import abc import os import stat from collections import deque from os import stat_result from typing import (Any, Dict, Iterable, Iterator, List, MutableMapping, Optional, Set, Text, Tuple, TYPE_CHECKING) from . import jsonlib from .utils import git # Cannot do `from ..gitignore import gitignore` because # relative import beyond toplevel throws *ImportError*! from gitignore import gitignore # type: ignore if TYPE_CHECKING: from .manifest import Manifest # avoid cyclic import GitIgnoreCacheType = MutableMapping[bytes, bool] def get_tree(tests_root: Text, manifest: "Manifest", manifest_path: Optional[Text], cache_root: Optional[Text], working_copy: bool = True, rebuild: bool = False) -> "FileSystem": tree = None if cache_root is None: cache_root = os.path.join(tests_root, ".wptcache") if not os.path.exists(cache_root): try: os.makedirs(cache_root) except OSError: cache_root = None if not working_copy: raise ValueError("working_copy=False unsupported") if tree is None: tree = FileSystem(tests_root, manifest.url_base, manifest_path=manifest_path, cache_path=cache_root, rebuild=rebuild) return tree class GitHasher: def __init__(self, path: Text) -> None: self.git = git(path) def _local_changes(self) -> Set[Text]: """get a set of files which have changed between HEAD and working copy""" assert self.git is not None # note that git runs the command with tests_root as the cwd, which may # not be the root of the git repo (e.g., within a browser repo) # # `git diff-index --relative` without a path still compares all tracked # files before non-WPT files are filtered out, which can be slow in # vendor repos. Explicitly pass the CWD (i.e., `tests_root`) as a path # argument to avoid unnecessary diffing. cmd = ["diff-index", "--relative", "--no-renames", "--name-only", "-z", "HEAD", os.curdir] data = self.git(*cmd) return set(data.split("\0")) def hash_cache(self) -> Dict[Text, Optional[Text]]: """ A dict of rel_path -> current git object id if the working tree matches HEAD else None """ hash_cache: Dict[Text, Optional[Text]] = {} if self.git is None: return hash_cache # note that git runs the command with tests_root as the cwd, which may # not be the root of the git repo (e.g., within a browser repo) cmd = ["ls-tree", "-r", "-z", "HEAD"] local_changes = self._local_changes() for result in self.git(*cmd).split("\0")[:-1]: # type: Text data, rel_path = result.rsplit("\t", 1) hash_cache[rel_path] = None if rel_path in local_changes else data.split(" ", 3)[2] return hash_cache class FileSystem: def __init__(self, tests_root: Text, url_base: Text, cache_path: Optional[Text], manifest_path: Optional[Text] = None, rebuild: bool = False) -> None: self.tests_root = tests_root self.url_base = url_base self.ignore_cache = None self.mtime_cache = None tests_root_bytes = tests_root.encode("utf8") if cache_path is not None: if manifest_path is not None: self.mtime_cache = MtimeCache(cache_path, tests_root, manifest_path, rebuild) if gitignore.has_ignore(tests_root_bytes): self.ignore_cache = GitIgnoreCache(cache_path, tests_root, rebuild) self.path_filter = gitignore.PathFilter(tests_root_bytes, extras=[b".git/"], cache=self.ignore_cache) git = GitHasher(tests_root) self.hash_cache = git.hash_cache() def __iter__(self) -> Iterator[Tuple[Text, Optional[Text], bool]]: mtime_cache = self.mtime_cache for dirpath, dirnames, filenames in self.path_filter( walk(self.tests_root.encode("utf8"))): for filename, path_stat in filenames: path = os.path.join(dirpath, filename).decode("utf8") if mtime_cache is None or mtime_cache.updated(path, path_stat): file_hash = self.hash_cache.get(path, None) yield path, file_hash, True else: yield path, None, False def dump_caches(self) -> None: for cache in [self.mtime_cache, self.ignore_cache]: if cache is not None: cache.dump() class CacheFile(metaclass=abc.ABCMeta): def __init__(self, cache_root: Text, tests_root: Text, rebuild: bool = False) -> None: self.tests_root = tests_root if not os.path.exists(cache_root): os.makedirs(cache_root) self.path = os.path.join(cache_root, self.file_name) self.modified = False self.data = self.load(rebuild) @abc.abstractproperty def file_name(self) -> Text: pass def dump(self) -> None: if not self.modified: return with open(self.path, 'w') as f: jsonlib.dump_local(self.data, f) def load(self, rebuild: bool = False) -> Dict[Text, Any]: data: Dict[Text, Any] = {} try: if not rebuild: with open(self.path) as f: try: data = jsonlib.load(f) except ValueError: pass data = self.check_valid(data) except OSError: pass return data def check_valid(self, data: Dict[Text, Any]) -> Dict[Text, Any]: """Check if the cached data is valid and return an updated copy of the cache containing only data that can be used.""" return data class MtimeCache(CacheFile): file_name = "mtime.json" def __init__(self, cache_root: Text, tests_root: Text, manifest_path: Text, rebuild: bool = False) -> None: self.manifest_path = manifest_path super().__init__(cache_root, tests_root, rebuild) def updated(self, rel_path: Text, stat: stat_result) -> bool: """Return a boolean indicating whether the file changed since the cache was last updated. This implicitly updates the cache with the new mtime data.""" mtime = stat.st_mtime if mtime != self.data.get(rel_path): self.modified = True self.data[rel_path] = mtime return True return False def check_valid(self, data: Dict[Any, Any]) -> Dict[Any, Any]: if data.get("/tests_root") != self.tests_root: self.modified = True else: if self.manifest_path is not None and os.path.exists(self.manifest_path): mtime = os.path.getmtime(self.manifest_path) if data.get("/manifest_path") != [self.manifest_path, mtime]: self.modified = True else: self.modified = True if self.modified: data = {} data["/tests_root"] = self.tests_root return data def dump(self) -> None: if self.manifest_path is None: raise ValueError if not os.path.exists(self.manifest_path): return mtime = os.path.getmtime(self.manifest_path) self.data["/manifest_path"] = [self.manifest_path, mtime] self.data["/tests_root"] = self.tests_root super().dump() class GitIgnoreCache(CacheFile, GitIgnoreCacheType): file_name = "gitignore2.json" def check_valid(self, data: Dict[Any, Any]) -> Dict[Any, Any]: ignore_path = os.path.join(self.tests_root, ".gitignore") mtime = os.path.getmtime(ignore_path) if data.get("/gitignore_file") != [ignore_path, mtime]: self.modified = True data = {} data["/gitignore_file"] = [ignore_path, mtime] return data def __contains__(self, key: Any) -> bool: try: key = key.decode("utf-8") except Exception: return False return key in self.data def __getitem__(self, key: bytes) -> bool: real_key = key.decode("utf-8") v = self.data[real_key] assert isinstance(v, bool) return v def __setitem__(self, key: bytes, value: bool) -> None: real_key = key.decode("utf-8") if self.data.get(real_key) != value: self.modified = True self.data[real_key] = value def __delitem__(self, key: bytes) -> None: real_key = key.decode("utf-8") del self.data[real_key] def __iter__(self) -> Iterator[bytes]: return (key.encode("utf-8") for key in self.data) def __len__(self) -> int: return len(self.data) def walk(root: bytes) -> Iterable[Tuple[bytes, List[Tuple[bytes, stat_result]], List[Tuple[bytes, stat_result]]]]: """Re-implementation of os.walk. Returns an iterator over (dirpath, dirnames, filenames), with some semantic differences to os.walk. This has a similar interface to os.walk, with the important difference that instead of lists of filenames and directory names, it yields lists of tuples of the form [(name, stat)] where stat is the result of os.stat for the file. That allows reusing the same stat data in the caller. It also always returns the dirpath relative to the root, with the root iself being returned as the empty string. Unlike os.walk the implementation is not recursive.""" get_stat = os.stat is_dir = stat.S_ISDIR is_link = stat.S_ISLNK join = os.path.join listdir = os.listdir relpath = os.path.relpath root = os.path.abspath(root) stack = deque([(root, b"")]) while stack: dir_path, rel_path = stack.popleft() try: # Note that listdir and error are globals in this module due # to earlier import-*. names = listdir(dir_path) except OSError: continue dirs, non_dirs = [], [] for name in names: path = join(dir_path, name) try: path_stat = get_stat(path) except OSError: continue if is_dir(path_stat.st_mode): dirs.append((name, path_stat)) else: non_dirs.append((name, path_stat)) yield rel_path, dirs, non_dirs for name, path_stat in dirs: new_path = join(dir_path, name) if not is_link(path_stat.st_mode): stack.append((new_path, relpath(new_path, root)))