec: Tuple[str, str] = ( + match.group("operator").strip(), + match.group("version").strip(), + ) + + # Store whether or not this Specifier should accept prereleases + self._prereleases = prereleases + + # https://github.com/python/mypy/pull/13475#pullrequestreview-1079784515 + @property # type: ignore[override] + def prereleases(self) -> bool: + # If there is an explicit prereleases set for this, then we'll just + # blindly use that. + if self._prereleases is not None: + return self._prereleases + + # Look at all of our specifiers and determine if they are inclusive + # operators, and if they are if they are including an explicit + # prerelease. + operator, version = self._spec + if operator in ["==", ">=", "<=", "~=", "==="]: + # The == specifier can include a trailing .*, if it does we + # want to remove before parsing. + if operator == "==" and version.endswith(".*"): + version = version[:-2] + + # Parse the version, and if it is a pre-release than this + # specifier allows pre-releases. + if Version(version).is_prerelease: + return True + + return False + + @prereleases.setter + def prereleases(self, value: bool) -> None: + self._prereleases = value + + @property + def operator(self) -> str: + """The operator of this specifier. + + >>> Specifier("==1.2.3").operator + '==' + """ + return self._spec[0] + + @property + def version(self) -> str: + """The version of this specifier. + + >>> Specifier("==1.2.3").version + '1.2.3' + """ + return self._spec[1] + + def __repr__(self) -> str: + """A representation of the Specifier that shows all internal state. + + >>> Specifier('>=1.0.0') + <Specifier('>=1.0.0')> + >>> Specifier('>=1.0.0', prereleases=False) + <Specifier('>=1.0.0', prereleases=False)> + >>> Specifier('>=1.0.0', prereleases=True) + <Specifier('>=1.0.0', prereleases=True)> + """ + pre = ( + f", prereleases={self.prereleases!r}" + if self._prereleases is not None + else "" + ) + + return f"<{self.__class__.__name__}({str(self)!r}{pre})>" + + def __str__(self) -> str: + """A string representation of the Specifier that can be round-tripped. + + >>> str(Specifier('>=1.0.0')) + '>=1.0.0' + >>> str(Specifier('>=1.0.0', prereleases=False)) + '>=1.0.0' + """ + return "{}{}".format(*self._spec) + + @property + def _canonical_spec(self) -> Tuple[str, str]: + canonical_version = canonicalize_version( + self._spec[1], + strip_trailing_zero=(self._spec[0] != "~="), + ) + return self._spec[0], canonical_version + + def __hash__(self) -> int: + return hash(self._canonical_spec) + + def __eq__(self, other: object) -> bool: + """Whether or not the two Specifier-like objects are equal. + + :param other: The other object to check against. + + The value of :attr:`prereleases` is ignored. + + >>> Specifier("==1.2.3") == Specifier("== 1.2.3.0") + True + >>> (Specifier("==1.2.3", prereleases=False) == + ... Specifier("==1.2.3", prereleases=True)) + True + >>> Specifier("==1.2.3") == "==1.2.3" + True + >>> Specifier("==1.2.3") == Specifier("==1.2.4") + False + >>> Specifier("==1.2.3") == Specifier("~=1.2.3") + False + """ + if isinstance(other, str): + try: + other = self.__class__(str(other)) + except InvalidSpecifier: + return NotImplemented + elif not isinstance(other, self.__class__): + return NotImplemented + + return self._canonical_spec == other._canonical_spec + + def _get_operator(self, op: str) -> CallableOperator: + operator_callable: CallableOperator = getattr( + self, f"_compare_{self._operators[op]}" + ) + return operator_callable + + def _compare_compatible(self, prospective: Version, spec: str) -> bool: + + # Compatible releases have an equivalent combination of >= and ==. That + # is that ~=2.2 is equivalent to >=2.2,==2.*. This allows us to + # implement this in terms of the other specifiers instead of + # implementing it ourselves. The only thing we need to do is construct + # the other specifiers. + + # We want everything but the last item in the version, but we want to + # ignore suffix segments. + prefix = ".".join( + list(itertools.takewhile(_is_not_suffix, _version_split(spec)))[:-1] + ) + + # Add the prefix notation to the end of our string + prefix += ".*" + + return self._get_operator(">=")(prospective, spec) and self._get_operator("==")( + prospective, prefix + ) + + def _compare_equal(self, prospective: Version, spec: str) -> bool: + + # We need special logic to handle prefix matching + if spec.endswith(".*"): + # In the case of prefix matching we want to ignore local segment. + normalized_prospective = canonicalize_version( + prospective.public, strip_trailing_zero=False + ) + # Get the normalized version string ignoring the trailing .* + normalized_spec = canonicalize_version(spec[:-2], strip_trailing_zero=False) + # Split the spec out by dots, and pretend that there is an implicit + # dot in between a release segment and a pre-release segment. + split_spec = _version_split(normalized_spec) + + # Split the prospective version out by dots, and pretend that there + # is an implicit dot in between a release segment and a pre-release + # segment. + split_prospective = _version_split(normalized_prospective) + + # 0-pad the prospective version before shortening it to get the correct + # shortened version. + padded_prospective, _ = _pad_version(split_prospective, split_spec) + + # Shorten the prospective version to be the same length as the spec + # so that we can determine if the specifier is a prefix of the + # prospective version or not. + shortened_prospective = padded_prospective[: len(split_spec)] + + return shortened_prospective == split_spec + else: + # Convert our spec string into a Version + spec_version = Version(spec) + + # If the specifier does not have a local segment, then we want to + # act as if the prospective version also does not have a local + # segment. + if not spec_version.local: + prospective = Version(prospective.public) + + return prospective == spec_version + + def _compare_not_equal(self, prospective: Version, spec: str) -> bool: + return not self._compare_equal(prospective, spec) + + def _compare_less_than_equal(self, prospective: Version, spec: str) -> bool: + + # NB: Local version identifiers are NOT permitted in the version + # specifier, so local version labels can be universally removed from + # the prospective version. + return Version(prospective.public) <= Version(spec) + + def _compare_greater_than_equal(self, prospective: Version, spec: str) -> bool: + + # NB: Local version identifiers are NOT permitted in the version + # specifier, so local version labels can be universally removed from + # the prospective version. + return Version(prospective.public) >= Version(spec) + + def _compare_less_than(self, prospective: Version, spec_str: str) -> bool: + + # Convert our spec to a Version instance, since we'll want to work with + # it as a version. + spec = Version(spec_str) + + # Check to see if the prospective version is less than the spec + # version. If it's not we can short circuit and just return False now + # instead of doing extra unneeded work. + if not prospective < spec: + return False + + # This special case is here so that, unless the specifier itself + # includes is a pre-release version, that we do not accept pre-release + # versions for the version mentioned in the specifier (e.g. <3.1 should + # not match 3.1.dev0, but should match 3.0.dev0). + if not spec.is_prerelease and prospective.is_prerelease: + if Version(prospective.base_version) == Version(spec.base_version): + return False + + # If we've gotten to here, it means that prospective version is both + # less than the spec version *and* it's not a pre-release of the same + # version in the spec. + return True + + def _compare_greater_than(self, prospective: Version, spec_str: str) -> bool: + + # Convert our spec to a Version instance, since we'll want to work with + # it as a version. + spec = Version(spec_str) + + # Check to see if the prospective version is greater than the spec + # version. If it's not we can short circuit and just return False now + # instead of doing extra unneeded work. + if not prospective > spec: + return False + + # This special case is here so that, unless the specifier itself + # includes is a post-release version, that we do not accept + # post-release versions for the version mentioned in the specifier + # (e.g. >3.1 should not match 3.0.post0, but should match 3.2.post0). + if not spec.is_postrelease and prospective.is_postrelease: + if Version(prospective.base_version) == Version(spec.base_version): + return False + + # Ensure that we do not allow a local version of the version mentioned + # in the specifier, which is technically greater than, to match. + if prospective.local is not None: + if Version(prospective.base_version) == Version(spec.base_version): + return False + + # If we've gotten to here, it means that prospective version is both + # greater than the spec version *and* it's not a pre-release of the + # same version in the spec. + return True + + def _compare_arbitrary(self, prospective: Version, spec: str) -> bool: + return str(prospective).lower() == str(spec).lower() + + def __contains__(self, item: Union[str, Version]) -> bool: + """Return whether or not the item is contained in this specifier. + + :param item: The item to check for. + + This is used for the ``in`` operator and behaves the same as + :meth:`contains` with no ``prereleases`` argument passed. + + >>> "1.2.3" in Specifier(">=1.2.3") + True + >>> Version("1.2.3") in Specifier(">=1.2.3") + True + >>> "1.0.0" in Specifier(">=1.2.3") + False + >>> "1.3.0a1" in Specifier(">=1.2.3") + False + >>> "1.3.0a1" in Specifier(">=1.2.3", prereleases=True) + True + """ + return self.contains(item) + + def contains( + self, item: UnparsedVersion, prereleases: Optional[bool] = None + ) -> bool: + """Return whether or not the item is contained in this specifier. + + :param item: + The item to check for, which can be a version string or a + :class:`Version` instance. + :param prereleases: + Whether or not to match prereleases with this Specifier. If set to + ``None`` (the default), it uses :attr:`prereleases` to determine + whether or not prereleases are allowed. + + >>> Specifier(">=1.2.3").contains("1.2.3") + True + >>> Specifier(">=1.2.3").contains(Version("1.2.3")) + True + >>> Specifier(">=1.2.3").contains("1.0.0") + False + >>> Specifier(">=1.2.3").contains("1.3.0a1") + False + >>> Specifier(">=1.2.3", prereleases=True).contains("1.3.0a1") + True + >>> Specifier(">=1.2.3").contains("1.3.0a1", prereleases=True) + True + """ + + # Determine if prereleases are to be allowed or not. + if prereleases is None: + prereleases = self.prereleases + + # Normalize item to a Version, this allows us to have a shortcut for + # "2.0" in Specifier(">=2") + normalized_item = _coerce_version(item) + + # Determine if we should be supporting prereleases in this specifier + # or not, if we do not support prereleases than we can short circuit + # logic if this version is a prereleases. + if normalized_item.is_prerelease and not prereleases: + return False + + # Actually do the comparison to determine if this item is contained + # within this Specifier or not. + operator_callable: CallableOperator = self._get_operator(self.operator) + return operator_callable(normalized_item, self.version) + + def filter( + self, iterable: Iterable[UnparsedVersionVar], prereleases: Optional[bool] = None + ) -> Iterator[UnparsedVersionVar]: + """Filter items in the given iterable, that match the specifier. + + :param iterable: + An iterable that can contain version strings and :class:`Version` instances. + The items in the iterable will be filtered according to the specifier. + :param prereleases: + Whether or not to allow prereleases in the returned iterator. If set to + ``None`` (the default), it will be intelligently decide whether to allow + prereleases or not (based on the :attr:`prereleases` attribute, and + whether the only versions matching are prereleases). + + This method is smarter than just ``filter(Specifier().contains, [...])`` + because it implements the rule from :pep:`440` that a prerelease item + SHOULD be accepted if no other versions match the given specifier. + + >>> list(Specifier(">=1.2.3").filter(["1.2", "1.3", "1.5a1"])) + ['1.3'] + >>> list(Specifier(">=1.2.3").filter(["1.2", "1.2.3", "1.3", Version("1.4")])) + ['1.2.3', '1.3', <Version('1.4')>] + >>> list(Specifier(">=1.2.3").filter(["1.2", "1.5a1"])) + ['1.5a1'] + >>> list(Specifier(">=1.2.3").filter(["1.3", "1.5a1"], prereleases=True)) + ['1.3', '1.5a1'] + >>> list(Specifier(">=1.2.3", prereleases=True).filter(["1.3", "1.5a1"])) + ['1.3', '1.5a1'] + """ + + yielded = False + found_prereleases = [] + + kw = {"prereleases": prereleases if prereleases is not None else True} + + # Attempt to iterate over all the values in the iterable and if any of + # them match, yield them. + for version in iterable: + parsed_version = _coerce_version(version) + + if self.contains(parsed_version, **kw): + # If our version is a prerelease, and we were not set to allow + # prereleases, then we'll store it for later in case nothing + # else matches this specifier. + if parsed_version.is_prerelease and not ( + prereleases or self.prereleases + ): + found_prereleases.append(version) + # Either this is not a prerelease, or we should have been + # accepting prereleases from the beginning. + else: + yielded = True + yield version + + # Now that we've iterated over everything, determine if we've yielded + # any values, and if we have not and we have any prereleases stored up + # then we will go ahead and yield the prereleases. + if not yielded and found_prereleases: + for version in found_prereleases: + yield version + + +_prefix_regex = re.compile(r"^([0-9]+)((?:a|b|c|rc)[0-9]+)$") + + +def _version_split(version: str) -> List[str]: + result: List[str] = [] + for item in version.split("."): + match = _prefix_regex.search(item) + if match: + result.extend(match.groups()) + else: + result.append(item) + return result + + +def _is_not_suffix(segment: str) -> bool: + return not any( + segment.startswith(prefix) for prefix in ("dev", "a", "b", "rc", "post") + ) + + +def _pad_version(left: List[str], right: List[str]) -> Tuple[List[str], List[str]]: + left_split, right_split = [], [] + + # Get the release segment of our versions + left_split.append(list(itertools.takewhile(lambda x: x.isdigit(), left))) + right_split.append(list(itertools.takewhile(lambda x: x.isdigit(), right))) + + # Get the rest of our versions + left_split.append(left[len(left_split[0]) :]) + right_split.append(right[len(right_split[0]) :]) + + # Insert our padding + left_split.insert(1, ["0"] * max(0, len(right_split[0]) - len(left_split[0]))) + right_split.insert(1, ["0"] * max(0, len(left_split[0]) - len(right_split[0]))) + + return (list(itertools.chain(*left_split)), list(itertools.chain(*right_split))) + + +class SpecifierSet(BaseSpecifier): + """This class abstracts handling of a set of version specifiers. + + It can be passed a single specifier (``>=3.0``), a comma-separated list of + specifiers (``>=3.0,!=3.1``), or no specifier at all. + """ + + def __init__( + self, specifiers: str = "", prereleases: Optional[bool] = None + ) -> None: + """Initialize a SpecifierSet instance. + + :param specifiers: + The string representation of a specifier or a comma-separated list of + specifiers which will be parsed and normalized before use. + :param prereleases: + This tells the SpecifierSet if it should accept prerelease versions if + applicable or not. The default of ``None`` will autodetect it from the + given specifiers. + + :raises InvalidSpecifier: + If the given ``specifiers`` are not parseable than this exception will be + raised. + """ + + # Split on `,` to break each individual specifier into it's own item, and + # strip each item to remove leading/trailing whitespace. + split_specifiers = [s.strip() for s in specifiers.split(",") if s.strip()] + + # Parsed each individual specifier, attempting first to make it a + # Specifier. + parsed: Set[Specifier] = set() + for specifier in split_specifiers: + parsed.add(Specifier(specifier)) + + # Turn our parsed specifiers into a frozen set and save them for later. + self._specs = frozenset(parsed) + + # Store our prereleases value so we can use it later to determine if + # we accept prereleases or not. + self._prereleases = prereleases + + @property + def prereleases(self) -> Optional[bool]: + # If we have been given an explicit prerelease modifier, then we'll + # pass that through here. + if self._prereleases is not None: + return self._prereleases + + # If we don't have any specifiers, and we don't have a forced value, + # then we'll just return None since we don't know if this should have + # pre-releases or not. + if not self._specs: + return None + + # Otherwise we'll see if any of the given specifiers accept + # prereleases, if any of them do we'll return True, otherwise False. + return any(s.prereleases for s in self._specs) + + @prereleases.setter + def prereleases(self, value: bool) -> None: + self._prereleases = value + + def __repr__(self) -> str: + """A representation of the specifier set that shows all internal state. + + Note that the ordering of the individual specifiers within the set may not + match the input string. + + >>> SpecifierSet('>=1.0.0,!=2.0.0') + <SpecifierSet('!=2.0.0,>=1.0.0')> + >>> SpecifierSet('>=1.0.0,!=2.0.0', prereleases=False) + <SpecifierSet('!=2.0.0,>=1.0.0', prereleases=False)> + >>> SpecifierSet('>=1.0.0,!=2.0.0', prereleases=True) + <SpecifierSet('!=2.0.0,>=1.0.0', prereleases=True)> + """ + pre = ( + f", prereleases={self.prereleases!r}" + if self._prereleases is not None + else "" + ) + + return f"<SpecifierSet({str(self)!r}{pre})>" + + def __str__(self) -> str: + """A string representation of the specifier set that can be round-tripped. + + Note that the ordering of the individual specifiers within the set may not + match the input string. + + >>> str(SpecifierSet(">=1.0.0,!=1.0.1")) + '!=1.0.1,>=1.0.0' + >>> str(SpecifierSet(">=1.0.0,!=1.0.1", prereleases=False)) + '!=1.0.1,>=1.0.0' + """ + return ",".join(sorted(str(s) for s in self._specs)) + + def __hash__(self) -> int: + return hash(self._specs) + + def __and__(self, other: Union["SpecifierSet", str]) -> "SpecifierSet": + """Return a SpecifierSet which is a combination of the two sets. + + :param other: The other object to combine with. + + >>> SpecifierSet(">=1.0.0,!=1.0.1") & '<=2.0.0,!=2.0.1' + <SpecifierSet('!=1.0.1,!=2.0.1,<=2.0.0,>=1.0.0')> + >>> SpecifierSet(">=1.0.0,!=1.0.1") & SpecifierSet('<=2.0.0,!=2.0.1') + <SpecifierSet('!=1.0.1,!=2.0.1,<=2.0.0,>=1.0.0')> + """ + if isinstance(other, str): + other = SpecifierSet(other) + elif not isinstance(other, SpecifierSet): + return NotImplemented + + specifier = SpecifierSet() + specifier._specs = frozenset(self._specs | other._specs) + + if self._prereleases is None and other._prereleases is not None: + specifier._prereleases = other._prereleases + elif self._prereleases is not None and other._prereleases is None: + specifier._prereleases = self._prereleases + elif self._prereleases == other._prereleases: + specifier._prereleases = self._prereleases + else: + raise ValueError( + "Cannot combine SpecifierSets with True and False prerelease " + "overrides." + ) + + return specifier + + def __eq__(self, other: object) -> bool: + """Whether or not the two SpecifierSet-like objects are equal. + + :param other: The other object to check against. + + The value of :attr:`prereleases` is ignored. + + >>> SpecifierSet(">=1.0.0,!=1.0.1") == SpecifierSet(">=1.0.0,!=1.0.1") + True + >>> (SpecifierSet(">=1.0.0,!=1.0.1", prereleases=False) == + ... SpecifierSet(">=1.0.0,!=1.0.1", prereleases=True)) + True + >>> SpecifierSet(">=1.0.0,!=1.0.1") == ">=1.0.0,!=1.0.1" + True + >>> SpecifierSet(">=1.0.0,!=1.0.1") == SpecifierSet(">=1.0.0") + False + >>> SpecifierSet(">=1.0.0,!=1.0.1") == SpecifierSet(">=1.0.0,!=1.0.2") + False + """ + if isinstance(other, (str, Specifier)): + other = SpecifierSet(str(other)) + elif not isinstance(other, SpecifierSet): + return NotImplemented + + return self._specs == other._specs + + def __len__(self) -> int: + """Returns the number of specifiers in this specifier set.""" + return len(self._specs) + + def __iter__(self) -> Iterator[Specifier]: + """ + Returns an iterator over all the underlying :class:`Specifier` instances + in this specifier set. + + >>> sorted(SpecifierSet(">=1.0.0,!=1.0.1"), key=str) + [<Specifier('!=1.0.1')>, <Specifier('>=1.0.0')>] + """ + return iter(self._specs) + + def __contains__(self, item: UnparsedVersion) -> bool: + """Return whether or not the item is contained in this specifier. + + :param item: The item to check for. + + This is used for the ``in`` operator and behaves the same as + :meth:`contains` with no ``prereleases`` argument passed. + + >>> "1.2.3" in SpecifierSet(">=1.0.0,!=1.0.1") + True + >>> Version("1.2.3") in SpecifierSet(">=1.0.0,!=1.0.1") + True + >>> "1.0.1" in SpecifierSet(">=1.0.0,!=1.0.1") + False + >>> "1.3.0a1" in SpecifierSet(">=1.0.0,!=1.0.1") + False + >>> "1.3.0a1" in SpecifierSet(">=1.0.0,!=1.0.1", prereleases=True) + True + """ + return self.contains(item) + + def contains( + self, + item: UnparsedVersion, + prereleases: Optional[bool] = None, + installed: Optional[bool] = None, + ) -> bool: + """Return whether or not the item is contained in this SpecifierSet. + + :param item: + The item to check for, which can be a version string or a + :class:`Version` instance. + :param prereleases: + Whether or not to match prereleases with this SpecifierSet. If set to + ``None`` (the default), it uses :attr:`prereleases` to determine + whether or not prereleases are allowed. + + >>> SpecifierSet(">=1.0.0,!=1.0.1").contains("1.2.3") + True + >>> SpecifierSet(">=1.0.0,!=1.0.1").contains(Version("1.2.3")) + True + >>> SpecifierSet(">=1.0.0,!=1.0.1").contains("1.0.1") + False + >>> SpecifierSet(">=1.0.0,!=1.0.1").contains("1.3.0a1") + False + >>> SpecifierSet(">=1.0.0,!=1.0.1", prereleases=True).contains("1.3.0a1") + True + >>> SpecifierSet(">=1.0.0,!=1.0.1").contains("1.3.0a1", prereleases=True) + True + """ + # Ensure that our item is a Version instance. + if not isinstance(item, Version): + item = Version(item) + + # Determine if we're forcing a prerelease or not, if we're not forcing + # one for this particular filter call, then we'll use whatever the + # SpecifierSet thinks for whether or not we should support prereleases. + if prereleases is None: + prereleases = self.prereleases + + # We can determine if we're going to allow pre-releases by looking to + # see if any of the underlying items supports them. If none of them do + # and this item is a pre-release then we do not allow it and we can + # short circuit that here. + # Note: This means that 1.0.dev1 would not be contained in something + # like >=1.0.devabc however it would be in >=1.0.debabc,>0.0.dev0 + if not prereleases and item.is_prerelease: + return False + + if installed and item.is_prerelease: + item = Version(item.base_version) + + # We simply dispatch to the underlying specs here to make sure that the + # given version is contained within all of them. + # Note: This use of all() here means that an empty set of specifiers + # will always return True, this is an explicit design decision. + return all(s.contains(item, prereleases=prereleases) for s in self._specs) + + def filter( + self, iterable: Iterable[UnparsedVersionVar], prereleases: Optional[bool] = None + ) -> Iterator[UnparsedVersionVar]: + """Filter items in the given iterable, that match the specifiers in this set. + + :param iterable: + An iterable that can contain version strings and :class:`Version` instances. + The items in the iterable will be filtered according to the specifier. + :param prereleases: + Whether or not to allow prereleases in the returned iterator. If set to + ``None`` (the default), it will be intelligently decide whether to allow + prereleases or not (based on the :attr:`prereleases` attribute, and + whether the only versions matching are prereleases). + + This method is smarter than just ``filter(SpecifierSet(...).contains, [...])`` + because it implements the rule from :pep:`440` that a prerelease item + SHOULD be accepted if no other versions match the given specifier. + + >>> list(SpecifierSet(">=1.2.3").filter(["1.2", "1.3", "1.5a1"])) + ['1.3'] + >>> list(SpecifierSet(">=1.2.3").filter(["1.2", "1.3", Version("1.4")])) + ['1.3', <Version('1.4')>] + >>> list(SpecifierSet(">=1.2.3").filter(["1.2", "1.5a1"])) + [] + >>> list(SpecifierSet(">=1.2.3").filter(["1.3", "1.5a1"], prereleases=True)) + ['1.3', '1.5a1'] + >>> list(SpecifierSet(">=1.2.3", prereleases=True).filter(["1.3", "1.5a1"])) + ['1.3', '1.5a1'] + + An "empty" SpecifierSet will filter items based on the presence of prerelease + versions in the set. + + >>> list(SpecifierSet("").filter(["1.3", "1.5a1"])) + ['1.3'] + >>> list(SpecifierSet("").filter(["1.5a1"])) + ['1.5a1'] + >>> list(SpecifierSet("", prereleases=True).filter(["1.3", "1.5a1"])) + ['1.3', '1.5a1'] + >>> list(SpecifierSet("").filter(["1.3", "1.5a1"], prereleases=True)) + ['1.3', '1.5a1'] + """ + # Determine if we're forcing a prerelease or not, if we're not forcing + # one for this particular filter call, then we'll use whatever the + # SpecifierSet thinks for whether or not we should support prereleases. + if prereleases is None: + prereleases = self.prereleases + + # If we have any specifiers, then we want to wrap our iterable in the + # filter method for each one, this will act as a logical AND amongst + # each specifier. + if self._specs: + for spec in self._specs: + iterable = spec.filter(iterable, prereleases=bool(prereleases)) + return iter(iterable) + # If we do not have any specifiers, then we need to have a rough filter + # which will filter out any pre-releases, unless there are no final + # releases. + else: + filtered: List[UnparsedVersionVar] = [] + found_prereleases: List[UnparsedVersionVar] = [] + + for item in iterable: + parsed_version = _coerce_version(item) + + # Store any item which is a pre-release for later unless we've + # already found a final version or we are accepting prereleases + if parsed_version.is_prerelease and not prereleases: + if not filtered: + found_prereleases.append(item) + else: + filtered.append(item) + + # If we've found no items except for pre-releases, then we'll go + # ahead and use the pre-releases + if not filtered and found_prereleases and prereleases is None: + return iter(found_prereleases) + + return iter(filtered) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/tags.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/tags.py new file mode 100644 index 0000000000..76d243414d --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/tags.py @@ -0,0 +1,546 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. + +import logging +import platform +import subprocess +import sys +import sysconfig +from importlib.machinery import EXTENSION_SUFFIXES +from typing import ( + Dict, + FrozenSet, + Iterable, + Iterator, + List, + Optional, + Sequence, + Tuple, + Union, + cast, +) + +from . import _manylinux, _musllinux + +logger = logging.getLogger(__name__) + +PythonVersion = Sequence[int] +MacVersion = Tuple[int, int] + +INTERPRETER_SHORT_NAMES: Dict[str, str] = { + "python": "py", # Generic. + "cpython": "cp", + "pypy": "pp", + "ironpython": "ip", + "jython": "jy", +} + + +_32_BIT_INTERPRETER = sys.maxsize <= 2**32 + + +class Tag: + """ + A representation of the tag triple for a wheel. + + Instances are considered immutable and thus are hashable. Equality checking + is also supported. + """ + + __slots__ = ["_interpreter", "_abi", "_platform", "_hash"] + + def __init__(self, interpreter: str, abi: str, platform: str) -> None: + self._interpreter = interpreter.lower() + self._abi = abi.lower() + self._platform = platform.lower() + # The __hash__ of every single element in a Set[Tag] will be evaluated each time + # that a set calls its `.disjoint()` method, which may be called hundreds of + # times when scanning a page of links for packages with tags matching that + # Set[Tag]. Pre-computing the value here produces significant speedups for + # downstream consumers. + self._hash = hash((self._interpreter, self._abi, self._platform)) + + @property + def interpreter(self) -> str: + return self._interpreter + + @property + def abi(self) -> str: + return self._abi + + @property + def platform(self) -> str: + return self._platform + + def __eq__(self, other: object) -> bool: + if not isinstance(other, Tag): + return NotImplemented + + return ( + (self._hash == other._hash) # Short-circuit ASAP for perf reasons. + and (self._platform == other._platform) + and (self._abi == other._abi) + and (self._interpreter == other._interpreter) + ) + + def __hash__(self) -> int: + return self._hash + + def __str__(self) -> str: + return f"{self._interpreter}-{self._abi}-{self._platform}" + + def __repr__(self) -> str: + return f"<{self} @ {id(self)}>" + + +def parse_tag(tag: str) -> FrozenSet[Tag]: + """ + Parses the provided tag (e.g. `py3-none-any`) into a frozenset of Tag instances. + + Returning a set is required due to the possibility that the tag is a + compressed tag set. + """ + tags = set() + interpreters, abis, platforms = tag.split("-") + for interpreter in interpreters.split("."): + for abi in abis.split("."): + for platform_ in platforms.split("."): + tags.add(Tag(interpreter, abi, platform_)) + return frozenset(tags) + + +def _get_config_var(name: str, warn: bool = False) -> Union[int, str, None]: + value: Union[int, str, None] = sysconfig.get_config_var(name) + if value is None and warn: + logger.debug( + "Config variable '%s' is unset, Python ABI tag may be incorrect", name + ) + return value + + +def _normalize_string(string: str) -> str: + return string.replace(".", "_").replace("-", "_").replace(" ", "_") + + +def _abi3_applies(python_version: PythonVersion) -> bool: + """ + Determine if the Python version supports abi3. + + PEP 384 was first implemented in Python 3.2. + """ + return len(python_version) > 1 and tuple(python_version) >= (3, 2) + + +def _cpython_abis(py_version: PythonVersion, warn: bool = False) -> List[str]: + py_version = tuple(py_version) # To allow for version comparison. + abis = [] + version = _version_nodot(py_version[:2]) + debug = pymalloc = ucs4 = "" + with_debug = _get_config_var("Py_DEBUG", warn) + has_refcount = hasattr(sys, "gettotalrefcount") + # Windows doesn't set Py_DEBUG, so checking for support of debug-compiled + # extension modules is the best option. + # https://github.com/pypa/pip/issues/3383#issuecomment-173267692 + has_ext = "_d.pyd" in EXTENSION_SUFFIXES + if with_debug or (with_debug is None and (has_refcount or has_ext)): + debug = "d" + if py_version < (3, 8): + with_pymalloc = _get_config_var("WITH_PYMALLOC", warn) + if with_pymalloc or with_pymalloc is None: + pymalloc = "m" + if py_version < (3, 3): + unicode_size = _get_config_var("Py_UNICODE_SIZE", warn) + if unicode_size == 4 or ( + unicode_size is None and sys.maxunicode == 0x10FFFF + ): + ucs4 = "u" + elif debug: + # Debug builds can also load "normal" extension modules. + # We can also assume no UCS-4 or pymalloc requirement. + abis.append(f"cp{version}") + abis.insert( + 0, + "cp{version}{debug}{pymalloc}{ucs4}".format( + version=version, debug=debug, pymalloc=pymalloc, ucs4=ucs4 + ), + ) + return abis + + +def cpython_tags( + python_version: Optional[PythonVersion] = None, + abis: Optional[Iterable[str]] = None, + platforms: Optional[Iterable[str]] = None, + *, + warn: bool = False, +) -> Iterator[Tag]: + """ + Yields the tags for a CPython interpreter. + + The tags consist of: + - cp<python_version>-<abi>-<platform> + - cp<python_version>-abi3-<platform> + - cp<python_version>-none-<platform> + - cp<less than python_version>-abi3-<platform> # Older Python versions down to 3.2. + + If python_version only specifies a major version then user-provided ABIs and + the 'none' ABItag will be used. + + If 'abi3' or 'none' are specified in 'abis' then they will be yielded at + their normal position and not at the beginning. + """ + if not python_version: + python_version = sys.version_info[:2] + + interpreter = f"cp{_version_nodot(python_version[:2])}" + + if abis is None: + if len(python_version) > 1: + abis = _cpython_abis(python_version, warn) + else: + abis = [] + abis = list(abis) + # 'abi3' and 'none' are explicitly handled later. + for explicit_abi in ("abi3", "none"): + try: + abis.remove(explicit_abi) + except ValueError: + pass + + platforms = list(platforms or platform_tags()) + for abi in abis: + for platform_ in platforms: + yield Tag(interpreter, abi, platform_) + if _abi3_applies(python_version): + yield from (Tag(interpreter, "abi3", platform_) for platform_ in platforms) + yield from (Tag(interpreter, "none", platform_) for platform_ in platforms) + + if _abi3_applies(python_version): + for minor_version in range(python_version[1] - 1, 1, -1): + for platform_ in platforms: + interpreter = "cp{version}".format( + version=_version_nodot((python_version[0], minor_version)) + ) + yield Tag(interpreter, "abi3", platform_) + + +def _generic_abi() -> List[str]: + """ + Return the ABI tag based on EXT_SUFFIX. + """ + # The following are examples of `EXT_SUFFIX`. + # We want to keep the parts which are related to the ABI and remove the + # parts which are related to the platform: + # - linux: '.cpython-310-x86_64-linux-gnu.so' => cp310 + # - mac: '.cpython-310-darwin.so' => cp310 + # - win: '.cp310-win_amd64.pyd' => cp310 + # - win: '.pyd' => cp37 (uses _cpython_abis()) + # - pypy: '.pypy38-pp73-x86_64-linux-gnu.so' => pypy38_pp73 + # - graalpy: '.graalpy-38-native-x86_64-darwin.dylib' + # => graalpy_38_native + + ext_suffix = _get_config_var("EXT_SUFFIX", warn=True) + if not isinstance(ext_suffix, str) or ext_suffix[0] != ".": + raise SystemError("invalid sysconfig.get_config_var('EXT_SUFFIX')") + parts = ext_suffix.split(".") + if len(parts) < 3: + # CPython3.7 and earlier uses ".pyd" on Windows. + return _cpython_abis(sys.version_info[:2]) + soabi = parts[1] + if soabi.startswith("cpython"): + # non-windows + abi = "cp" + soabi.split("-")[1] + elif soabi.startswith("cp"): + # windows + abi = soabi.split("-")[0] + elif soabi.startswith("pypy"): + abi = "-".join(soabi.split("-")[:2]) + elif soabi.startswith("graalpy"): + abi = "-".join(soabi.split("-")[:3]) + elif soabi: + # pyston, ironpython, others? + abi = soabi + else: + return [] + return [_normalize_string(abi)] + + +def generic_tags( + interpreter: Optional[str] = None, + abis: Optional[Iterable[str]] = None, + platforms: Optional[Iterable[str]] = None, + *, + warn: bool = False, +) -> Iterator[Tag]: + """ + Yields the tags for a generic interpreter. + + The tags consist of: + - <interpreter>-<abi>-<platform> + + The "none" ABI will be added if it was not explicitly provided. + """ + if not interpreter: + interp_name = interpreter_name() + interp_version = interpreter_version(warn=warn) + interpreter = "".join([interp_name, interp_version]) + if abis is None: + abis = _generic_abi() + else: + abis = list(abis) + platforms = list(platforms or platform_tags()) + if "none" not in abis: + abis.append("none") + for abi in abis: + for platform_ in platforms: + yield Tag(interpreter, abi, platform_) + + +def _py_interpreter_range(py_version: PythonVersion) -> Iterator[str]: + """ + Yields Python versions in descending order. + + After the latest version, the major-only version will be yielded, and then + all previous versions of that major version. + """ + if len(py_version) > 1: + yield f"py{_version_nodot(py_version[:2])}" + yield f"py{py_version[0]}" + if len(py_version) > 1: + for minor in range(py_version[1] - 1, -1, -1): + yield f"py{_version_nodot((py_version[0], minor))}" + + +def compatible_tags( + python_version: Optional[PythonVersion] = None, + interpreter: Optional[str] = None, + platforms: Optional[Iterable[str]] = None, +) -> Iterator[Tag]: + """ + Yields the sequence of tags that are compatible with a specific version of Python. + + The tags consist of: + - py*-none-<platform> + - <interpreter>-none-any # ... if `interpreter` is provided. + - py*-none-any + """ + if not python_version: + python_version = sys.version_info[:2] + platforms = list(platforms or platform_tags()) + for version in _py_interpreter_range(python_version): + for platform_ in platforms: + yield Tag(version, "none", platform_) + if interpreter: + yield Tag(interpreter, "none", "any") + for version in _py_interpreter_range(python_version): + yield Tag(version, "none", "any") + + +def _mac_arch(arch: str, is_32bit: bool = _32_BIT_INTERPRETER) -> str: + if not is_32bit: + return arch + + if arch.startswith("ppc"): + return "ppc" + + return "i386" + + +def _mac_binary_formats(version: MacVersion, cpu_arch: str) -> List[str]: + formats = [cpu_arch] + if cpu_arch == "x86_64": + if version < (10, 4): + return [] + formats.extend(["intel", "fat64", "fat32"]) + + elif cpu_arch == "i386": + if version < (10, 4): + return [] + formats.extend(["intel", "fat32", "fat"]) + + elif cpu_arch == "ppc64": + # TODO: Need to care about 32-bit PPC for ppc64 through 10.2? + if version > (10, 5) or version < (10, 4): + return [] + formats.append("fat64") + + elif cpu_arch == "ppc": + if version > (10, 6): + return [] + formats.extend(["fat32", "fat"]) + + if cpu_arch in {"arm64", "x86_64"}: + formats.append("universal2") + + if cpu_arch in {"x86_64", "i386", "ppc64", "ppc", "intel"}: + formats.append("universal") + + return formats + + +def mac_platforms( + version: Optional[MacVersion] = None, arch: Optional[str] = None +) -> Iterator[str]: + """ + Yields the platform tags for a macOS system. + + The `version` parameter is a two-item tuple specifying the macOS version to + generate platform tags for. The `arch` parameter is the CPU architecture to + generate platform tags for. Both parameters default to the appropriate value + for the current system. + """ + version_str, _, cpu_arch = platform.mac_ver() + if version is None: + version = cast("MacVersion", tuple(map(int, version_str.split(".")[:2]))) + if version == (10, 16): + # When built against an older macOS SDK, Python will report macOS 10.16 + # instead of the real version. + version_str = subprocess.run( + [ + sys.executable, + "-sS", + "-c", + "import platform; print(platform.mac_ver()[0])", + ], + check=True, + env={"SYSTEM_VERSION_COMPAT": "0"}, + stdout=subprocess.PIPE, + universal_newlines=True, + ).stdout + version = cast("MacVersion", tuple(map(int, version_str.split(".")[:2]))) + else: + version = version + if arch is None: + arch = _mac_arch(cpu_arch) + else: + arch = arch + + if (10, 0) <= version and version < (11, 0): + # Prior to Mac OS 11, each yearly release of Mac OS bumped the + # "minor" version number. The major version was always 10. + for minor_version in range(version[1], -1, -1): + compat_version = 10, minor_version + binary_formats = _mac_binary_formats(compat_version, arch) + for binary_format in binary_formats: + yield "macosx_{major}_{minor}_{binary_format}".format( + major=10, minor=minor_version, binary_format=binary_format + ) + + if version >= (11, 0): + # Starting with Mac OS 11, each yearly release bumps the major version + # number. The minor versions are now the midyear updates. + for major_version in range(version[0], 10, -1): + compat_version = major_version, 0 + binary_formats = _mac_binary_formats(compat_version, arch) + for binary_format in binary_formats: + yield "macosx_{major}_{minor}_{binary_format}".format( + major=major_version, minor=0, binary_format=binary_format + ) + + if version >= (11, 0): + # Mac OS 11 on x86_64 is compatible with binaries from previous releases. + # Arm64 support was introduced in 11.0, so no Arm binaries from previous + # releases exist. + # + # However, the "universal2" binary format can have a + # macOS version earlier than 11.0 when the x86_64 part of the binary supports + # that version of macOS. + if arch == "x86_64": + for minor_version in range(16, 3, -1): + compat_version = 10, minor_version + binary_formats = _mac_binary_formats(compat_version, arch) + for binary_format in binary_formats: + yield "macosx_{major}_{minor}_{binary_format}".format( + major=compat_version[0], + minor=compat_version[1], + binary_format=binary_format, + ) + else: + for minor_version in range(16, 3, -1): + compat_version = 10, minor_version + binary_format = "universal2" + yield "macosx_{major}_{minor}_{binary_format}".format( + major=compat_version[0], + minor=compat_version[1], + binary_format=binary_format, + ) + + +def _linux_platforms(is_32bit: bool = _32_BIT_INTERPRETER) -> Iterator[str]: + linux = _normalize_string(sysconfig.get_platform()) + if is_32bit: + if linux == "linux_x86_64": + linux = "linux_i686" + elif linux == "linux_aarch64": + linux = "linux_armv7l" + _, arch = linux.split("_", 1) + yield from _manylinux.platform_tags(linux, arch) + yield from _musllinux.platform_tags(arch) + yield linux + + +def _generic_platforms() -> Iterator[str]: + yield _normalize_string(sysconfig.get_platform()) + + +def platform_tags() -> Iterator[str]: + """ + Provides the platform tags for this installation. + """ + if platform.system() == "Darwin": + return mac_platforms() + elif platform.system() == "Linux": + return _linux_platforms() + else: + return _generic_platforms() + + +def interpreter_name() -> str: + """ + Returns the name of the running interpreter. + + Some implementations have a reserved, two-letter abbreviation which will + be returned when appropriate. + """ + name = sys.implementation.name + return INTERPRETER_SHORT_NAMES.get(name) or name + + +def interpreter_version(*, warn: bool = False) -> str: + """ + Returns the version of the running interpreter. + """ + version = _get_config_var("py_version_nodot", warn=warn) + if version: + version = str(version) + else: + version = _version_nodot(sys.version_info[:2]) + return version + + +def _version_nodot(version: PythonVersion) -> str: + return "".join(map(str, version)) + + +def sys_tags(*, warn: bool = False) -> Iterator[Tag]: + """ + Returns the sequence of tag triples for the running interpreter. + + The order of the sequence corresponds to priority order for the + interpreter, from most to least important. + """ + + interp_name = interpreter_name() + if interp_name == "cp": + yield from cpython_tags(warn=warn) + else: + yield from generic_tags() + + if interp_name == "pp": + interp = "pp3" + elif interp_name == "cp": + interp = "cp" + interpreter_version(warn=warn) + else: + interp = None + yield from compatible_tags(interpreter=interp) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/utils.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/utils.py new file mode 100644 index 0000000000..33c613b749 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/utils.py @@ -0,0 +1,141 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. + +import re +from typing import FrozenSet, NewType, Tuple, Union, cast + +from .tags import Tag, parse_tag +from .version import InvalidVersion, Version + +BuildTag = Union[Tuple[()], Tuple[int, str]] +NormalizedName = NewType("NormalizedName", str) + + +class InvalidWheelFilename(ValueError): + """ + An invalid wheel filename was found, users should refer to PEP 427. + """ + + +class InvalidSdistFilename(ValueError): + """ + An invalid sdist filename was found, users should refer to the packaging user guide. + """ + + +_canonicalize_regex = re.compile(r"[-_.]+") +# PEP 427: The build number must start with a digit. +_build_tag_regex = re.compile(r"(\d+)(.*)") + + +def canonicalize_name(name: str) -> NormalizedName: + # This is taken from PEP 503. + value = _canonicalize_regex.sub("-", name).lower() + return cast(NormalizedName, value) + + +def canonicalize_version( + version: Union[Version, str], *, strip_trailing_zero: bool = True +) -> str: + """ + This is very similar to Version.__str__, but has one subtle difference + with the way it handles the release segment. + """ + if isinstance(version, str): + try: + parsed = Version(version) + except InvalidVersion: + # Legacy versions cannot be normalized + return version + else: + parsed = version + + parts = [] + + # Epoch + if parsed.epoch != 0: + parts.append(f"{parsed.epoch}!") + + # Release segment + release_segment = ".".join(str(x) for x in parsed.release) + if strip_trailing_zero: + # NB: This strips trailing '.0's to normalize + release_segment = re.sub(r"(\.0)+$", "", release_segment) + parts.append(release_segment) + + # Pre-release + if parsed.pre is not None: + parts.append("".join(str(x) for x in parsed.pre)) + + # Post-release + if parsed.post is not None: + parts.append(f".post{parsed.post}") + + # Development release + if parsed.dev is not None: + parts.append(f".dev{parsed.dev}") + + # Local version segment + if parsed.local is not None: + parts.append(f"+{parsed.local}") + + return "".join(parts) + + +def parse_wheel_filename( + filename: str, +) -> Tuple[NormalizedName, Version, BuildTag, FrozenSet[Tag]]: + if not filename.endswith(".whl"): + raise InvalidWheelFilename( + f"Invalid wheel filename (extension must be '.whl'): {filename}" + ) + + filename = filename[:-4] + dashes = filename.count("-") + if dashes not in (4, 5): + raise InvalidWheelFilename( + f"Invalid wheel filename (wrong number of parts): {filename}" + ) + + parts = filename.split("-", dashes - 2) + name_part = parts[0] + # See PEP 427 for the rules on escaping the project name + if "__" in name_part or re.match(r"^[\w\d._]*$", name_part, re.UNICODE) is None: + raise InvalidWheelFilename(f"Invalid project name: {filename}") + name = canonicalize_name(name_part) + version = Version(parts[1]) + if dashes == 5: + build_part = parts[2] + build_match = _build_tag_regex.match(build_part) + if build_match is None: + raise InvalidWheelFilename( + f"Invalid build number: {build_part} in '{filename}'" + ) + build = cast(BuildTag, (int(build_match.group(1)), build_match.group(2))) + else: + build = () + tags = parse_tag(parts[-1]) + return (name, version, build, tags) + + +def parse_sdist_filename(filename: str) -> Tuple[NormalizedName, Version]: + if filename.endswith(".tar.gz"): + file_stem = filename[: -len(".tar.gz")] + elif filename.endswith(".zip"): + file_stem = filename[: -len(".zip")] + else: + raise InvalidSdistFilename( + f"Invalid sdist filename (extension must be '.tar.gz' or '.zip'):" + f" {filename}" + ) + + # We are requiring a PEP 440 version, which cannot contain dashes, + # so we split on the last dash. + name_part, sep, version_part = file_stem.rpartition("-") + if not sep: + raise InvalidSdistFilename(f"Invalid sdist filename: {filename}") + + name = canonicalize_name(name_part) + version = Version(version_part) + return (name, version) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/packaging/version.py b/third_party/python/setuptools/pkg_resources/_vendor/packaging/version.py new file mode 100644 index 0000000000..b30e8cbf84 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/packaging/version.py @@ -0,0 +1,564 @@ +# This file is dual licensed under the terms of the Apache License, Version +# 2.0, and the BSD License. See the LICENSE file in the root of this repository +# for complete details. +""" +.. testsetup:: + + from packaging.version import parse, Version +""" + +import collections +import itertools +import re +from typing import Any, Callable, Optional, SupportsInt, Tuple, Union + +from ._structures import Infinity, InfinityType, NegativeInfinity, NegativeInfinityType + +__all__ = ["VERSION_PATTERN", "parse", "Version", "InvalidVersion"] + +InfiniteTypes = Union[InfinityType, NegativeInfinityType] +PrePostDevType = Union[InfiniteTypes, Tuple[str, int]] +SubLocalType = Union[InfiniteTypes, int, str] +LocalType = Union[ + NegativeInfinityType, + Tuple[ + Union[ + SubLocalType, + Tuple[SubLocalType, str], + Tuple[NegativeInfinityType, SubLocalType], + ], + ..., + ], +] +CmpKey = Tuple[ + int, Tuple[int, ...], PrePostDevType, PrePostDevType, PrePostDevType, LocalType +] +VersionComparisonMethod = Callable[[CmpKey, CmpKey], bool] + +_Version = collections.namedtuple( + "_Version", ["epoch", "release", "dev", "pre", "post", "local"] +) + + +def parse(version: str) -> "Version": + """Parse the given version string. + + >>> parse('1.0.dev1') + <Version('1.0.dev1')> + + :param version: The version string to parse. + :raises InvalidVersion: When the version string is not a valid version. + """ + return Version(version) + + +class InvalidVersion(ValueError): + """Raised when a version string is not a valid version. + + >>> Version("invalid") + Traceback (most recent call last): + ... + packaging.version.InvalidVersion: Invalid version: 'invalid' + """ + + +class _BaseVersion: + _key: Tuple[Any, ...] + + def __hash__(self) -> int: + return hash(self._key) + + # Please keep the duplicated `isinstance` check + # in the six comparisons hereunder + # unless you find a way to avoid adding overhead function calls. + def __lt__(self, other: "_BaseVersion") -> bool: + if not isinstance(other, _BaseVersion): + return NotImplemented + + return self._key < other._key + + def __le__(self, other: "_BaseVersion") -> bool: + if not isinstance(other, _BaseVersion): + return NotImplemented + + return self._key <= other._key + + def __eq__(self, other: object) -> bool: + if not isinstance(other, _BaseVersion): + return NotImplemented + + return self._key == other._key + + def __ge__(self, other: "_BaseVersion") -> bool: + if not isinstance(other, _BaseVersion): + return NotImplemented + + return self._key >= other._key + + def __gt__(self, other: "_BaseVersion") -> bool: + if not isinstance(other, _BaseVersion): + return NotImplemented + + return self._key > other._key + + def __ne__(self, other: object) -> bool: + if not isinstance(other, _BaseVersion): + return NotImplemented + + return self._key != other._key + + +# Deliberately not anchored to the start and end of the string, to make it +# easier for 3rd party code to reuse +_VERSION_PATTERN = r""" + v? + (?: + (?:(?P<epoch>[0-9]+)!)? # epoch + (?P<release>[0-9]+(?:\.[0-9]+)*) # release segment + (?P<pre> # pre-release + [-_\.]? + (?P<pre_l>(a|b|c|rc|alpha|beta|pre|preview)) + [-_\.]? + (?P<pre_n>[0-9]+)? + )? + (?P<post> # post release + (?:-(?P<post_n1>[0-9]+)) + | + (?: + [-_\.]? + (?P<post_l>post|rev|r) + [-_\.]? + (?P<post_n2>[0-9]+)? + ) + )? + (?P<dev> # dev release + [-_\.]? + (?P<dev_l>dev) + [-_\.]? + (?P<dev_n>[0-9]+)? + )? + ) + (?:\+(?P<local>[a-z0-9]+(?:[-_\.][a-z0-9]+)*))? # local version +""" + +VERSION_PATTERN = _VERSION_PATTERN +""" +A string containing the regular expression used to match a valid version. + +The pattern is not anchored at either end, and is intended for embedding in larger +expressions (for example, matching a version number as part of a file name). The +regular expression should be compiled with the ``re.VERBOSE`` and ``re.IGNORECASE`` +flags set. + +:meta hide-value: +""" + + +class Version(_BaseVersion): + """This class abstracts handling of a project's versions. + + A :class:`Version` instance is comparison aware and can be compared and + sorted using the standard Python interfaces. + + >>> v1 = Version("1.0a5") + >>> v2 = Version("1.0") + >>> v1 + <Version('1.0a5')> + >>> v2 + <Version('1.0')> + >>> v1 < v2 + True + >>> v1 == v2 + False + >>> v1 > v2 + False + >>> v1 >= v2 + False + >>> v1 <= v2 + True + """ + + _regex = re.compile(r"^\s*" + VERSION_PATTERN + r"\s*$", re.VERBOSE | re.IGNORECASE) + _key: CmpKey + + def __init__(self, version: str) -> None: + """Initialize a Version object. + + :param version: + The string representation of a version which will be parsed and normalized + before use. + :raises InvalidVersion: + If the ``version`` does not conform to PEP 440 in any way then this + exception will be raised. + """ + + # Validate the version and parse it into pieces + match = self._regex.search(version) + if not match: + raise InvalidVersion(f"Invalid version: '{version}'") + + # Store the parsed out pieces of the version + self._version = _Version( + epoch=int(match.group("epoch")) if match.group("epoch") else 0, + release=tuple(int(i) for i in match.group("release").split(".")), + pre=_parse_letter_version(match.group("pre_l"), match.group("pre_n")), + post=_parse_letter_version( + match.group("post_l"), match.group("post_n1") or match.group("post_n2") + ), + dev=_parse_letter_version(match.group("dev_l"), match.group("dev_n")), + local=_parse_local_version(match.group("local")), + ) + + # Generate a key which will be used for sorting + self._key = _cmpkey( + self._version.epoch, + self._version.release, + self._version.pre, + self._version.post, + self._version.dev, + self._version.local, + ) + + def __repr__(self) -> str: + """A representation of the Version that shows all internal state. + + >>> Version('1.0.0') + <Version('1.0.0')> + """ + return f"<Version('{self}')>" + + def __str__(self) -> str: + """A string representation of the version that can be rounded-tripped. + + >>> str(Version("1.0a5")) + '1.0a5' + """ + parts = [] + + # Epoch + if self.epoch != 0: + parts.append(f"{self.epoch}!") + + # Release segment + parts.append(".".join(str(x) for x in self.release)) + + # Pre-release + if self.pre is not None: + parts.append("".join(str(x) for x in self.pre)) + + # Post-release + if self.post is not None: + parts.append(f".post{self.post}") + + # Development release + if self.dev is not None: + parts.append(f".dev{self.dev}") + + # Local version segment + if self.local is not None: + parts.append(f"+{self.local}") + + return "".join(parts) + + @property + def epoch(self) -> int: + """The epoch of the version. + + >>> Version("2.0.0").epoch + 0 + >>> Version("1!2.0.0").epoch + 1 + """ + _epoch: int = self._version.epoch + return _epoch + + @property + def release(self) -> Tuple[int, ...]: + """The components of the "release" segment of the version. + + >>> Version("1.2.3").release + (1, 2, 3) + >>> Version("2.0.0").release + (2, 0, 0) + >>> Version("1!2.0.0.post0").release + (2, 0, 0) + + Includes trailing zeroes but not the epoch or any pre-release / development / + post-release suffixes. + """ + _release: Tuple[int, ...] = self._version.release + return _release + + @property + def pre(self) -> Optional[Tuple[str, int]]: + """The pre-release segment of the version. + + >>> print(Version("1.2.3").pre) + None + >>> Version("1.2.3a1").pre + ('a', 1) + >>> Version("1.2.3b1").pre + ('b', 1) + >>> Version("1.2.3rc1").pre + ('rc', 1) + """ + _pre: Optional[Tuple[str, int]] = self._version.pre + return _pre + + @property + def post(self) -> Optional[int]: + """The post-release number of the version. + + >>> print(Version("1.2.3").post) + None + >>> Version("1.2.3.post1").post + 1 + """ + return self._version.post[1] if self._version.post else None + + @property + def dev(self) -> Optional[int]: + """The development number of the version. + + >>> print(Version("1.2.3").dev) + None + >>> Version("1.2.3.dev1").dev + 1 + """ + return self._version.dev[1] if self._version.dev else None + + @property + def local(self) -> Optional[str]: + """The local version segment of the version. + + >>> print(Version("1.2.3").local) + None + >>> Version("1.2.3+abc").local + 'abc' + """ + if self._version.local: + return ".".join(str(x) for x in self._version.local) + else: + return None + + @property + def public(self) -> str: + """The public portion of the version. + + >>> Version("1.2.3").public + '1.2.3' + >>> Version("1.2.3+abc").public + '1.2.3' + >>> Version("1.2.3+abc.dev1").public + '1.2.3' + """ + return str(self).split("+", 1)[0] + + @property + def base_version(self) -> str: + """The "base version" of the version. + + >>> Version("1.2.3").base_version + '1.2.3' + >>> Version("1.2.3+abc").base_version + '1.2.3' + >>> Version("1!1.2.3+abc.dev1").base_version + '1!1.2.3' + + The "base version" is the public version of the project without any pre or post + release markers. + """ + parts = [] + + # Epoch + if self.epoch != 0: + parts.append(f"{self.epoch}!") + + # Release segment + parts.append(".".join(str(x) for x in self.release)) + + return "".join(parts) + + @property + def is_prerelease(self) -> bool: + """Whether this version is a pre-release. + + >>> Version("1.2.3").is_prerelease + False + >>> Version("1.2.3a1").is_prerelease + True + >>> Version("1.2.3b1").is_prerelease + True + >>> Version("1.2.3rc1").is_prerelease + True + >>> Version("1.2.3dev1").is_prerelease + True + """ + return self.dev is not None or self.pre is not None + + @property + def is_postrelease(self) -> bool: + """Whether this version is a post-release. + + >>> Version("1.2.3").is_postrelease + False + >>> Version("1.2.3.post1").is_postrelease + True + """ + return self.post is not None + + @property + def is_devrelease(self) -> bool: + """Whether this version is a development release. + + >>> Version("1.2.3").is_devrelease + False + >>> Version("1.2.3.dev1").is_devrelease + True + """ + return self.dev is not None + + @property + def major(self) -> int: + """The first item of :attr:`release` or ``0`` if unavailable. + + >>> Version("1.2.3").major + 1 + """ + return self.release[0] if len(self.release) >= 1 else 0 + + @property + def minor(self) -> int: + """The second item of :attr:`release` or ``0`` if unavailable. + + >>> Version("1.2.3").minor + 2 + >>> Version("1").minor + 0 + """ + return self.release[1] if len(self.release) >= 2 else 0 + + @property + def micro(self) -> int: + """The third item of :attr:`release` or ``0`` if unavailable. + + >>> Version("1.2.3").micro + 3 + >>> Version("1").micro + 0 + """ + return self.release[2] if len(self.release) >= 3 else 0 + + +def _parse_letter_version( + letter: str, number: Union[str, bytes, SupportsInt] +) -> Optional[Tuple[str, int]]: + + if letter: + # We consider there to be an implicit 0 in a pre-release if there is + # not a numeral associated with it. + if number is None: + number = 0 + + # We normalize any letters to their lower case form + letter = letter.lower() + + # We consider some words to be alternate spellings of other words and + # in those cases we want to normalize the spellings to our preferred + # spelling. + if letter == "alpha": + letter = "a" + elif letter == "beta": + letter = "b" + elif letter in ["c", "pre", "preview"]: + letter = "rc" + elif letter in ["rev", "r"]: + letter = "post" + + return letter, int(number) + if not letter and number: + # We assume if we are given a number, but we are not given a letter + # then this is using the implicit post release syntax (e.g. 1.0-1) + letter = "post" + + return letter, int(number) + + return None + + +_local_version_separators = re.compile(r"[\._-]") + + +def _parse_local_version(local: str) -> Optional[LocalType]: + """ + Takes a string like abc.1.twelve and turns it into ("abc", 1, "twelve"). + """ + if local is not None: + return tuple( + part.lower() if not part.isdigit() else int(part) + for part in _local_version_separators.split(local) + ) + return None + + +def _cmpkey( + epoch: int, + release: Tuple[int, ...], + pre: Optional[Tuple[str, int]], + post: Optional[Tuple[str, int]], + dev: Optional[Tuple[str, int]], + local: Optional[Tuple[SubLocalType]], +) -> CmpKey: + + # When we compare a release version, we want to compare it with all of the + # trailing zeros removed. So we'll use a reverse the list, drop all the now + # leading zeros until we come to something non zero, then take the rest + # re-reverse it back into the correct order and make it a tuple and use + # that for our sorting key. + _release = tuple( + reversed(list(itertools.dropwhile(lambda x: x == 0, reversed(release)))) + ) + + # We need to "trick" the sorting algorithm to put 1.0.dev0 before 1.0a0. + # We'll do this by abusing the pre segment, but we _only_ want to do this + # if there is not a pre or a post segment. If we have one of those then + # the normal sorting rules will handle this case correctly. + if pre is None and post is None and dev is not None: + _pre: PrePostDevType = NegativeInfinity + # Versions without a pre-release (except as noted above) should sort after + # those with one. + elif pre is None: + _pre = Infinity + else: + _pre = pre + + # Versions without a post segment should sort before those with one. + if post is None: + _post: PrePostDevType = NegativeInfinity + + else: + _post = post + + # Versions without a development segment should sort after those with one. + if dev is None: + _dev: PrePostDevType = Infinity + + else: + _dev = dev + + if local is None: + # Versions without a local segment should sort before those with one. + _local: LocalType = NegativeInfinity + else: + # Versions with a local segment need that segment parsed to implement + # the sorting rules in PEP440. + # - Alpha numeric segments sort before numeric segments + # - Alpha numeric segments sort lexicographically + # - Numeric segments sort numerically + # - Shorter versions sort before longer versions when the prefixes + # match exactly + _local = tuple( + (i, "") if isinstance(i, int) else (NegativeInfinity, i) for i in local + ) + + return epoch, _release, _pre, _post, _dev, _local diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/__init__.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/__init__.py new file mode 100644 index 0000000000..aef2821b83 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/__init__.py @@ -0,0 +1,342 @@ +""" +Utilities for determining application-specific dirs. See <https://github.com/platformdirs/platformdirs> for details and +usage. +""" +from __future__ import annotations + +import os +import sys +from pathlib import Path + +if sys.version_info >= (3, 8): # pragma: no cover (py38+) + from typing import Literal +else: # pragma: no cover (py38+) + from ..typing_extensions import Literal + +from .api import PlatformDirsABC +from .version import __version__ +from .version import __version_tuple__ as __version_info__ + + +def _set_platform_dir_class() -> type[PlatformDirsABC]: + if sys.platform == "win32": + from .windows import Windows as Result + elif sys.platform == "darwin": + from .macos import MacOS as Result + else: + from .unix import Unix as Result + + if os.getenv("ANDROID_DATA") == "/data" and os.getenv("ANDROID_ROOT") == "/system": + + if os.getenv("SHELL") or os.getenv("PREFIX"): + return Result + + from .android import _android_folder + + if _android_folder() is not None: + from .android import Android + + return Android # return to avoid redefinition of result + + return Result + + +PlatformDirs = _set_platform_dir_class() #: Currently active platform +AppDirs = PlatformDirs #: Backwards compatibility with appdirs + + +def user_data_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param roaming: See `roaming <platformdirs.api.PlatformDirsABC.version>`. + :returns: data directory tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, roaming=roaming).user_data_dir + + +def site_data_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + multipath: bool = False, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param multipath: See `roaming <platformdirs.api.PlatformDirsABC.multipath>`. + :returns: data directory shared by users + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, multipath=multipath).site_data_dir + + +def user_config_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param roaming: See `roaming <platformdirs.api.PlatformDirsABC.version>`. + :returns: config directory tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, roaming=roaming).user_config_dir + + +def site_config_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + multipath: bool = False, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param multipath: See `roaming <platformdirs.api.PlatformDirsABC.multipath>`. + :returns: config directory shared by the users + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, multipath=multipath).site_config_dir + + +def user_cache_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + opinion: bool = True, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param opinion: See `roaming <platformdirs.api.PlatformDirsABC.opinion>`. + :returns: cache directory tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, opinion=opinion).user_cache_dir + + +def user_state_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param roaming: See `roaming <platformdirs.api.PlatformDirsABC.version>`. + :returns: state directory tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, roaming=roaming).user_state_dir + + +def user_log_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + opinion: bool = True, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param opinion: See `roaming <platformdirs.api.PlatformDirsABC.opinion>`. + :returns: log directory tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, opinion=opinion).user_log_dir + + +def user_documents_dir() -> str: + """ + :returns: documents directory tied to the user + """ + return PlatformDirs().user_documents_dir + + +def user_runtime_dir( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + opinion: bool = True, +) -> str: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param opinion: See `opinion <platformdirs.api.PlatformDirsABC.opinion>`. + :returns: runtime directory tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, opinion=opinion).user_runtime_dir + + +def user_data_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param roaming: See `roaming <platformdirs.api.PlatformDirsABC.version>`. + :returns: data path tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, roaming=roaming).user_data_path + + +def site_data_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + multipath: bool = False, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param multipath: See `multipath <platformdirs.api.PlatformDirsABC.multipath>`. + :returns: data path shared by users + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, multipath=multipath).site_data_path + + +def user_config_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param roaming: See `roaming <platformdirs.api.PlatformDirsABC.version>`. + :returns: config path tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, roaming=roaming).user_config_path + + +def site_config_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + multipath: bool = False, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param multipath: See `roaming <platformdirs.api.PlatformDirsABC.multipath>`. + :returns: config path shared by the users + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, multipath=multipath).site_config_path + + +def user_cache_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + opinion: bool = True, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param opinion: See `roaming <platformdirs.api.PlatformDirsABC.opinion>`. + :returns: cache path tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, opinion=opinion).user_cache_path + + +def user_state_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param roaming: See `roaming <platformdirs.api.PlatformDirsABC.version>`. + :returns: state path tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, roaming=roaming).user_state_path + + +def user_log_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + opinion: bool = True, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param opinion: See `roaming <platformdirs.api.PlatformDirsABC.opinion>`. + :returns: log path tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, opinion=opinion).user_log_path + + +def user_documents_path() -> Path: + """ + :returns: documents path tied to the user + """ + return PlatformDirs().user_documents_path + + +def user_runtime_path( + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + opinion: bool = True, +) -> Path: + """ + :param appname: See `appname <platformdirs.api.PlatformDirsABC.appname>`. + :param appauthor: See `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`. + :param version: See `version <platformdirs.api.PlatformDirsABC.version>`. + :param opinion: See `opinion <platformdirs.api.PlatformDirsABC.opinion>`. + :returns: runtime path tied to the user + """ + return PlatformDirs(appname=appname, appauthor=appauthor, version=version, opinion=opinion).user_runtime_path + + +__all__ = [ + "__version__", + "__version_info__", + "PlatformDirs", + "AppDirs", + "PlatformDirsABC", + "user_data_dir", + "user_config_dir", + "user_cache_dir", + "user_state_dir", + "user_log_dir", + "user_documents_dir", + "user_runtime_dir", + "site_data_dir", + "site_config_dir", + "user_data_path", + "user_config_path", + "user_cache_path", + "user_state_path", + "user_log_path", + "user_documents_path", + "user_runtime_path", + "site_data_path", + "site_config_path", +] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/__main__.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/__main__.py new file mode 100644 index 0000000000..0fc1edd59c --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/__main__.py @@ -0,0 +1,46 @@ +from __future__ import annotations + +from platformdirs import PlatformDirs, __version__ + +PROPS = ( + "user_data_dir", + "user_config_dir", + "user_cache_dir", + "user_state_dir", + "user_log_dir", + "user_documents_dir", + "user_runtime_dir", + "site_data_dir", + "site_config_dir", +) + + +def main() -> None: + app_name = "MyApp" + app_author = "MyCompany" + + print(f"-- platformdirs {__version__} --") + + print("-- app dirs (with optional 'version')") + dirs = PlatformDirs(app_name, app_author, version="1.0") + for prop in PROPS: + print(f"{prop}: {getattr(dirs, prop)}") + + print("\n-- app dirs (without optional 'version')") + dirs = PlatformDirs(app_name, app_author) + for prop in PROPS: + print(f"{prop}: {getattr(dirs, prop)}") + + print("\n-- app dirs (without optional 'appauthor')") + dirs = PlatformDirs(app_name) + for prop in PROPS: + print(f"{prop}: {getattr(dirs, prop)}") + + print("\n-- app dirs (with disabled 'appauthor')") + dirs = PlatformDirs(app_name, appauthor=False) + for prop in PROPS: + print(f"{prop}: {getattr(dirs, prop)}") + + +if __name__ == "__main__": + main() diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/android.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/android.py new file mode 100644 index 0000000000..eda8093512 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/android.py @@ -0,0 +1,120 @@ +from __future__ import annotations + +import os +import re +import sys +from functools import lru_cache +from typing import cast + +from .api import PlatformDirsABC + + +class Android(PlatformDirsABC): + """ + Follows the guidance `from here <https://android.stackexchange.com/a/216132>`_. Makes use of the + `appname <platformdirs.api.PlatformDirsABC.appname>` and + `version <platformdirs.api.PlatformDirsABC.version>`. + """ + + @property + def user_data_dir(self) -> str: + """:return: data directory tied to the user, e.g. ``/data/user/<userid>/<packagename>/files/<AppName>``""" + return self._append_app_name_and_version(cast(str, _android_folder()), "files") + + @property + def site_data_dir(self) -> str: + """:return: data directory shared by users, same as `user_data_dir`""" + return self.user_data_dir + + @property + def user_config_dir(self) -> str: + """ + :return: config directory tied to the user, e.g. ``/data/user/<userid>/<packagename>/shared_prefs/<AppName>`` + """ + return self._append_app_name_and_version(cast(str, _android_folder()), "shared_prefs") + + @property + def site_config_dir(self) -> str: + """:return: config directory shared by the users, same as `user_config_dir`""" + return self.user_config_dir + + @property + def user_cache_dir(self) -> str: + """:return: cache directory tied to the user, e.g. e.g. ``/data/user/<userid>/<packagename>/cache/<AppName>``""" + return self._append_app_name_and_version(cast(str, _android_folder()), "cache") + + @property + def user_state_dir(self) -> str: + """:return: state directory tied to the user, same as `user_data_dir`""" + return self.user_data_dir + + @property + def user_log_dir(self) -> str: + """ + :return: log directory tied to the user, same as `user_cache_dir` if not opinionated else ``log`` in it, + e.g. ``/data/user/<userid>/<packagename>/cache/<AppName>/log`` + """ + path = self.user_cache_dir + if self.opinion: + path = os.path.join(path, "log") + return path + + @property + def user_documents_dir(self) -> str: + """ + :return: documents directory tied to the user e.g. ``/storage/emulated/0/Documents`` + """ + return _android_documents_folder() + + @property + def user_runtime_dir(self) -> str: + """ + :return: runtime directory tied to the user, same as `user_cache_dir` if not opinionated else ``tmp`` in it, + e.g. ``/data/user/<userid>/<packagename>/cache/<AppName>/tmp`` + """ + path = self.user_cache_dir + if self.opinion: + path = os.path.join(path, "tmp") + return path + + +@lru_cache(maxsize=1) +def _android_folder() -> str | None: + """:return: base folder for the Android OS or None if cannot be found""" + try: + # First try to get path to android app via pyjnius + from jnius import autoclass + + Context = autoclass("android.content.Context") # noqa: N806 + result: str | None = Context.getFilesDir().getParentFile().getAbsolutePath() + except Exception: + # if fails find an android folder looking path on the sys.path + pattern = re.compile(r"/data/(data|user/\d+)/(.+)/files") + for path in sys.path: + if pattern.match(path): + result = path.split("/files")[0] + break + else: + result = None + return result + + +@lru_cache(maxsize=1) +def _android_documents_folder() -> str: + """:return: documents folder for the Android OS""" + # Get directories with pyjnius + try: + from jnius import autoclass + + Context = autoclass("android.content.Context") # noqa: N806 + Environment = autoclass("android.os.Environment") # noqa: N806 + documents_dir: str = Context.getExternalFilesDir(Environment.DIRECTORY_DOCUMENTS).getAbsolutePath() + except Exception: + documents_dir = "/storage/emulated/0/Documents" + + return documents_dir + + +__all__ = [ + "Android", +] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/api.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/api.py new file mode 100644 index 0000000000..6f6e2c2c69 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/api.py @@ -0,0 +1,156 @@ +from __future__ import annotations + +import os +import sys +from abc import ABC, abstractmethod +from pathlib import Path + +if sys.version_info >= (3, 8): # pragma: no branch + from typing import Literal # pragma: no cover + + +class PlatformDirsABC(ABC): + """ + Abstract base class for platform directories. + """ + + def __init__( + self, + appname: str | None = None, + appauthor: str | None | Literal[False] = None, + version: str | None = None, + roaming: bool = False, + multipath: bool = False, + opinion: bool = True, + ): + """ + Create a new platform directory. + + :param appname: See `appname`. + :param appauthor: See `appauthor`. + :param version: See `version`. + :param roaming: See `roaming`. + :param multipath: See `multipath`. + :param opinion: See `opinion`. + """ + self.appname = appname #: The name of application. + self.appauthor = appauthor + """ + The name of the app author or distributing body for this application. Typically, it is the owning company name. + Defaults to `appname`. You may pass ``False`` to disable it. + """ + self.version = version + """ + An optional version path element to append to the path. You might want to use this if you want multiple versions + of your app to be able to run independently. If used, this would typically be ``<major>.<minor>``. + """ + self.roaming = roaming + """ + Whether to use the roaming appdata directory on Windows. That means that for users on a Windows network setup + for roaming profiles, this user data will be synced on login (see + `here <http://technet.microsoft.com/en-us/library/cc766489(WS.10).aspx>`_). + """ + self.multipath = multipath + """ + An optional parameter only applicable to Unix/Linux which indicates that the entire list of data dirs should be + returned. By default, the first item would only be returned. + """ + self.opinion = opinion #: A flag to indicating to use opinionated values. + + def _append_app_name_and_version(self, *base: str) -> str: + params = list(base[1:]) + if self.appname: + params.append(self.appname) + if self.version: + params.append(self.version) + return os.path.join(base[0], *params) + + @property + @abstractmethod + def user_data_dir(self) -> str: + """:return: data directory tied to the user""" + + @property + @abstractmethod + def site_data_dir(self) -> str: + """:return: data directory shared by users""" + + @property + @abstractmethod + def user_config_dir(self) -> str: + """:return: config directory tied to the user""" + + @property + @abstractmethod + def site_config_dir(self) -> str: + """:return: config directory shared by the users""" + + @property + @abstractmethod + def user_cache_dir(self) -> str: + """:return: cache directory tied to the user""" + + @property + @abstractmethod + def user_state_dir(self) -> str: + """:return: state directory tied to the user""" + + @property + @abstractmethod + def user_log_dir(self) -> str: + """:return: log directory tied to the user""" + + @property + @abstractmethod + def user_documents_dir(self) -> str: + """:return: documents directory tied to the user""" + + @property + @abstractmethod + def user_runtime_dir(self) -> str: + """:return: runtime directory tied to the user""" + + @property + def user_data_path(self) -> Path: + """:return: data path tied to the user""" + return Path(self.user_data_dir) + + @property + def site_data_path(self) -> Path: + """:return: data path shared by users""" + return Path(self.site_data_dir) + + @property + def user_config_path(self) -> Path: + """:return: config path tied to the user""" + return Path(self.user_config_dir) + + @property + def site_config_path(self) -> Path: + """:return: config path shared by the users""" + return Path(self.site_config_dir) + + @property + def user_cache_path(self) -> Path: + """:return: cache path tied to the user""" + return Path(self.user_cache_dir) + + @property + def user_state_path(self) -> Path: + """:return: state path tied to the user""" + return Path(self.user_state_dir) + + @property + def user_log_path(self) -> Path: + """:return: log path tied to the user""" + return Path(self.user_log_dir) + + @property + def user_documents_path(self) -> Path: + """:return: documents path tied to the user""" + return Path(self.user_documents_dir) + + @property + def user_runtime_path(self) -> Path: + """:return: runtime path tied to the user""" + return Path(self.user_runtime_dir) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/macos.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/macos.py new file mode 100644 index 0000000000..a01337c776 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/macos.py @@ -0,0 +1,64 @@ +from __future__ import annotations + +import os + +from .api import PlatformDirsABC + + +class MacOS(PlatformDirsABC): + """ + Platform directories for the macOS operating system. Follows the guidance from `Apple documentation + <https://developer.apple.com/library/archive/documentation/FileManagement/Conceptual/FileSystemProgrammingGuide/MacOSXDirectories/MacOSXDirectories.html>`_. + Makes use of the `appname <platformdirs.api.PlatformDirsABC.appname>` and + `version <platformdirs.api.PlatformDirsABC.version>`. + """ + + @property + def user_data_dir(self) -> str: + """:return: data directory tied to the user, e.g. ``~/Library/Application Support/$appname/$version``""" + return self._append_app_name_and_version(os.path.expanduser("~/Library/Application Support/")) + + @property + def site_data_dir(self) -> str: + """:return: data directory shared by users, e.g. ``/Library/Application Support/$appname/$version``""" + return self._append_app_name_and_version("/Library/Application Support") + + @property + def user_config_dir(self) -> str: + """:return: config directory tied to the user, e.g. ``~/Library/Preferences/$appname/$version``""" + return self._append_app_name_and_version(os.path.expanduser("~/Library/Preferences/")) + + @property + def site_config_dir(self) -> str: + """:return: config directory shared by the users, e.g. ``/Library/Preferences/$appname``""" + return self._append_app_name_and_version("/Library/Preferences") + + @property + def user_cache_dir(self) -> str: + """:return: cache directory tied to the user, e.g. ``~/Library/Caches/$appname/$version``""" + return self._append_app_name_and_version(os.path.expanduser("~/Library/Caches")) + + @property + def user_state_dir(self) -> str: + """:return: state directory tied to the user, same as `user_data_dir`""" + return self.user_data_dir + + @property + def user_log_dir(self) -> str: + """:return: log directory tied to the user, e.g. ``~/Library/Logs/$appname/$version``""" + return self._append_app_name_and_version(os.path.expanduser("~/Library/Logs")) + + @property + def user_documents_dir(self) -> str: + """:return: documents directory tied to the user, e.g. ``~/Documents``""" + return os.path.expanduser("~/Documents") + + @property + def user_runtime_dir(self) -> str: + """:return: runtime directory tied to the user, e.g. ``~/Library/Caches/TemporaryItems/$appname/$version``""" + return self._append_app_name_and_version(os.path.expanduser("~/Library/Caches/TemporaryItems")) + + +__all__ = [ + "MacOS", +] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/unix.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/unix.py new file mode 100644 index 0000000000..9aca5a0305 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/unix.py @@ -0,0 +1,181 @@ +from __future__ import annotations + +import os +import sys +from configparser import ConfigParser +from pathlib import Path + +from .api import PlatformDirsABC + +if sys.platform.startswith("linux"): # pragma: no branch # no op check, only to please the type checker + from os import getuid +else: + + def getuid() -> int: + raise RuntimeError("should only be used on Linux") + + +class Unix(PlatformDirsABC): + """ + On Unix/Linux, we follow the + `XDG Basedir Spec <https://specifications.freedesktop.org/basedir-spec/basedir-spec-latest.html>`_. The spec allows + overriding directories with environment variables. The examples show are the default values, alongside the name of + the environment variable that overrides them. Makes use of the + `appname <platformdirs.api.PlatformDirsABC.appname>`, + `version <platformdirs.api.PlatformDirsABC.version>`, + `multipath <platformdirs.api.PlatformDirsABC.multipath>`, + `opinion <platformdirs.api.PlatformDirsABC.opinion>`. + """ + + @property + def user_data_dir(self) -> str: + """ + :return: data directory tied to the user, e.g. ``~/.local/share/$appname/$version`` or + ``$XDG_DATA_HOME/$appname/$version`` + """ + path = os.environ.get("XDG_DATA_HOME", "") + if not path.strip(): + path = os.path.expanduser("~/.local/share") + return self._append_app_name_and_version(path) + + @property + def site_data_dir(self) -> str: + """ + :return: data directories shared by users (if `multipath <platformdirs.api.PlatformDirsABC.multipath>` is + enabled and ``XDG_DATA_DIR`` is set and a multi path the response is also a multi path separated by the OS + path separator), e.g. ``/usr/local/share/$appname/$version`` or ``/usr/share/$appname/$version`` + """ + # XDG default for $XDG_DATA_DIRS; only first, if multipath is False + path = os.environ.get("XDG_DATA_DIRS", "") + if not path.strip(): + path = f"/usr/local/share{os.pathsep}/usr/share" + return self._with_multi_path(path) + + def _with_multi_path(self, path: str) -> str: + path_list = path.split(os.pathsep) + if not self.multipath: + path_list = path_list[0:1] + path_list = [self._append_app_name_and_version(os.path.expanduser(p)) for p in path_list] + return os.pathsep.join(path_list) + + @property + def user_config_dir(self) -> str: + """ + :return: config directory tied to the user, e.g. ``~/.config/$appname/$version`` or + ``$XDG_CONFIG_HOME/$appname/$version`` + """ + path = os.environ.get("XDG_CONFIG_HOME", "") + if not path.strip(): + path = os.path.expanduser("~/.config") + return self._append_app_name_and_version(path) + + @property + def site_config_dir(self) -> str: + """ + :return: config directories shared by users (if `multipath <platformdirs.api.PlatformDirsABC.multipath>` + is enabled and ``XDG_DATA_DIR`` is set and a multi path the response is also a multi path separated by the OS + path separator), e.g. ``/etc/xdg/$appname/$version`` + """ + # XDG default for $XDG_CONFIG_DIRS only first, if multipath is False + path = os.environ.get("XDG_CONFIG_DIRS", "") + if not path.strip(): + path = "/etc/xdg" + return self._with_multi_path(path) + + @property + def user_cache_dir(self) -> str: + """ + :return: cache directory tied to the user, e.g. ``~/.cache/$appname/$version`` or + ``~/$XDG_CACHE_HOME/$appname/$version`` + """ + path = os.environ.get("XDG_CACHE_HOME", "") + if not path.strip(): + path = os.path.expanduser("~/.cache") + return self._append_app_name_and_version(path) + + @property + def user_state_dir(self) -> str: + """ + :return: state directory tied to the user, e.g. ``~/.local/state/$appname/$version`` or + ``$XDG_STATE_HOME/$appname/$version`` + """ + path = os.environ.get("XDG_STATE_HOME", "") + if not path.strip(): + path = os.path.expanduser("~/.local/state") + return self._append_app_name_and_version(path) + + @property + def user_log_dir(self) -> str: + """ + :return: log directory tied to the user, same as `user_state_dir` if not opinionated else ``log`` in it + """ + path = self.user_state_dir + if self.opinion: + path = os.path.join(path, "log") + return path + + @property + def user_documents_dir(self) -> str: + """ + :return: documents directory tied to the user, e.g. ``~/Documents`` + """ + documents_dir = _get_user_dirs_folder("XDG_DOCUMENTS_DIR") + if documents_dir is None: + documents_dir = os.environ.get("XDG_DOCUMENTS_DIR", "").strip() + if not documents_dir: + documents_dir = os.path.expanduser("~/Documents") + + return documents_dir + + @property + def user_runtime_dir(self) -> str: + """ + :return: runtime directory tied to the user, e.g. ``/run/user/$(id -u)/$appname/$version`` or + ``$XDG_RUNTIME_DIR/$appname/$version`` + """ + path = os.environ.get("XDG_RUNTIME_DIR", "") + if not path.strip(): + path = f"/run/user/{getuid()}" + return self._append_app_name_and_version(path) + + @property + def site_data_path(self) -> Path: + """:return: data path shared by users. Only return first item, even if ``multipath`` is set to ``True``""" + return self._first_item_as_path_if_multipath(self.site_data_dir) + + @property + def site_config_path(self) -> Path: + """:return: config path shared by the users. Only return first item, even if ``multipath`` is set to ``True``""" + return self._first_item_as_path_if_multipath(self.site_config_dir) + + def _first_item_as_path_if_multipath(self, directory: str) -> Path: + if self.multipath: + # If multipath is True, the first path is returned. + directory = directory.split(os.pathsep)[0] + return Path(directory) + + +def _get_user_dirs_folder(key: str) -> str | None: + """Return directory from user-dirs.dirs config file. See https://freedesktop.org/wiki/Software/xdg-user-dirs/""" + user_dirs_config_path = os.path.join(Unix().user_config_dir, "user-dirs.dirs") + if os.path.exists(user_dirs_config_path): + parser = ConfigParser() + + with open(user_dirs_config_path) as stream: + # Add fake section header, so ConfigParser doesn't complain + parser.read_string(f"[top]\n{stream.read()}") + + if key not in parser["top"]: + return None + + path = parser["top"][key].strip('"') + # Handle relative home paths + path = path.replace("$HOME", os.path.expanduser("~")) + return path + + return None + + +__all__ = [ + "Unix", +] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/version.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/version.py new file mode 100644 index 0000000000..9f6eb98e8f --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/version.py @@ -0,0 +1,4 @@ +# file generated by setuptools_scm +# don't change, don't track in version control +__version__ = version = '2.6.2' +__version_tuple__ = version_tuple = (2, 6, 2) diff --git a/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/windows.py b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/windows.py new file mode 100644 index 0000000000..d5c27b3414 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/platformdirs/windows.py @@ -0,0 +1,184 @@ +from __future__ import annotations + +import ctypes +import os +import sys +from functools import lru_cache +from typing import Callable + +from .api import PlatformDirsABC + + +class Windows(PlatformDirsABC): + """`MSDN on where to store app data files + <http://support.microsoft.com/default.aspx?scid=kb;en-us;310294#XSLTH3194121123120121120120>`_. + Makes use of the + `appname <platformdirs.api.PlatformDirsABC.appname>`, + `appauthor <platformdirs.api.PlatformDirsABC.appauthor>`, + `version <platformdirs.api.PlatformDirsABC.version>`, + `roaming <platformdirs.api.PlatformDirsABC.roaming>`, + `opinion <platformdirs.api.PlatformDirsABC.opinion>`.""" + + @property + def user_data_dir(self) -> str: + """ + :return: data directory tied to the user, e.g. + ``%USERPROFILE%\\AppData\\Local\\$appauthor\\$appname`` (not roaming) or + ``%USERPROFILE%\\AppData\\Roaming\\$appauthor\\$appname`` (roaming) + """ + const = "CSIDL_APPDATA" if self.roaming else "CSIDL_LOCAL_APPDATA" + path = os.path.normpath(get_win_folder(const)) + return self._append_parts(path) + + def _append_parts(self, path: str, *, opinion_value: str | None = None) -> str: + params = [] + if self.appname: + if self.appauthor is not False: + author = self.appauthor or self.appname + params.append(author) + params.append(self.appname) + if opinion_value is not None and self.opinion: + params.append(opinion_value) + if self.version: + params.append(self.version) + return os.path.join(path, *params) + + @property + def site_data_dir(self) -> str: + """:return: data directory shared by users, e.g. ``C:\\ProgramData\\$appauthor\\$appname``""" + path = os.path.normpath(get_win_folder("CSIDL_COMMON_APPDATA")) + return self._append_parts(path) + + @property + def user_config_dir(self) -> str: + """:return: config directory tied to the user, same as `user_data_dir`""" + return self.user_data_dir + + @property + def site_config_dir(self) -> str: + """:return: config directory shared by the users, same as `site_data_dir`""" + return self.site_data_dir + + @property + def user_cache_dir(self) -> str: + """ + :return: cache directory tied to the user (if opinionated with ``Cache`` folder within ``$appname``) e.g. + ``%USERPROFILE%\\AppData\\Local\\$appauthor\\$appname\\Cache\\$version`` + """ + path = os.path.normpath(get_win_folder("CSIDL_LOCAL_APPDATA")) + return self._append_parts(path, opinion_value="Cache") + + @property + def user_state_dir(self) -> str: + """:return: state directory tied to the user, same as `user_data_dir`""" + return self.user_data_dir + + @property + def user_log_dir(self) -> str: + """ + :return: log directory tied to the user, same as `user_data_dir` if not opinionated else ``Logs`` in it + """ + path = self.user_data_dir + if self.opinion: + path = os.path.join(path, "Logs") + return path + + @property + def user_documents_dir(self) -> str: + """ + :return: documents directory tied to the user e.g. ``%USERPROFILE%\\Documents`` + """ + return os.path.normpath(get_win_folder("CSIDL_PERSONAL")) + + @property + def user_runtime_dir(self) -> str: + """ + :return: runtime directory tied to the user, e.g. + ``%USERPROFILE%\\AppData\\Local\\Temp\\$appauthor\\$appname`` + """ + path = os.path.normpath(os.path.join(get_win_folder("CSIDL_LOCAL_APPDATA"), "Temp")) + return self._append_parts(path) + + +def get_win_folder_from_env_vars(csidl_name: str) -> str: + """Get folder from environment variables.""" + if csidl_name == "CSIDL_PERSONAL": # does not have an environment name + return os.path.join(os.path.normpath(os.environ["USERPROFILE"]), "Documents") + + env_var_name = { + "CSIDL_APPDATA": "APPDATA", + "CSIDL_COMMON_APPDATA": "ALLUSERSPROFILE", + "CSIDL_LOCAL_APPDATA": "LOCALAPPDATA", + }.get(csidl_name) + if env_var_name is None: + raise ValueError(f"Unknown CSIDL name: {csidl_name}") + result = os.environ.get(env_var_name) + if result is None: + raise ValueError(f"Unset environment variable: {env_var_name}") + return result + + +def get_win_folder_from_registry(csidl_name: str) -> str: + """Get folder from the registry. + + This is a fallback technique at best. I'm not sure if using the + registry for this guarantees us the correct answer for all CSIDL_* + names. + """ + shell_folder_name = { + "CSIDL_APPDATA": "AppData", + "CSIDL_COMMON_APPDATA": "Common AppData", + "CSIDL_LOCAL_APPDATA": "Local AppData", + "CSIDL_PERSONAL": "Personal", + }.get(csidl_name) + if shell_folder_name is None: + raise ValueError(f"Unknown CSIDL name: {csidl_name}") + if sys.platform != "win32": # only needed for mypy type checker to know that this code runs only on Windows + raise NotImplementedError + import winreg + + key = winreg.OpenKey(winreg.HKEY_CURRENT_USER, r"Software\Microsoft\Windows\CurrentVersion\Explorer\Shell Folders") + directory, _ = winreg.QueryValueEx(key, shell_folder_name) + return str(directory) + + +def get_win_folder_via_ctypes(csidl_name: str) -> str: + """Get folder with ctypes.""" + csidl_const = { + "CSIDL_APPDATA": 26, + "CSIDL_COMMON_APPDATA": 35, + "CSIDL_LOCAL_APPDATA": 28, + "CSIDL_PERSONAL": 5, + }.get(csidl_name) + if csidl_const is None: + raise ValueError(f"Unknown CSIDL name: {csidl_name}") + + buf = ctypes.create_unicode_buffer(1024) + windll = getattr(ctypes, "windll") # noqa: B009 # using getattr to avoid false positive with mypy type checker + windll.shell32.SHGetFolderPathW(None, csidl_const, None, 0, buf) + + # Downgrade to short path name if it has highbit chars. + if any(ord(c) > 255 for c in buf): + buf2 = ctypes.create_unicode_buffer(1024) + if windll.kernel32.GetShortPathNameW(buf.value, buf2, 1024): + buf = buf2 + + return buf.value + + +def _pick_get_win_folder() -> Callable[[str], str]: + if hasattr(ctypes, "windll"): + return get_win_folder_via_ctypes + try: + import winreg # noqa: F401 + except ImportError: + return get_win_folder_from_env_vars + else: + return get_win_folder_from_registry + + +get_win_folder = lru_cache(maxsize=None)(_pick_get_win_folder()) + +__all__ = [ + "Windows", +] diff --git a/third_party/python/setuptools/pkg_resources/_vendor/typing_extensions.py b/third_party/python/setuptools/pkg_resources/_vendor/typing_extensions.py new file mode 100644 index 0000000000..ef42417c20 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/typing_extensions.py @@ -0,0 +1,2209 @@ +import abc +import collections +import collections.abc +import functools +import operator +import sys +import types as _types +import typing + + +__all__ = [ + # Super-special typing primitives. + 'Any', + 'ClassVar', + 'Concatenate', + 'Final', + 'LiteralString', + 'ParamSpec', + 'ParamSpecArgs', + 'ParamSpecKwargs', + 'Self', + 'Type', + 'TypeVar', + 'TypeVarTuple', + 'Unpack', + + # ABCs (from collections.abc). + 'Awaitable', + 'AsyncIterator', + 'AsyncIterable', + 'Coroutine', + 'AsyncGenerator', + 'AsyncContextManager', + 'ChainMap', + + # Concrete collection types. + 'ContextManager', + 'Counter', + 'Deque', + 'DefaultDict', + 'NamedTuple', + 'OrderedDict', + 'TypedDict', + + # Structural checks, a.k.a. protocols. + 'SupportsIndex', + + # One-off things. + 'Annotated', + 'assert_never', + 'assert_type', + 'clear_overloads', + 'dataclass_transform', + 'get_overloads', + 'final', + 'get_args', + 'get_origin', + 'get_type_hints', + 'IntVar', + 'is_typeddict', + 'Literal', + 'NewType', + 'overload', + 'override', + 'Protocol', + 'reveal_type', + 'runtime', + 'runtime_checkable', + 'Text', + 'TypeAlias', + 'TypeGuard', + 'TYPE_CHECKING', + 'Never', + 'NoReturn', + 'Required', + 'NotRequired', +] + +# for backward compatibility +PEP_560 = True +GenericMeta = type + +# The functions below are modified copies of typing internal helpers. +# They are needed by _ProtocolMeta and they provide support for PEP 646. + +_marker = object() + + +def _check_generic(cls, parameters, elen=_marker): + """Check correct count for parameters of a generic cls (internal helper). + This gives a nice error message in case of count mismatch. + """ + if not elen: + raise TypeError(f"{cls} is not a generic class") + if elen is _marker: + if not hasattr(cls, "__parameters__") or not cls.__parameters__: + raise TypeError(f"{cls} is not a generic class") + elen = len(cls.__parameters__) + alen = len(parameters) + if alen != elen: + if hasattr(cls, "__parameters__"): + parameters = [p for p in cls.__parameters__ if not _is_unpack(p)] + num_tv_tuples = sum(isinstance(p, TypeVarTuple) for p in parameters) + if (num_tv_tuples > 0) and (alen >= elen - num_tv_tuples): + return + raise TypeError(f"Too {'many' if alen > elen else 'few'} parameters for {cls};" + f" actual {alen}, expected {elen}") + + +if sys.version_info >= (3, 10): + def _should_collect_from_parameters(t): + return isinstance( + t, (typing._GenericAlias, _types.GenericAlias, _types.UnionType) + ) +elif sys.version_info >= (3, 9): + def _should_collect_from_parameters(t): + return isinstance(t, (typing._GenericAlias, _types.GenericAlias)) +else: + def _should_collect_from_parameters(t): + return isinstance(t, typing._GenericAlias) and not t._special + + +def _collect_type_vars(types, typevar_types=None): + """Collect all type variable contained in types in order of + first appearance (lexicographic order). For example:: + + _collect_type_vars((T, List[S, T])) == (T, S) + """ + if typevar_types is None: + typevar_types = typing.TypeVar + tvars = [] + for t in types: + if ( + isinstance(t, typevar_types) and + t not in tvars and + not _is_unpack(t) + ): + tvars.append(t) + if _should_collect_from_parameters(t): + tvars.extend([t for t in t.__parameters__ if t not in tvars]) + return tuple(tvars) + + +NoReturn = typing.NoReturn + +# Some unconstrained type variables. These are used by the container types. +# (These are not for export.) +T = typing.TypeVar('T') # Any type. +KT = typing.TypeVar('KT') # Key type. +VT = typing.TypeVar('VT') # Value type. +T_co = typing.TypeVar('T_co', covariant=True) # Any type covariant containers. +T_contra = typing.TypeVar('T_contra', contravariant=True) # Ditto contravariant. + + +if sys.version_info >= (3, 11): + from typing import Any +else: + + class _AnyMeta(type): + def __instancecheck__(self, obj): + if self is Any: + raise TypeError("typing_extensions.Any cannot be used with isinstance()") + return super().__instancecheck__(obj) + + def __repr__(self): + if self is Any: + return "typing_extensions.Any" + return super().__repr__() + + class Any(metaclass=_AnyMeta): + """Special type indicating an unconstrained type. + - Any is compatible with every type. + - Any assumed to have all methods. + - All values assumed to be instances of Any. + Note that all the above statements are true from the point of view of + static type checkers. At runtime, Any should not be used with instance + checks. + """ + def __new__(cls, *args, **kwargs): + if cls is Any: + raise TypeError("Any cannot be instantiated") + return super().__new__(cls, *args, **kwargs) + + +ClassVar = typing.ClassVar + +# On older versions of typing there is an internal class named "Final". +# 3.8+ +if hasattr(typing, 'Final') and sys.version_info[:2] >= (3, 7): + Final = typing.Final +# 3.7 +else: + class _FinalForm(typing._SpecialForm, _root=True): + + def __repr__(self): + return 'typing_extensions.' + self._name + + def __getitem__(self, parameters): + item = typing._type_check(parameters, + f'{self._name} accepts only a single type.') + return typing._GenericAlias(self, (item,)) + + Final = _FinalForm('Final', + doc="""A special typing construct to indicate that a name + cannot be re-assigned or overridden in a subclass. + For example: + + MAX_SIZE: Final = 9000 + MAX_SIZE += 1 # Error reported by type checker + + class Connection: + TIMEOUT: Final[int] = 10 + class FastConnector(Connection): + TIMEOUT = 1 # Error reported by type checker + + There is no runtime checking of these properties.""") + +if sys.version_info >= (3, 11): + final = typing.final +else: + # @final exists in 3.8+, but we backport it for all versions + # before 3.11 to keep support for the __final__ attribute. + # See https://bugs.python.org/issue46342 + def final(f): + """This decorator can be used to indicate to type checkers that + the decorated method cannot be overridden, and decorated class + cannot be subclassed. For example: + + class Base: + @final + def done(self) -> None: + ... + class Sub(Base): + def done(self) -> None: # Error reported by type checker + ... + @final + class Leaf: + ... + class Other(Leaf): # Error reported by type checker + ... + + There is no runtime checking of these properties. The decorator + sets the ``__final__`` attribute to ``True`` on the decorated object + to allow runtime introspection. + """ + try: + f.__final__ = True + except (AttributeError, TypeError): + # Skip the attribute silently if it is not writable. + # AttributeError happens if the object has __slots__ or a + # read-only property, TypeError if it's a builtin class. + pass + return f + + +def IntVar(name): + return typing.TypeVar(name) + + +# 3.8+: +if hasattr(typing, 'Literal'): + Literal = typing.Literal +# 3.7: +else: + class _LiteralForm(typing._SpecialForm, _root=True): + + def __repr__(self): + return 'typing_extensions.' + self._name + + def __getitem__(self, parameters): + return typing._GenericAlias(self, parameters) + + Literal = _LiteralForm('Literal', + doc="""A type that can be used to indicate to type checkers + that the corresponding value has a value literally equivalent + to the provided parameter. For example: + + var: Literal[4] = 4 + + The type checker understands that 'var' is literally equal to + the value 4 and no other value. + + Literal[...] cannot be subclassed. There is no runtime + checking verifying that the parameter is actually a value + instead of a type.""") + + +_overload_dummy = typing._overload_dummy # noqa + + +if hasattr(typing, "get_overloads"): # 3.11+ + overload = typing.overload + get_overloads = typing.get_overloads + clear_overloads = typing.clear_overloads +else: + # {module: {qualname: {firstlineno: func}}} + _overload_registry = collections.defaultdict( + functools.partial(collections.defaultdict, dict) + ) + + def overload(func): + """Decorator for overloaded functions/methods. + + In a stub file, place two or more stub definitions for the same + function in a row, each decorated with @overload. For example: + + @overload + def utf8(value: None) -> None: ... + @overload + def utf8(value: bytes) -> bytes: ... + @overload + def utf8(value: str) -> bytes: ... + + In a non-stub file (i.e. a regular .py file), do the same but + follow it with an implementation. The implementation should *not* + be decorated with @overload. For example: + + @overload + def utf8(value: None) -> None: ... + @overload + def utf8(value: bytes) -> bytes: ... + @overload + def utf8(value: str) -> bytes: ... + def utf8(value): + # implementation goes here + + The overloads for a function can be retrieved at runtime using the + get_overloads() function. + """ + # classmethod and staticmethod + f = getattr(func, "__func__", func) + try: + _overload_registry[f.__module__][f.__qualname__][ + f.__code__.co_firstlineno + ] = func + except AttributeError: + # Not a normal function; ignore. + pass + return _overload_dummy + + def get_overloads(func): + """Return all defined overloads for *func* as a sequence.""" + # classmethod and staticmethod + f = getattr(func, "__func__", func) + if f.__module__ not in _overload_registry: + return [] + mod_dict = _overload_registry[f.__module__] + if f.__qualname__ not in mod_dict: + return [] + return list(mod_dict[f.__qualname__].values()) + + def clear_overloads(): + """Clear all overloads in the registry.""" + _overload_registry.clear() + + +# This is not a real generic class. Don't use outside annotations. +Type = typing.Type + +# Various ABCs mimicking those in collections.abc. +# A few are simply re-exported for completeness. + + +Awaitable = typing.Awaitable +Coroutine = typing.Coroutine +AsyncIterable = typing.AsyncIterable +AsyncIterator = typing.AsyncIterator +Deque = typing.Deque +ContextManager = typing.ContextManager +AsyncContextManager = typing.AsyncContextManager +DefaultDict = typing.DefaultDict + +# 3.7.2+ +if hasattr(typing, 'OrderedDict'): + OrderedDict = typing.OrderedDict +# 3.7.0-3.7.2 +else: + OrderedDict = typing._alias(collections.OrderedDict, (KT, VT)) + +Counter = typing.Counter +ChainMap = typing.ChainMap +AsyncGenerator = typing.AsyncGenerator +NewType = typing.NewType +Text = typing.Text +TYPE_CHECKING = typing.TYPE_CHECKING + + +_PROTO_WHITELIST = ['Callable', 'Awaitable', + 'Iterable', 'Iterator', 'AsyncIterable', 'AsyncIterator', + 'Hashable', 'Sized', 'Container', 'Collection', 'Reversible', + 'ContextManager', 'AsyncContextManager'] + + +def _get_protocol_attrs(cls): + attrs = set() + for base in cls.__mro__[:-1]: # without object + if base.__name__ in ('Protocol', 'Generic'): + continue + annotations = getattr(base, '__annotations__', {}) + for attr in list(base.__dict__.keys()) + list(annotations.keys()): + if (not attr.startswith('_abc_') and attr not in ( + '__abstractmethods__', '__annotations__', '__weakref__', + '_is_protocol', '_is_runtime_protocol', '__dict__', + '__args__', '__slots__', + '__next_in_mro__', '__parameters__', '__origin__', + '__orig_bases__', '__extra__', '__tree_hash__', + '__doc__', '__subclasshook__', '__init__', '__new__', + '__module__', '_MutableMapping__marker', '_gorg')): + attrs.add(attr) + return attrs + + +def _is_callable_members_only(cls): + return all(callable(getattr(cls, attr, None)) for attr in _get_protocol_attrs(cls)) + + +def _maybe_adjust_parameters(cls): + """Helper function used in Protocol.__init_subclass__ and _TypedDictMeta.__new__. + + The contents of this function are very similar + to logic found in typing.Generic.__init_subclass__ + on the CPython main branch. + """ + tvars = [] + if '__orig_bases__' in cls.__dict__: + tvars = typing._collect_type_vars(cls.__orig_bases__) + # Look for Generic[T1, ..., Tn] or Protocol[T1, ..., Tn]. + # If found, tvars must be a subset of it. + # If not found, tvars is it. + # Also check for and reject plain Generic, + # and reject multiple Generic[...] and/or Protocol[...]. + gvars = None + for base in cls.__orig_bases__: + if (isinstance(base, typing._GenericAlias) and + base.__origin__ in (typing.Generic, Protocol)): + # for error messages + the_base = base.__origin__.__name__ + if gvars is not None: + raise TypeError( + "Cannot inherit from Generic[...]" + " and/or Protocol[...] multiple types.") + gvars = base.__parameters__ + if gvars is None: + gvars = tvars + else: + tvarset = set(tvars) + gvarset = set(gvars) + if not tvarset <= gvarset: + s_vars = ', '.join(str(t) for t in tvars if t not in gvarset) + s_args = ', '.join(str(g) for g in gvars) + raise TypeError(f"Some type variables ({s_vars}) are" + f" not listed in {the_base}[{s_args}]") + tvars = gvars + cls.__parameters__ = tuple(tvars) + + +# 3.8+ +if hasattr(typing, 'Protocol'): + Protocol = typing.Protocol +# 3.7 +else: + + def _no_init(self, *args, **kwargs): + if type(self)._is_protocol: + raise TypeError('Protocols cannot be instantiated') + + class _ProtocolMeta(abc.ABCMeta): # noqa: B024 + # This metaclass is a bit unfortunate and exists only because of the lack + # of __instancehook__. + def __instancecheck__(cls, instance): + # We need this method for situations where attributes are + # assigned in __init__. + if ((not getattr(cls, '_is_protocol', False) or + _is_callable_members_only(cls)) and + issubclass(instance.__class__, cls)): + return True + if cls._is_protocol: + if all(hasattr(instance, attr) and + (not callable(getattr(cls, attr, None)) or + getattr(instance, attr) is not None) + for attr in _get_protocol_attrs(cls)): + return True + return super().__instancecheck__(instance) + + class Protocol(metaclass=_ProtocolMeta): + # There is quite a lot of overlapping code with typing.Generic. + # Unfortunately it is hard to avoid this while these live in two different + # modules. The duplicated code will be removed when Protocol is moved to typing. + """Base class for protocol classes. Protocol classes are defined as:: + + class Proto(Protocol): + def meth(self) -> int: + ... + + Such classes are primarily used with static type checkers that recognize + structural subtyping (static duck-typing), for example:: + + class C: + def meth(self) -> int: + return 0 + + def func(x: Proto) -> int: + return x.meth() + + func(C()) # Passes static type check + + See PEP 544 for details. Protocol classes decorated with + @typing_extensions.runtime act as simple-minded runtime protocol that checks + only the presence of given attributes, ignoring their type signatures. + + Protocol classes can be generic, they are defined as:: + + class GenProto(Protocol[T]): + def meth(self) -> T: + ... + """ + __slots__ = () + _is_protocol = True + + def __new__(cls, *args, **kwds): + if cls is Protocol: + raise TypeError("Type Protocol cannot be instantiated; " + "it can only be used as a base class") + return super().__new__(cls) + + @typing._tp_cache + def __class_getitem__(cls, params): + if not isinstance(params, tuple): + params = (params,) + if not params and cls is not typing.Tuple: + raise TypeError( + f"Parameter list to {cls.__qualname__}[...] cannot be empty") + msg = "Parameters to generic types must be types." + params = tuple(typing._type_check(p, msg) for p in params) # noqa + if cls is Protocol: + # Generic can only be subscripted with unique type variables. + if not all(isinstance(p, typing.TypeVar) for p in params): + i = 0 + while isinstance(params[i], typing.TypeVar): + i += 1 + raise TypeError( + "Parameters to Protocol[...] must all be type variables." + f" Parameter {i + 1} is {params[i]}") + if len(set(params)) != len(params): + raise TypeError( + "Parameters to Protocol[...] must all be unique") + else: + # Subscripting a regular Generic subclass. + _check_generic(cls, params, len(cls.__parameters__)) + return typing._GenericAlias(cls, params) + + def __init_subclass__(cls, *args, **kwargs): + if '__orig_bases__' in cls.__dict__: + error = typing.Generic in cls.__orig_bases__ + else: + error = typing.Generic in cls.__bases__ + if error: + raise TypeError("Cannot inherit from plain Generic") + _maybe_adjust_parameters(cls) + + # Determine if this is a protocol or a concrete subclass. + if not cls.__dict__.get('_is_protocol', None): + cls._is_protocol = any(b is Protocol for b in cls.__bases__) + + # Set (or override) the protocol subclass hook. + def _proto_hook(other): + if not cls.__dict__.get('_is_protocol', None): + return NotImplemented + if not getattr(cls, '_is_runtime_protocol', False): + if sys._getframe(2).f_globals['__name__'] in ['abc', 'functools']: + return NotImplemented + raise TypeError("Instance and class checks can only be used with" + " @runtime protocols") + if not _is_callable_members_only(cls): + if sys._getframe(2).f_globals['__name__'] in ['abc', 'functools']: + return NotImplemented + raise TypeError("Protocols with non-method members" + " don't support issubclass()") + if not isinstance(other, type): + # Same error as for issubclass(1, int) + raise TypeError('issubclass() arg 1 must be a class') + for attr in _get_protocol_attrs(cls): + for base in other.__mro__: + if attr in base.__dict__: + if base.__dict__[attr] is None: + return NotImplemented + break + annotations = getattr(base, '__annotations__', {}) + if (isinstance(annotations, typing.Mapping) and + attr in annotations and + isinstance(other, _ProtocolMeta) and + other._is_protocol): + break + else: + return NotImplemented + return True + if '__subclasshook__' not in cls.__dict__: + cls.__subclasshook__ = _proto_hook + + # We have nothing more to do for non-protocols. + if not cls._is_protocol: + return + + # Check consistency of bases. + for base in cls.__bases__: + if not (base in (object, typing.Generic) or + base.__module__ == 'collections.abc' and + base.__name__ in _PROTO_WHITELIST or + isinstance(base, _ProtocolMeta) and base._is_protocol): + raise TypeError('Protocols can only inherit from other' + f' protocols, got {repr(base)}') + cls.__init__ = _no_init + + +# 3.8+ +if hasattr(typing, 'runtime_checkable'): + runtime_checkable = typing.runtime_checkable +# 3.7 +else: + def runtime_checkable(cls): + """Mark a protocol class as a runtime protocol, so that it + can be used with isinstance() and issubclass(). Raise TypeError + if applied to a non-protocol class. + + This allows a simple-minded structural check very similar to the + one-offs in collections.abc such as Hashable. + """ + if not isinstance(cls, _ProtocolMeta) or not cls._is_protocol: + raise TypeError('@runtime_checkable can be only applied to protocol classes,' + f' got {cls!r}') + cls._is_runtime_protocol = True + return cls + + +# Exists for backwards compatibility. +runtime = runtime_checkable + + +# 3.8+ +if hasattr(typing, 'SupportsIndex'): + SupportsIndex = typing.SupportsIndex +# 3.7 +else: + @runtime_checkable + class SupportsIndex(Protocol): + __slots__ = () + + @abc.abstractmethod + def __index__(self) -> int: + pass + + +if hasattr(typing, "Required"): + # The standard library TypedDict in Python 3.8 does not store runtime information + # about which (if any) keys are optional. See https://bugs.python.org/issue38834 + # The standard library TypedDict in Python 3.9.0/1 does not honour the "total" + # keyword with old-style TypedDict(). See https://bugs.python.org/issue42059 + # The standard library TypedDict below Python 3.11 does not store runtime + # information about optional and required keys when using Required or NotRequired. + # Generic TypedDicts are also impossible using typing.TypedDict on Python <3.11. + TypedDict = typing.TypedDict + _TypedDictMeta = typing._TypedDictMeta + is_typeddict = typing.is_typeddict +else: + def _check_fails(cls, other): + try: + if sys._getframe(1).f_globals['__name__'] not in ['abc', + 'functools', + 'typing']: + # Typed dicts are only for static structural subtyping. + raise TypeError('TypedDict does not support instance and class checks') + except (AttributeError, ValueError): + pass + return False + + def _dict_new(*args, **kwargs): + if not args: + raise TypeError('TypedDict.__new__(): not enough arguments') + _, args = args[0], args[1:] # allow the "cls" keyword be passed + return dict(*args, **kwargs) + + _dict_new.__text_signature__ = '($cls, _typename, _fields=None, /, **kwargs)' + + def _typeddict_new(*args, total=True, **kwargs): + if not args: + raise TypeError('TypedDict.__new__(): not enough arguments') + _, args = args[0], args[1:] # allow the "cls" keyword be passed + if args: + typename, args = args[0], args[1:] # allow the "_typename" keyword be passed + elif '_typename' in kwargs: + typename = kwargs.pop('_typename') + import warnings + warnings.warn("Passing '_typename' as keyword argument is deprecated", + DeprecationWarning, stacklevel=2) + else: + raise TypeError("TypedDict.__new__() missing 1 required positional " + "argument: '_typename'") + if args: + try: + fields, = args # allow the "_fields" keyword be passed + except ValueError: + raise TypeError('TypedDict.__new__() takes from 2 to 3 ' + f'positional arguments but {len(args) + 2} ' + 'were given') + elif '_fields' in kwargs and len(kwargs) == 1: + fields = kwargs.pop('_fields') + import warnings + warnings.warn("Passing '_fields' as keyword argument is deprecated", + DeprecationWarning, stacklevel=2) + else: + fields = None + + if fields is None: + fields = kwargs + elif kwargs: + raise TypeError("TypedDict takes either a dict or keyword arguments," + " but not both") + + ns = {'__annotations__': dict(fields)} + try: + # Setting correct module is necessary to make typed dict classes pickleable. + ns['__module__'] = sys._getframe(1).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): + pass + + return _TypedDictMeta(typename, (), ns, total=total) + + _typeddict_new.__text_signature__ = ('($cls, _typename, _fields=None,' + ' /, *, total=True, **kwargs)') + + class _TypedDictMeta(type): + def __init__(cls, name, bases, ns, total=True): + super().__init__(name, bases, ns) + + def __new__(cls, name, bases, ns, total=True): + # Create new typed dict class object. + # This method is called directly when TypedDict is subclassed, + # or via _typeddict_new when TypedDict is instantiated. This way + # TypedDict supports all three syntaxes described in its docstring. + # Subclasses and instances of TypedDict return actual dictionaries + # via _dict_new. + ns['__new__'] = _typeddict_new if name == 'TypedDict' else _dict_new + # Don't insert typing.Generic into __bases__ here, + # or Generic.__init_subclass__ will raise TypeError + # in the super().__new__() call. + # Instead, monkey-patch __bases__ onto the class after it's been created. + tp_dict = super().__new__(cls, name, (dict,), ns) + + if any(issubclass(base, typing.Generic) for base in bases): + tp_dict.__bases__ = (typing.Generic, dict) + _maybe_adjust_parameters(tp_dict) + + annotations = {} + own_annotations = ns.get('__annotations__', {}) + msg = "TypedDict('Name', {f0: t0, f1: t1, ...}); each t must be a type" + own_annotations = { + n: typing._type_check(tp, msg) for n, tp in own_annotations.items() + } + required_keys = set() + optional_keys = set() + + for base in bases: + annotations.update(base.__dict__.get('__annotations__', {})) + required_keys.update(base.__dict__.get('__required_keys__', ())) + optional_keys.update(base.__dict__.get('__optional_keys__', ())) + + annotations.update(own_annotations) + for annotation_key, annotation_type in own_annotations.items(): + annotation_origin = get_origin(annotation_type) + if annotation_origin is Annotated: + annotation_args = get_args(annotation_type) + if annotation_args: + annotation_type = annotation_args[0] + annotation_origin = get_origin(annotation_type) + + if annotation_origin is Required: + required_keys.add(annotation_key) + elif annotation_origin is NotRequired: + optional_keys.add(annotation_key) + elif total: + required_keys.add(annotation_key) + else: + optional_keys.add(annotation_key) + + tp_dict.__annotations__ = annotations + tp_dict.__required_keys__ = frozenset(required_keys) + tp_dict.__optional_keys__ = frozenset(optional_keys) + if not hasattr(tp_dict, '__total__'): + tp_dict.__total__ = total + return tp_dict + + __instancecheck__ = __subclasscheck__ = _check_fails + + TypedDict = _TypedDictMeta('TypedDict', (dict,), {}) + TypedDict.__module__ = __name__ + TypedDict.__doc__ = \ + """A simple typed name space. At runtime it is equivalent to a plain dict. + + TypedDict creates a dictionary type that expects all of its + instances to have a certain set of keys, with each key + associated with a value of a consistent type. This expectation + is not checked at runtime but is only enforced by type checkers. + Usage:: + + class Point2D(TypedDict): + x: int + y: int + label: str + + a: Point2D = {'x': 1, 'y': 2, 'label': 'good'} # OK + b: Point2D = {'z': 3, 'label': 'bad'} # Fails type check + + assert Point2D(x=1, y=2, label='first') == dict(x=1, y=2, label='first') + + The type info can be accessed via the Point2D.__annotations__ dict, and + the Point2D.__required_keys__ and Point2D.__optional_keys__ frozensets. + TypedDict supports two additional equivalent forms:: + + Point2D = TypedDict('Point2D', x=int, y=int, label=str) + Point2D = TypedDict('Point2D', {'x': int, 'y': int, 'label': str}) + + The class syntax is only supported in Python 3.6+, while two other + syntax forms work for Python 2.7 and 3.2+ + """ + + if hasattr(typing, "_TypedDictMeta"): + _TYPEDDICT_TYPES = (typing._TypedDictMeta, _TypedDictMeta) + else: + _TYPEDDICT_TYPES = (_TypedDictMeta,) + + def is_typeddict(tp): + """Check if an annotation is a TypedDict class + + For example:: + class Film(TypedDict): + title: str + year: int + + is_typeddict(Film) # => True + is_typeddict(Union[list, str]) # => False + """ + return isinstance(tp, tuple(_TYPEDDICT_TYPES)) + + +if hasattr(typing, "assert_type"): + assert_type = typing.assert_type + +else: + def assert_type(__val, __typ): + """Assert (to the type checker) that the value is of the given type. + + When the type checker encounters a call to assert_type(), it + emits an error if the value is not of the specified type:: + + def greet(name: str) -> None: + assert_type(name, str) # ok + assert_type(name, int) # type checker error + + At runtime this returns the first argument unchanged and otherwise + does nothing. + """ + return __val + + +if hasattr(typing, "Required"): + get_type_hints = typing.get_type_hints +else: + import functools + import types + + # replaces _strip_annotations() + def _strip_extras(t): + """Strips Annotated, Required and NotRequired from a given type.""" + if isinstance(t, _AnnotatedAlias): + return _strip_extras(t.__origin__) + if hasattr(t, "__origin__") and t.__origin__ in (Required, NotRequired): + return _strip_extras(t.__args__[0]) + if isinstance(t, typing._GenericAlias): + stripped_args = tuple(_strip_extras(a) for a in t.__args__) + if stripped_args == t.__args__: + return t + return t.copy_with(stripped_args) + if hasattr(types, "GenericAlias") and isinstance(t, types.GenericAlias): + stripped_args = tuple(_strip_extras(a) for a in t.__args__) + if stripped_args == t.__args__: + return t + return types.GenericAlias(t.__origin__, stripped_args) + if hasattr(types, "UnionType") and isinstance(t, types.UnionType): + stripped_args = tuple(_strip_extras(a) for a in t.__args__) + if stripped_args == t.__args__: + return t + return functools.reduce(operator.or_, stripped_args) + + return t + + def get_type_hints(obj, globalns=None, localns=None, include_extras=False): + """Return type hints for an object. + + This is often the same as obj.__annotations__, but it handles + forward references encoded as string literals, adds Optional[t] if a + default value equal to None is set and recursively replaces all + 'Annotated[T, ...]', 'Required[T]' or 'NotRequired[T]' with 'T' + (unless 'include_extras=True'). + + The argument may be a module, class, method, or function. The annotations + are returned as a dictionary. For classes, annotations include also + inherited members. + + TypeError is raised if the argument is not of a type that can contain + annotations, and an empty dictionary is returned if no annotations are + present. + + BEWARE -- the behavior of globalns and localns is counterintuitive + (unless you are familiar with how eval() and exec() work). The + search order is locals first, then globals. + + - If no dict arguments are passed, an attempt is made to use the + globals from obj (or the respective module's globals for classes), + and these are also used as the locals. If the object does not appear + to have globals, an empty dictionary is used. + + - If one dict argument is passed, it is used for both globals and + locals. + + - If two dict arguments are passed, they specify globals and + locals, respectively. + """ + if hasattr(typing, "Annotated"): + hint = typing.get_type_hints( + obj, globalns=globalns, localns=localns, include_extras=True + ) + else: + hint = typing.get_type_hints(obj, globalns=globalns, localns=localns) + if include_extras: + return hint + return {k: _strip_extras(t) for k, t in hint.items()} + + +# Python 3.9+ has PEP 593 (Annotated) +if hasattr(typing, 'Annotated'): + Annotated = typing.Annotated + # Not exported and not a public API, but needed for get_origin() and get_args() + # to work. + _AnnotatedAlias = typing._AnnotatedAlias +# 3.7-3.8 +else: + class _AnnotatedAlias(typing._GenericAlias, _root=True): + """Runtime representation of an annotated type. + + At its core 'Annotated[t, dec1, dec2, ...]' is an alias for the type 't' + with extra annotations. The alias behaves like a normal typing alias, + instantiating is the same as instantiating the underlying type, binding + it to types is also the same. + """ + def __init__(self, origin, metadata): + if isinstance(origin, _AnnotatedAlias): + metadata = origin.__metadata__ + metadata + origin = origin.__origin__ + super().__init__(origin, origin) + self.__metadata__ = metadata + + def copy_with(self, params): + assert len(params) == 1 + new_type = params[0] + return _AnnotatedAlias(new_type, self.__metadata__) + + def __repr__(self): + return (f"typing_extensions.Annotated[{typing._type_repr(self.__origin__)}, " + f"{', '.join(repr(a) for a in self.__metadata__)}]") + + def __reduce__(self): + return operator.getitem, ( + Annotated, (self.__origin__,) + self.__metadata__ + ) + + def __eq__(self, other): + if not isinstance(other, _AnnotatedAlias): + return NotImplemented + if self.__origin__ != other.__origin__: + return False + return self.__metadata__ == other.__metadata__ + + def __hash__(self): + return hash((self.__origin__, self.__metadata__)) + + class Annotated: + """Add context specific metadata to a type. + + Example: Annotated[int, runtime_check.Unsigned] indicates to the + hypothetical runtime_check module that this type is an unsigned int. + Every other consumer of this type can ignore this metadata and treat + this type as int. + + The first argument to Annotated must be a valid type (and will be in + the __origin__ field), the remaining arguments are kept as a tuple in + the __extra__ field. + + Details: + + - It's an error to call `Annotated` with less than two arguments. + - Nested Annotated are flattened:: + + Annotated[Annotated[T, Ann1, Ann2], Ann3] == Annotated[T, Ann1, Ann2, Ann3] + + - Instantiating an annotated type is equivalent to instantiating the + underlying type:: + + Annotated[C, Ann1](5) == C(5) + + - Annotated can be used as a generic type alias:: + + Optimized = Annotated[T, runtime.Optimize()] + Optimized[int] == Annotated[int, runtime.Optimize()] + + OptimizedList = Annotated[List[T], runtime.Optimize()] + OptimizedList[int] == Annotated[List[int], runtime.Optimize()] + """ + + __slots__ = () + + def __new__(cls, *args, **kwargs): + raise TypeError("Type Annotated cannot be instantiated.") + + @typing._tp_cache + def __class_getitem__(cls, params): + if not isinstance(params, tuple) or len(params) < 2: + raise TypeError("Annotated[...] should be used " + "with at least two arguments (a type and an " + "annotation).") + allowed_special_forms = (ClassVar, Final) + if get_origin(params[0]) in allowed_special_forms: + origin = params[0] + else: + msg = "Annotated[t, ...]: t must be a type." + origin = typing._type_check(params[0], msg) + metadata = tuple(params[1:]) + return _AnnotatedAlias(origin, metadata) + + def __init_subclass__(cls, *args, **kwargs): + raise TypeError( + f"Cannot subclass {cls.__module__}.Annotated" + ) + +# Python 3.8 has get_origin() and get_args() but those implementations aren't +# Annotated-aware, so we can't use those. Python 3.9's versions don't support +# ParamSpecArgs and ParamSpecKwargs, so only Python 3.10's versions will do. +if sys.version_info[:2] >= (3, 10): + get_origin = typing.get_origin + get_args = typing.get_args +# 3.7-3.9 +else: + try: + # 3.9+ + from typing import _BaseGenericAlias + except ImportError: + _BaseGenericAlias = typing._GenericAlias + try: + # 3.9+ + from typing import GenericAlias as _typing_GenericAlias + except ImportError: + _typing_GenericAlias = typing._GenericAlias + + def get_origin(tp): + """Get the unsubscripted version of a type. + + This supports generic types, Callable, Tuple, Union, Literal, Final, ClassVar + and Annotated. Return None for unsupported types. Examples:: + + get_origin(Literal[42]) is Literal + get_origin(int) is None + get_origin(ClassVar[int]) is ClassVar + get_origin(Generic) is Generic + get_origin(Generic[T]) is Generic + get_origin(Union[T, int]) is Union + get_origin(List[Tuple[T, T]][int]) == list + get_origin(P.args) is P + """ + if isinstance(tp, _AnnotatedAlias): + return Annotated + if isinstance(tp, (typing._GenericAlias, _typing_GenericAlias, _BaseGenericAlias, + ParamSpecArgs, ParamSpecKwargs)): + return tp.__origin__ + if tp is typing.Generic: + return typing.Generic + return None + + def get_args(tp): + """Get type arguments with all substitutions performed. + + For unions, basic simplifications used by Union constructor are performed. + Examples:: + get_args(Dict[str, int]) == (str, int) + get_args(int) == () + get_args(Union[int, Union[T, int], str][int]) == (int, str) + get_args(Union[int, Tuple[T, int]][str]) == (int, Tuple[str, int]) + get_args(Callable[[], T][int]) == ([], int) + """ + if isinstance(tp, _AnnotatedAlias): + return (tp.__origin__,) + tp.__metadata__ + if isinstance(tp, (typing._GenericAlias, _typing_GenericAlias)): + if getattr(tp, "_special", False): + return () + res = tp.__args__ + if get_origin(tp) is collections.abc.Callable and res[0] is not Ellipsis: + res = (list(res[:-1]), res[-1]) + return res + return () + + +# 3.10+ +if hasattr(typing, 'TypeAlias'): + TypeAlias = typing.TypeAlias +# 3.9 +elif sys.version_info[:2] >= (3, 9): + class _TypeAliasForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + @_TypeAliasForm + def TypeAlias(self, parameters): + """Special marker indicating that an assignment should + be recognized as a proper type alias definition by type + checkers. + + For example:: + + Predicate: TypeAlias = Callable[..., bool] + + It's invalid when used anywhere except as in the example above. + """ + raise TypeError(f"{self} is not subscriptable") +# 3.7-3.8 +else: + class _TypeAliasForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + TypeAlias = _TypeAliasForm('TypeAlias', + doc="""Special marker indicating that an assignment should + be recognized as a proper type alias definition by type + checkers. + + For example:: + + Predicate: TypeAlias = Callable[..., bool] + + It's invalid when used anywhere except as in the example + above.""") + + +class _DefaultMixin: + """Mixin for TypeVarLike defaults.""" + + __slots__ = () + + def __init__(self, default): + if isinstance(default, (tuple, list)): + self.__default__ = tuple((typing._type_check(d, "Default must be a type") + for d in default)) + elif default: + self.__default__ = typing._type_check(default, "Default must be a type") + else: + self.__default__ = None + + +# Add default and infer_variance parameters from PEP 696 and 695 +class TypeVar(typing.TypeVar, _DefaultMixin, _root=True): + """Type variable.""" + + __module__ = 'typing' + + def __init__(self, name, *constraints, bound=None, + covariant=False, contravariant=False, + default=None, infer_variance=False): + super().__init__(name, *constraints, bound=bound, covariant=covariant, + contravariant=contravariant) + _DefaultMixin.__init__(self, default) + self.__infer_variance__ = infer_variance + + # for pickling: + try: + def_mod = sys._getframe(1).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): + def_mod = None + if def_mod != 'typing_extensions': + self.__module__ = def_mod + + +# Python 3.10+ has PEP 612 +if hasattr(typing, 'ParamSpecArgs'): + ParamSpecArgs = typing.ParamSpecArgs + ParamSpecKwargs = typing.ParamSpecKwargs +# 3.7-3.9 +else: + class _Immutable: + """Mixin to indicate that object should not be copied.""" + __slots__ = () + + def __copy__(self): + return self + + def __deepcopy__(self, memo): + return self + + class ParamSpecArgs(_Immutable): + """The args for a ParamSpec object. + + Given a ParamSpec object P, P.args is an instance of ParamSpecArgs. + + ParamSpecArgs objects have a reference back to their ParamSpec: + + P.args.__origin__ is P + + This type is meant for runtime introspection and has no special meaning to + static type checkers. + """ + def __init__(self, origin): + self.__origin__ = origin + + def __repr__(self): + return f"{self.__origin__.__name__}.args" + + def __eq__(self, other): + if not isinstance(other, ParamSpecArgs): + return NotImplemented + return self.__origin__ == other.__origin__ + + class ParamSpecKwargs(_Immutable): + """The kwargs for a ParamSpec object. + + Given a ParamSpec object P, P.kwargs is an instance of ParamSpecKwargs. + + ParamSpecKwargs objects have a reference back to their ParamSpec: + + P.kwargs.__origin__ is P + + This type is meant for runtime introspection and has no special meaning to + static type checkers. + """ + def __init__(self, origin): + self.__origin__ = origin + + def __repr__(self): + return f"{self.__origin__.__name__}.kwargs" + + def __eq__(self, other): + if not isinstance(other, ParamSpecKwargs): + return NotImplemented + return self.__origin__ == other.__origin__ + +# 3.10+ +if hasattr(typing, 'ParamSpec'): + + # Add default Parameter - PEP 696 + class ParamSpec(typing.ParamSpec, _DefaultMixin, _root=True): + """Parameter specification variable.""" + + __module__ = 'typing' + + def __init__(self, name, *, bound=None, covariant=False, contravariant=False, + default=None): + super().__init__(name, bound=bound, covariant=covariant, + contravariant=contravariant) + _DefaultMixin.__init__(self, default) + + # for pickling: + try: + def_mod = sys._getframe(1).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): + def_mod = None + if def_mod != 'typing_extensions': + self.__module__ = def_mod + +# 3.7-3.9 +else: + + # Inherits from list as a workaround for Callable checks in Python < 3.9.2. + class ParamSpec(list, _DefaultMixin): + """Parameter specification variable. + + Usage:: + + P = ParamSpec('P') + + Parameter specification variables exist primarily for the benefit of static + type checkers. They are used to forward the parameter types of one + callable to another callable, a pattern commonly found in higher order + functions and decorators. They are only valid when used in ``Concatenate``, + or s the first argument to ``Callable``. In Python 3.10 and higher, + they are also supported in user-defined Generics at runtime. + See class Generic for more information on generic types. An + example for annotating a decorator:: + + T = TypeVar('T') + P = ParamSpec('P') + + def add_logging(f: Callable[P, T]) -> Callable[P, T]: + '''A type-safe decorator to add logging to a function.''' + def inner(*args: P.args, **kwargs: P.kwargs) -> T: + logging.info(f'{f.__name__} was called') + return f(*args, **kwargs) + return inner + + @add_logging + def add_two(x: float, y: float) -> float: + '''Add two numbers together.''' + return x + y + + Parameter specification variables defined with covariant=True or + contravariant=True can be used to declare covariant or contravariant + generic types. These keyword arguments are valid, but their actual semantics + are yet to be decided. See PEP 612 for details. + + Parameter specification variables can be introspected. e.g.: + + P.__name__ == 'T' + P.__bound__ == None + P.__covariant__ == False + P.__contravariant__ == False + + Note that only parameter specification variables defined in global scope can + be pickled. + """ + + # Trick Generic __parameters__. + __class__ = typing.TypeVar + + @property + def args(self): + return ParamSpecArgs(self) + + @property + def kwargs(self): + return ParamSpecKwargs(self) + + def __init__(self, name, *, bound=None, covariant=False, contravariant=False, + default=None): + super().__init__([self]) + self.__name__ = name + self.__covariant__ = bool(covariant) + self.__contravariant__ = bool(contravariant) + if bound: + self.__bound__ = typing._type_check(bound, 'Bound must be a type.') + else: + self.__bound__ = None + _DefaultMixin.__init__(self, default) + + # for pickling: + try: + def_mod = sys._getframe(1).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): + def_mod = None + if def_mod != 'typing_extensions': + self.__module__ = def_mod + + def __repr__(self): + if self.__covariant__: + prefix = '+' + elif self.__contravariant__: + prefix = '-' + else: + prefix = '~' + return prefix + self.__name__ + + def __hash__(self): + return object.__hash__(self) + + def __eq__(self, other): + return self is other + + def __reduce__(self): + return self.__name__ + + # Hack to get typing._type_check to pass. + def __call__(self, *args, **kwargs): + pass + + +# 3.7-3.9 +if not hasattr(typing, 'Concatenate'): + # Inherits from list as a workaround for Callable checks in Python < 3.9.2. + class _ConcatenateGenericAlias(list): + + # Trick Generic into looking into this for __parameters__. + __class__ = typing._GenericAlias + + # Flag in 3.8. + _special = False + + def __init__(self, origin, args): + super().__init__(args) + self.__origin__ = origin + self.__args__ = args + + def __repr__(self): + _type_repr = typing._type_repr + return (f'{_type_repr(self.__origin__)}' + f'[{", ".join(_type_repr(arg) for arg in self.__args__)}]') + + def __hash__(self): + return hash((self.__origin__, self.__args__)) + + # Hack to get typing._type_check to pass in Generic. + def __call__(self, *args, **kwargs): + pass + + @property + def __parameters__(self): + return tuple( + tp for tp in self.__args__ if isinstance(tp, (typing.TypeVar, ParamSpec)) + ) + + +# 3.7-3.9 +@typing._tp_cache +def _concatenate_getitem(self, parameters): + if parameters == (): + raise TypeError("Cannot take a Concatenate of no types.") + if not isinstance(parameters, tuple): + parameters = (parameters,) + if not isinstance(parameters[-1], ParamSpec): + raise TypeError("The last parameter to Concatenate should be a " + "ParamSpec variable.") + msg = "Concatenate[arg, ...]: each arg must be a type." + parameters = tuple(typing._type_check(p, msg) for p in parameters) + return _ConcatenateGenericAlias(self, parameters) + + +# 3.10+ +if hasattr(typing, 'Concatenate'): + Concatenate = typing.Concatenate + _ConcatenateGenericAlias = typing._ConcatenateGenericAlias # noqa +# 3.9 +elif sys.version_info[:2] >= (3, 9): + @_TypeAliasForm + def Concatenate(self, parameters): + """Used in conjunction with ``ParamSpec`` and ``Callable`` to represent a + higher order function which adds, removes or transforms parameters of a + callable. + + For example:: + + Callable[Concatenate[int, P], int] + + See PEP 612 for detailed information. + """ + return _concatenate_getitem(self, parameters) +# 3.7-8 +else: + class _ConcatenateForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + def __getitem__(self, parameters): + return _concatenate_getitem(self, parameters) + + Concatenate = _ConcatenateForm( + 'Concatenate', + doc="""Used in conjunction with ``ParamSpec`` and ``Callable`` to represent a + higher order function which adds, removes or transforms parameters of a + callable. + + For example:: + + Callable[Concatenate[int, P], int] + + See PEP 612 for detailed information. + """) + +# 3.10+ +if hasattr(typing, 'TypeGuard'): + TypeGuard = typing.TypeGuard +# 3.9 +elif sys.version_info[:2] >= (3, 9): + class _TypeGuardForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + @_TypeGuardForm + def TypeGuard(self, parameters): + """Special typing form used to annotate the return type of a user-defined + type guard function. ``TypeGuard`` only accepts a single type argument. + At runtime, functions marked this way should return a boolean. + + ``TypeGuard`` aims to benefit *type narrowing* -- a technique used by static + type checkers to determine a more precise type of an expression within a + program's code flow. Usually type narrowing is done by analyzing + conditional code flow and applying the narrowing to a block of code. The + conditional expression here is sometimes referred to as a "type guard". + + Sometimes it would be convenient to use a user-defined boolean function + as a type guard. Such a function should use ``TypeGuard[...]`` as its + return type to alert static type checkers to this intention. + + Using ``-> TypeGuard`` tells the static type checker that for a given + function: + + 1. The return value is a boolean. + 2. If the return value is ``True``, the type of its argument + is the type inside ``TypeGuard``. + + For example:: + + def is_str(val: Union[str, float]): + # "isinstance" type guard + if isinstance(val, str): + # Type of ``val`` is narrowed to ``str`` + ... + else: + # Else, type of ``val`` is narrowed to ``float``. + ... + + Strict type narrowing is not enforced -- ``TypeB`` need not be a narrower + form of ``TypeA`` (it can even be a wider form) and this may lead to + type-unsafe results. The main reason is to allow for things like + narrowing ``List[object]`` to ``List[str]`` even though the latter is not + a subtype of the former, since ``List`` is invariant. The responsibility of + writing type-safe type guards is left to the user. + + ``TypeGuard`` also works with type variables. For more information, see + PEP 647 (User-Defined Type Guards). + """ + item = typing._type_check(parameters, f'{self} accepts only a single type.') + return typing._GenericAlias(self, (item,)) +# 3.7-3.8 +else: + class _TypeGuardForm(typing._SpecialForm, _root=True): + + def __repr__(self): + return 'typing_extensions.' + self._name + + def __getitem__(self, parameters): + item = typing._type_check(parameters, + f'{self._name} accepts only a single type') + return typing._GenericAlias(self, (item,)) + + TypeGuard = _TypeGuardForm( + 'TypeGuard', + doc="""Special typing form used to annotate the return type of a user-defined + type guard function. ``TypeGuard`` only accepts a single type argument. + At runtime, functions marked this way should return a boolean. + + ``TypeGuard`` aims to benefit *type narrowing* -- a technique used by static + type checkers to determine a more precise type of an expression within a + program's code flow. Usually type narrowing is done by analyzing + conditional code flow and applying the narrowing to a block of code. The + conditional expression here is sometimes referred to as a "type guard". + + Sometimes it would be convenient to use a user-defined boolean function + as a type guard. Such a function should use ``TypeGuard[...]`` as its + return type to alert static type checkers to this intention. + + Using ``-> TypeGuard`` tells the static type checker that for a given + function: + + 1. The return value is a boolean. + 2. If the return value is ``True``, the type of its argument + is the type inside ``TypeGuard``. + + For example:: + + def is_str(val: Union[str, float]): + # "isinstance" type guard + if isinstance(val, str): + # Type of ``val`` is narrowed to ``str`` + ... + else: + # Else, type of ``val`` is narrowed to ``float``. + ... + + Strict type narrowing is not enforced -- ``TypeB`` need not be a narrower + form of ``TypeA`` (it can even be a wider form) and this may lead to + type-unsafe results. The main reason is to allow for things like + narrowing ``List[object]`` to ``List[str]`` even though the latter is not + a subtype of the former, since ``List`` is invariant. The responsibility of + writing type-safe type guards is left to the user. + + ``TypeGuard`` also works with type variables. For more information, see + PEP 647 (User-Defined Type Guards). + """) + + +# Vendored from cpython typing._SpecialFrom +class _SpecialForm(typing._Final, _root=True): + __slots__ = ('_name', '__doc__', '_getitem') + + def __init__(self, getitem): + self._getitem = getitem + self._name = getitem.__name__ + self.__doc__ = getitem.__doc__ + + def __getattr__(self, item): + if item in {'__name__', '__qualname__'}: + return self._name + + raise AttributeError(item) + + def __mro_entries__(self, bases): + raise TypeError(f"Cannot subclass {self!r}") + + def __repr__(self): + return f'typing_extensions.{self._name}' + + def __reduce__(self): + return self._name + + def __call__(self, *args, **kwds): + raise TypeError(f"Cannot instantiate {self!r}") + + def __or__(self, other): + return typing.Union[self, other] + + def __ror__(self, other): + return typing.Union[other, self] + + def __instancecheck__(self, obj): + raise TypeError(f"{self} cannot be used with isinstance()") + + def __subclasscheck__(self, cls): + raise TypeError(f"{self} cannot be used with issubclass()") + + @typing._tp_cache + def __getitem__(self, parameters): + return self._getitem(self, parameters) + + +if hasattr(typing, "LiteralString"): + LiteralString = typing.LiteralString +else: + @_SpecialForm + def LiteralString(self, params): + """Represents an arbitrary literal string. + + Example:: + + from typing_extensions import LiteralString + + def query(sql: LiteralString) -> ...: + ... + + query("SELECT * FROM table") # ok + query(f"SELECT * FROM {input()}") # not ok + + See PEP 675 for details. + + """ + raise TypeError(f"{self} is not subscriptable") + + +if hasattr(typing, "Self"): + Self = typing.Self +else: + @_SpecialForm + def Self(self, params): + """Used to spell the type of "self" in classes. + + Example:: + + from typing import Self + + class ReturnsSelf: + def parse(self, data: bytes) -> Self: + ... + return self + + """ + + raise TypeError(f"{self} is not subscriptable") + + +if hasattr(typing, "Never"): + Never = typing.Never +else: + @_SpecialForm + def Never(self, params): + """The bottom type, a type that has no members. + + This can be used to define a function that should never be + called, or a function that never returns:: + + from typing_extensions import Never + + def never_call_me(arg: Never) -> None: + pass + + def int_or_str(arg: int | str) -> None: + never_call_me(arg) # type checker error + match arg: + case int(): + print("It's an int") + case str(): + print("It's a str") + case _: + never_call_me(arg) # ok, arg is of type Never + + """ + + raise TypeError(f"{self} is not subscriptable") + + +if hasattr(typing, 'Required'): + Required = typing.Required + NotRequired = typing.NotRequired +elif sys.version_info[:2] >= (3, 9): + class _ExtensionsSpecialForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + @_ExtensionsSpecialForm + def Required(self, parameters): + """A special typing construct to mark a key of a total=False TypedDict + as required. For example: + + class Movie(TypedDict, total=False): + title: Required[str] + year: int + + m = Movie( + title='The Matrix', # typechecker error if key is omitted + year=1999, + ) + + There is no runtime checking that a required key is actually provided + when instantiating a related TypedDict. + """ + item = typing._type_check(parameters, f'{self._name} accepts only a single type.') + return typing._GenericAlias(self, (item,)) + + @_ExtensionsSpecialForm + def NotRequired(self, parameters): + """A special typing construct to mark a key of a TypedDict as + potentially missing. For example: + + class Movie(TypedDict): + title: str + year: NotRequired[int] + + m = Movie( + title='The Matrix', # typechecker error if key is omitted + year=1999, + ) + """ + item = typing._type_check(parameters, f'{self._name} accepts only a single type.') + return typing._GenericAlias(self, (item,)) + +else: + class _RequiredForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + def __getitem__(self, parameters): + item = typing._type_check(parameters, + f'{self._name} accepts only a single type.') + return typing._GenericAlias(self, (item,)) + + Required = _RequiredForm( + 'Required', + doc="""A special typing construct to mark a key of a total=False TypedDict + as required. For example: + + class Movie(TypedDict, total=False): + title: Required[str] + year: int + + m = Movie( + title='The Matrix', # typechecker error if key is omitted + year=1999, + ) + + There is no runtime checking that a required key is actually provided + when instantiating a related TypedDict. + """) + NotRequired = _RequiredForm( + 'NotRequired', + doc="""A special typing construct to mark a key of a TypedDict as + potentially missing. For example: + + class Movie(TypedDict): + title: str + year: NotRequired[int] + + m = Movie( + title='The Matrix', # typechecker error if key is omitted + year=1999, + ) + """) + + +if hasattr(typing, "Unpack"): # 3.11+ + Unpack = typing.Unpack +elif sys.version_info[:2] >= (3, 9): + class _UnpackSpecialForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + class _UnpackAlias(typing._GenericAlias, _root=True): + __class__ = typing.TypeVar + + @_UnpackSpecialForm + def Unpack(self, parameters): + """A special typing construct to unpack a variadic type. For example: + + Shape = TypeVarTuple('Shape') + Batch = NewType('Batch', int) + + def add_batch_axis( + x: Array[Unpack[Shape]] + ) -> Array[Batch, Unpack[Shape]]: ... + + """ + item = typing._type_check(parameters, f'{self._name} accepts only a single type.') + return _UnpackAlias(self, (item,)) + + def _is_unpack(obj): + return isinstance(obj, _UnpackAlias) + +else: + class _UnpackAlias(typing._GenericAlias, _root=True): + __class__ = typing.TypeVar + + class _UnpackForm(typing._SpecialForm, _root=True): + def __repr__(self): + return 'typing_extensions.' + self._name + + def __getitem__(self, parameters): + item = typing._type_check(parameters, + f'{self._name} accepts only a single type.') + return _UnpackAlias(self, (item,)) + + Unpack = _UnpackForm( + 'Unpack', + doc="""A special typing construct to unpack a variadic type. For example: + + Shape = TypeVarTuple('Shape') + Batch = NewType('Batch', int) + + def add_batch_axis( + x: Array[Unpack[Shape]] + ) -> Array[Batch, Unpack[Shape]]: ... + + """) + + def _is_unpack(obj): + return isinstance(obj, _UnpackAlias) + + +if hasattr(typing, "TypeVarTuple"): # 3.11+ + + # Add default Parameter - PEP 696 + class TypeVarTuple(typing.TypeVarTuple, _DefaultMixin, _root=True): + """Type variable tuple.""" + + def __init__(self, name, *, default=None): + super().__init__(name) + _DefaultMixin.__init__(self, default) + + # for pickling: + try: + def_mod = sys._getframe(1).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): + def_mod = None + if def_mod != 'typing_extensions': + self.__module__ = def_mod + +else: + class TypeVarTuple(_DefaultMixin): + """Type variable tuple. + + Usage:: + + Ts = TypeVarTuple('Ts') + + In the same way that a normal type variable is a stand-in for a single + type such as ``int``, a type variable *tuple* is a stand-in for a *tuple* + type such as ``Tuple[int, str]``. + + Type variable tuples can be used in ``Generic`` declarations. + Consider the following example:: + + class Array(Generic[*Ts]): ... + + The ``Ts`` type variable tuple here behaves like ``tuple[T1, T2]``, + where ``T1`` and ``T2`` are type variables. To use these type variables + as type parameters of ``Array``, we must *unpack* the type variable tuple using + the star operator: ``*Ts``. The signature of ``Array`` then behaves + as if we had simply written ``class Array(Generic[T1, T2]): ...``. + In contrast to ``Generic[T1, T2]``, however, ``Generic[*Shape]`` allows + us to parameterise the class with an *arbitrary* number of type parameters. + + Type variable tuples can be used anywhere a normal ``TypeVar`` can. + This includes class definitions, as shown above, as well as function + signatures and variable annotations:: + + class Array(Generic[*Ts]): + + def __init__(self, shape: Tuple[*Ts]): + self._shape: Tuple[*Ts] = shape + + def get_shape(self) -> Tuple[*Ts]: + return self._shape + + shape = (Height(480), Width(640)) + x: Array[Height, Width] = Array(shape) + y = abs(x) # Inferred type is Array[Height, Width] + z = x + x # ... is Array[Height, Width] + x.get_shape() # ... is tuple[Height, Width] + + """ + + # Trick Generic __parameters__. + __class__ = typing.TypeVar + + def __iter__(self): + yield self.__unpacked__ + + def __init__(self, name, *, default=None): + self.__name__ = name + _DefaultMixin.__init__(self, default) + + # for pickling: + try: + def_mod = sys._getframe(1).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): + def_mod = None + if def_mod != 'typing_extensions': + self.__module__ = def_mod + + self.__unpacked__ = Unpack[self] + + def __repr__(self): + return self.__name__ + + def __hash__(self): + return object.__hash__(self) + + def __eq__(self, other): + return self is other + + def __reduce__(self): + return self.__name__ + + def __init_subclass__(self, *args, **kwds): + if '_root' not in kwds: + raise TypeError("Cannot subclass special typing classes") + + +if hasattr(typing, "reveal_type"): + reveal_type = typing.reveal_type +else: + def reveal_type(__obj: T) -> T: + """Reveal the inferred type of a variable. + + When a static type checker encounters a call to ``reveal_type()``, + it will emit the inferred type of the argument:: + + x: int = 1 + reveal_type(x) + + Running a static type checker (e.g., ``mypy``) on this example + will produce output similar to 'Revealed type is "builtins.int"'. + + At runtime, the function prints the runtime type of the + argument and returns it unchanged. + + """ + print(f"Runtime type is {type(__obj).__name__!r}", file=sys.stderr) + return __obj + + +if hasattr(typing, "assert_never"): + assert_never = typing.assert_never +else: + def assert_never(__arg: Never) -> Never: + """Assert to the type checker that a line of code is unreachable. + + Example:: + + def int_or_str(arg: int | str) -> None: + match arg: + case int(): + print("It's an int") + case str(): + print("It's a str") + case _: + assert_never(arg) + + If a type checker finds that a call to assert_never() is + reachable, it will emit an error. + + At runtime, this throws an exception when called. + + """ + raise AssertionError("Expected code to be unreachable") + + +if hasattr(typing, 'dataclass_transform'): + dataclass_transform = typing.dataclass_transform +else: + def dataclass_transform( + *, + eq_default: bool = True, + order_default: bool = False, + kw_only_default: bool = False, + field_specifiers: typing.Tuple[ + typing.Union[typing.Type[typing.Any], typing.Callable[..., typing.Any]], + ... + ] = (), + **kwargs: typing.Any, + ) -> typing.Callable[[T], T]: + """Decorator that marks a function, class, or metaclass as providing + dataclass-like behavior. + + Example: + + from typing_extensions import dataclass_transform + + _T = TypeVar("_T") + + # Used on a decorator function + @dataclass_transform() + def create_model(cls: type[_T]) -> type[_T]: + ... + return cls + + @create_model + class CustomerModel: + id: int + name: str + + # Used on a base class + @dataclass_transform() + class ModelBase: ... + + class CustomerModel(ModelBase): + id: int + name: str + + # Used on a metaclass + @dataclass_transform() + class ModelMeta(type): ... + + class ModelBase(metaclass=ModelMeta): ... + + class CustomerModel(ModelBase): + id: int + name: str + + Each of the ``CustomerModel`` classes defined in this example will now + behave similarly to a dataclass created with the ``@dataclasses.dataclass`` + decorator. For example, the type checker will synthesize an ``__init__`` + method. + + The arguments to this decorator can be used to customize this behavior: + - ``eq_default`` indicates whether the ``eq`` parameter is assumed to be + True or False if it is omitted by the caller. + - ``order_default`` indicates whether the ``order`` parameter is + assumed to be True or False if it is omitted by the caller. + - ``kw_only_default`` indicates whether the ``kw_only`` parameter is + assumed to be True or False if it is omitted by the caller. + - ``field_specifiers`` specifies a static list of supported classes + or functions that describe fields, similar to ``dataclasses.field()``. + + At runtime, this decorator records its arguments in the + ``__dataclass_transform__`` attribute on the decorated object. + + See PEP 681 for details. + + """ + def decorator(cls_or_fn): + cls_or_fn.__dataclass_transform__ = { + "eq_default": eq_default, + "order_default": order_default, + "kw_only_default": kw_only_default, + "field_specifiers": field_specifiers, + "kwargs": kwargs, + } + return cls_or_fn + return decorator + + +if hasattr(typing, "override"): + override = typing.override +else: + _F = typing.TypeVar("_F", bound=typing.Callable[..., typing.Any]) + + def override(__arg: _F) -> _F: + """Indicate that a method is intended to override a method in a base class. + + Usage: + + class Base: + def method(self) -> None: ... + pass + + class Child(Base): + @override + def method(self) -> None: + super().method() + + When this decorator is applied to a method, the type checker will + validate that it overrides a method with the same name on a base class. + This helps prevent bugs that may occur when a base class is changed + without an equivalent change to a child class. + + See PEP 698 for details. + + """ + return __arg + + +# We have to do some monkey patching to deal with the dual nature of +# Unpack/TypeVarTuple: +# - We want Unpack to be a kind of TypeVar so it gets accepted in +# Generic[Unpack[Ts]] +# - We want it to *not* be treated as a TypeVar for the purposes of +# counting generic parameters, so that when we subscript a generic, +# the runtime doesn't try to substitute the Unpack with the subscripted type. +if not hasattr(typing, "TypeVarTuple"): + typing._collect_type_vars = _collect_type_vars + typing._check_generic = _check_generic + + +# Backport typing.NamedTuple as it exists in Python 3.11. +# In 3.11, the ability to define generic `NamedTuple`s was supported. +# This was explicitly disallowed in 3.9-3.10, and only half-worked in <=3.8. +if sys.version_info >= (3, 11): + NamedTuple = typing.NamedTuple +else: + def _caller(): + try: + return sys._getframe(2).f_globals.get('__name__', '__main__') + except (AttributeError, ValueError): # For platforms without _getframe() + return None + + def _make_nmtuple(name, types, module, defaults=()): + fields = [n for n, t in types] + annotations = {n: typing._type_check(t, f"field {n} annotation must be a type") + for n, t in types} + nm_tpl = collections.namedtuple(name, fields, + defaults=defaults, module=module) + nm_tpl.__annotations__ = nm_tpl.__new__.__annotations__ = annotations + # The `_field_types` attribute was removed in 3.9; + # in earlier versions, it is the same as the `__annotations__` attribute + if sys.version_info < (3, 9): + nm_tpl._field_types = annotations + return nm_tpl + + _prohibited_namedtuple_fields = typing._prohibited + _special_namedtuple_fields = frozenset({'__module__', '__name__', '__annotations__'}) + + class _NamedTupleMeta(type): + def __new__(cls, typename, bases, ns): + assert _NamedTuple in bases + for base in bases: + if base is not _NamedTuple and base is not typing.Generic: + raise TypeError( + 'can only inherit from a NamedTuple type and Generic') + bases = tuple(tuple if base is _NamedTuple else base for base in bases) + types = ns.get('__annotations__', {}) + default_names = [] + for field_name in types: + if field_name in ns: + default_names.append(field_name) + elif default_names: + raise TypeError(f"Non-default namedtuple field {field_name} " + f"cannot follow default field" + f"{'s' if len(default_names) > 1 else ''} " + f"{', '.join(default_names)}") + nm_tpl = _make_nmtuple( + typename, types.items(), + defaults=[ns[n] for n in default_names], + module=ns['__module__'] + ) + nm_tpl.__bases__ = bases + if typing.Generic in bases: + class_getitem = typing.Generic.__class_getitem__.__func__ + nm_tpl.__class_getitem__ = classmethod(class_getitem) + # update from user namespace without overriding special namedtuple attributes + for key in ns: + if key in _prohibited_namedtuple_fields: + raise AttributeError("Cannot overwrite NamedTuple attribute " + key) + elif key not in _special_namedtuple_fields and key not in nm_tpl._fields: + setattr(nm_tpl, key, ns[key]) + if typing.Generic in bases: + nm_tpl.__init_subclass__() + return nm_tpl + + def NamedTuple(__typename, __fields=None, **kwargs): + if __fields is None: + __fields = kwargs.items() + elif kwargs: + raise TypeError("Either list of fields or keywords" + " can be provided to NamedTuple, not both") + return _make_nmtuple(__typename, __fields, module=_caller()) + + NamedTuple.__doc__ = typing.NamedTuple.__doc__ + _NamedTuple = type.__new__(_NamedTupleMeta, 'NamedTuple', (), {}) + + # On 3.8+, alter the signature so that it matches typing.NamedTuple. + # The signature of typing.NamedTuple on >=3.8 is invalid syntax in Python 3.7, + # so just leave the signature as it is on 3.7. + if sys.version_info >= (3, 8): + NamedTuple.__text_signature__ = '(typename, fields=None, /, **kwargs)' + + def _namedtuple_mro_entries(bases): + assert NamedTuple in bases + return (_NamedTuple,) + + NamedTuple.__mro_entries__ = _namedtuple_mro_entries diff --git a/third_party/python/setuptools/pkg_resources/_vendor/zipp.py b/third_party/python/setuptools/pkg_resources/_vendor/zipp.py new file mode 100644 index 0000000000..26b723c1fd --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/_vendor/zipp.py @@ -0,0 +1,329 @@ +import io +import posixpath +import zipfile +import itertools +import contextlib +import sys +import pathlib + +if sys.version_info < (3, 7): + from collections import OrderedDict +else: + OrderedDict = dict + + +__all__ = ['Path'] + + +def _parents(path): + """ + Given a path with elements separated by + posixpath.sep, generate all parents of that path. + + >>> list(_parents('b/d')) + ['b'] + >>> list(_parents('/b/d/')) + ['/b'] + >>> list(_parents('b/d/f/')) + ['b/d', 'b'] + >>> list(_parents('b')) + [] + >>> list(_parents('')) + [] + """ + return itertools.islice(_ancestry(path), 1, None) + + +def _ancestry(path): + """ + Given a path with elements separated by + posixpath.sep, generate all elements of that path + + >>> list(_ancestry('b/d')) + ['b/d', 'b'] + >>> list(_ancestry('/b/d/')) + ['/b/d', '/b'] + >>> list(_ancestry('b/d/f/')) + ['b/d/f', 'b/d', 'b'] + >>> list(_ancestry('b')) + ['b'] + >>> list(_ancestry('')) + [] + """ + path = path.rstrip(posixpath.sep) + while path and path != posixpath.sep: + yield path + path, tail = posixpath.split(path) + + +_dedupe = OrderedDict.fromkeys +"""Deduplicate an iterable in original order""" + + +def _difference(minuend, subtrahend): + """ + Return items in minuend not in subtrahend, retaining order + with O(1) lookup. + """ + return itertools.filterfalse(set(subtrahend).__contains__, minuend) + + +class CompleteDirs(zipfile.ZipFile): + """ + A ZipFile subclass that ensures that implied directories + are always included in the namelist. + """ + + @staticmethod + def _implied_dirs(names): + parents = itertools.chain.from_iterable(map(_parents, names)) + as_dirs = (p + posixpath.sep for p in parents) + return _dedupe(_difference(as_dirs, names)) + + def namelist(self): + names = super(CompleteDirs, self).namelist() + return names + list(self._implied_dirs(names)) + + def _name_set(self): + return set(self.namelist()) + + def resolve_dir(self, name): + """ + If the name represents a directory, return that name + as a directory (with the trailing slash). + """ + names = self._name_set() + dirname = name + '/' + dir_match = name not in names and dirname in names + return dirname if dir_match else name + + @classmethod + def make(cls, source): + """ + Given a source (filename or zipfile), return an + appropriate CompleteDirs subclass. + """ + if isinstance(source, CompleteDirs): + return source + + if not isinstance(source, zipfile.ZipFile): + return cls(_pathlib_compat(source)) + + # Only allow for FastLookup when supplied zipfile is read-only + if 'r' not in source.mode: + cls = CompleteDirs + + source.__class__ = cls + return source + + +class FastLookup(CompleteDirs): + """ + ZipFile subclass to ensure implicit + dirs exist and are resolved rapidly. + """ + + def namelist(self): + with contextlib.suppress(AttributeError): + return self.__names + self.__names = super(FastLookup, self).namelist() + return self.__names + + def _name_set(self): + with contextlib.suppress(AttributeError): + return self.__lookup + self.__lookup = super(FastLookup, self)._name_set() + return self.__lookup + + +def _pathlib_compat(path): + """ + For path-like objects, convert to a filename for compatibility + on Python 3.6.1 and earlier. + """ + try: + return path.__fspath__() + except AttributeError: + return str(path) + + +class Path: + """ + A pathlib-compatible interface for zip files. + + Consider a zip file with this structure:: + + . + ├── a.txt + └── b + ├── c.txt + └── d + └── e.txt + + >>> data = io.BytesIO() + >>> zf = zipfile.ZipFile(data, 'w') + >>> zf.writestr('a.txt', 'content of a') + >>> zf.writestr('b/c.txt', 'content of c') + >>> zf.writestr('b/d/e.txt', 'content of e') + >>> zf.filename = 'mem/abcde.zip' + + Path accepts the zipfile object itself or a filename + + >>> root = Path(zf) + + From there, several path operations are available. + + Directory iteration (including the zip file itself): + + >>> a, b = root.iterdir() + >>> a + Path('mem/abcde.zip', 'a.txt') + >>> b + Path('mem/abcde.zip', 'b/') + + name property: + + >>> b.name + 'b' + + join with divide operator: + + >>> c = b / 'c.txt' + >>> c + Path('mem/abcde.zip', 'b/c.txt') + >>> c.name + 'c.txt' + + Read text: + + >>> c.read_text() + 'content of c' + + existence: + + >>> c.exists() + True + >>> (b / 'missing.txt').exists() + False + + Coercion to string: + + >>> import os + >>> str(c).replace(os.sep, posixpath.sep) + 'mem/abcde.zip/b/c.txt' + + At the root, ``name``, ``filename``, and ``parent`` + resolve to the zipfile. Note these attributes are not + valid and will raise a ``ValueError`` if the zipfile + has no filename. + + >>> root.name + 'abcde.zip' + >>> str(root.filename).replace(os.sep, posixpath.sep) + 'mem/abcde.zip' + >>> str(root.parent) + 'mem' + """ + + __repr = "{self.__class__.__name__}({self.root.filename!r}, {self.at!r})" + + def __init__(self, root, at=""): + """ + Construct a Path from a ZipFile or filename. + + Note: When the source is an existing ZipFile object, + its type (__class__) will be mutated to a + specialized type. If the caller wishes to retain the + original type, the caller should either create a + separate ZipFile object or pass a filename. + """ + self.root = FastLookup.make(root) + self.at = at + + def open(self, mode='r', *args, pwd=None, **kwargs): + """ + Open this entry as text or binary following the semantics + of ``pathlib.Path.open()`` by passing arguments through + to io.TextIOWrapper(). + """ + if self.is_dir(): + raise IsADirectoryError(self) + zip_mode = mode[0] + if not self.exists() and zip_mode == 'r': + raise FileNotFoundError(self) + stream = self.root.open(self.at, zip_mode, pwd=pwd) + if 'b' in mode: + if args or kwargs: + raise ValueError("encoding args invalid for binary operation") + return stream + return io.TextIOWrapper(stream, *args, **kwargs) + + @property + def name(self): + return pathlib.Path(self.at).name or self.filename.name + + @property + def suffix(self): + return pathlib.Path(self.at).suffix or self.filename.suffix + + @property + def suffixes(self): + return pathlib.Path(self.at).suffixes or self.filename.suffixes + + @property + def stem(self): + return pathlib.Path(self.at).stem or self.filename.stem + + @property + def filename(self): + return pathlib.Path(self.root.filename).joinpath(self.at) + + def read_text(self, *args, **kwargs): + with self.open('r', *args, **kwargs) as strm: + return strm.read() + + def read_bytes(self): + with self.open('rb') as strm: + return strm.read() + + def _is_child(self, path): + return posixpath.dirname(path.at.rstrip("/")) == self.at.rstrip("/") + + def _next(self, at): + return self.__class__(self.root, at) + + def is_dir(self): + return not self.at or self.at.endswith("/") + + def is_file(self): + return self.exists() and not self.is_dir() + + def exists(self): + return self.at in self.root._name_set() + + def iterdir(self): + if not self.is_dir(): + raise ValueError("Can't listdir a file") + subs = map(self._next, self.root.namelist()) + return filter(self._is_child, subs) + + def __str__(self): + return posixpath.join(self.root.filename, self.at) + + def __repr__(self): + return self.__repr.format(self=self) + + def joinpath(self, *other): + next = posixpath.join(self.at, *map(_pathlib_compat, other)) + return self._next(self.root.resolve_dir(next)) + + __truediv__ = joinpath + + @property + def parent(self): + if not self.at: + return self.filename.parent + parent_at = posixpath.dirname(self.at.rstrip('/')) + if parent_at: + parent_at += '/' + return self._next(parent_at) diff --git a/third_party/python/setuptools/pkg_resources/extern/__init__.py b/third_party/python/setuptools/pkg_resources/extern/__init__.py new file mode 100644 index 0000000000..948bcc6094 --- /dev/null +++ b/third_party/python/setuptools/pkg_resources/extern/__init__.py @@ -0,0 +1,80 @@ +import importlib.util +import sys + + +class VendorImporter: + """ + A PEP 302 meta path importer for finding optionally-vendored + or otherwise naturally-installed packages from root_name. + """ + + def __init__(self, root_name, vendored_names=(), vendor_pkg=None): + self.root_name = root_name + self.vendored_names = set(vendored_names) + self.vendor_pkg = vendor_pkg or root_name.replace('extern', '_vendor') + + @property + def search_path(self): + """ + Search first the vendor package then as a natural package. + """ + yield self.vendor_pkg + '.' + yield '' + + def _module_matches_namespace(self, fullname): + """Figure out if the target module is vendored.""" + root, base, target = fullname.partition(self.root_name + '.') + return not root and any(map(target.startswith, self.vendored_names)) + + def load_module(self, fullname): + """ + Iterate over the search path to locate and load fullname. + """ + root, base, target = fullname.partition(self.root_name + '.') + for prefix in self.search_path: + try: + extant = prefix + target + __import__(extant) + mod = sys.modules[extant] + sys.modules[fullname] = mod + return mod + except ImportError: + pass + else: + raise ImportError( + "The '{target}' package is required; " + "normally this is bundled with this package so if you get " + "this warning, consult the packager of your " + "distribution.".format(**locals()) + ) + + def create_module(self, spec): + return self.load_module(spec.name) + + def exec_module(self, module): + pass + + def find_spec(self, fullname, path=None, target=None): + """Return a module spec for vendored names.""" + return ( + importlib.util.spec_from_loader(fullname, self) + if self._module_matches_namespace(fullname) + else None + ) + + def install(self): + """ + Install this importer into sys.meta_path if not already present. + """ + if self not in sys.meta_path: + sys.meta_path.append(self) + + +names = ( + 'packaging', + 'platformdirs', + 'jaraco', + 'importlib_resources', + 'more_itertools', +) +VendorImporter(__name__, names).install() |