diff options
Diffstat (limited to 'third_party/python/ply')
58 files changed, 12753 insertions, 0 deletions
diff --git a/third_party/python/ply/ANNOUNCE b/third_party/python/ply/ANNOUNCE new file mode 100644 index 0000000000..c430051cf4 --- /dev/null +++ b/third_party/python/ply/ANNOUNCE @@ -0,0 +1,40 @@ +January 31, 2017 + + Announcing : PLY-3.10 (Python Lex-Yacc) + + http://www.dabeaz.com/ply + +I'm pleased to announce PLY-3.10--a pure Python implementation of the +common parsing tools lex and yacc. PLY-3.10 is a minor bug fix +release. It supports both Python 2 and Python 3. + +If you are new to PLY, here are a few highlights: + +- PLY is closely modeled after traditional lex/yacc. If you know how + to use these or similar tools in other languages, you will find + PLY to be comparable. + +- PLY provides very extensive error reporting and diagnostic + information to assist in parser construction. The original + implementation was developed for instructional purposes. As + a result, the system tries to identify the most common types + of errors made by novice users. + +- PLY provides full support for empty productions, error recovery, + precedence rules, and ambiguous grammars. + +- Parsing is based on LR-parsing which is fast, memory efficient, + better suited to large grammars, and which has a number of nice + properties when dealing with syntax errors and other parsing + problems. Currently, PLY can build its parsing tables using + either SLR or LALR(1) algorithms. + +More information about PLY can be obtained on the PLY webpage at: + + http://www.dabeaz.com/ply + +PLY is freely available. + +Cheers, + +David Beazley (http://www.dabeaz.com)
\ No newline at end of file diff --git a/third_party/python/ply/CHANGES b/third_party/python/ply/CHANGES new file mode 100644 index 0000000000..815c23184e --- /dev/null +++ b/third_party/python/ply/CHANGES @@ -0,0 +1,1394 @@ +Version 3.10 +--------------------- +01/31/17: beazley + Changed grammar signature computation to not involve hashing + functions. Parts are just combined into a big string. + +10/07/16: beazley + Fixed Issue #101: Incorrect shift-reduce conflict resolution with + precedence specifier. + + PLY was incorrectly resolving shift-reduce conflicts in certain + cases. For example, in the example/calc/calc.py example, you + could trigger it doing this: + + calc > -3 - 4 + 1 (correct answer should be -7) + calc > + + Issue and suggested patch contributed by https://github.com/RomaVis + +Version 3.9 +--------------------- +08/30/16: beazley + Exposed the parser state number as the parser.state attribute + in productions and error functions. For example: + + def p_somerule(p): + ''' + rule : A B C + ''' + print('State:', p.parser.state) + + May address issue #65 (publish current state in error callback). + +08/30/16: beazley + Fixed Issue #88. Python3 compatibility with ply/cpp. + +08/30/16: beazley + Fixed Issue #93. Ply can crash if SyntaxError is raised inside + a production. Not actually sure if the original implementation + worked as documented at all. Yacc has been modified to follow + the spec as outlined in the CHANGES noted for 11/27/07 below. + +08/30/16: beazley + Fixed Issue #97. Failure with code validation when the original + source files aren't present. Validation step now ignores + the missing file. + +08/30/16: beazley + Minor fixes to version numbers. + +Version 3.8 +--------------------- +10/02/15: beazley + Fixed issues related to Python 3.5. Patch contributed by Barry Warsaw. + +Version 3.7 +--------------------- +08/25/15: beazley + Fixed problems when reading table files from pickled data. + +05/07/15: beazley + Fixed regression in handling of table modules if specified as module + objects. See https://github.com/dabeaz/ply/issues/63 + +Version 3.6 +--------------------- +04/25/15: beazley + If PLY is unable to create the 'parser.out' or 'parsetab.py' files due + to permission issues, it now just issues a warning message and + continues to operate. This could happen if a module using PLY + is installed in a funny way where tables have to be regenerated, but + for whatever reason, the user doesn't have write permission on + the directory where PLY wants to put them. + +04/24/15: beazley + Fixed some issues related to use of packages and table file + modules. Just to emphasize, PLY now generates its special + files such as 'parsetab.py' and 'lextab.py' in the *SAME* + directory as the source file that uses lex() and yacc(). + + If for some reason, you want to change the name of the table + module, use the tabmodule and lextab options: + + lexer = lex.lex(lextab='spamlextab') + parser = yacc.yacc(tabmodule='spamparsetab') + + If you specify a simple name as shown, the module will still be + created in the same directory as the file invoking lex() or yacc(). + If you want the table files to be placed into a different package, + then give a fully qualified package name. For example: + + lexer = lex.lex(lextab='pkgname.files.lextab') + parser = yacc.yacc(tabmodule='pkgname.files.parsetab') + + For this to work, 'pkgname.files' must already exist as a valid + Python package (i.e., the directories must already exist and be + set up with the proper __init__.py files, etc.). + +Version 3.5 +--------------------- +04/21/15: beazley + Added support for defaulted_states in the parser. A + defaulted_state is a state where the only legal action is a + reduction of a single grammar rule across all valid input + tokens. For such states, the rule is reduced and the + reading of the next lookahead token is delayed until it is + actually needed at a later point in time. + + This delay in consuming the next lookahead token is a + potentially important feature in advanced parsing + applications that require tight interaction between the + lexer and the parser. For example, a grammar rule change + modify the lexer state upon reduction and have such changes + take effect before the next input token is read. + + *** POTENTIAL INCOMPATIBILITY *** + One potential danger of defaulted_states is that syntax + errors might be deferred to a a later point of processing + than where they were detected in past versions of PLY. + Thus, it's possible that your error handling could change + slightly on the same inputs. defaulted_states do not change + the overall parsing of the input (i.e., the same grammar is + accepted). + + If for some reason, you need to disable defaulted states, + you can do this: + + parser = yacc.yacc() + parser.defaulted_states = {} + +04/21/15: beazley + Fixed debug logging in the parser. It wasn't properly reporting goto states + on grammar rule reductions. + +04/20/15: beazley + Added actions to be defined to character literals (Issue #32). For example: + + literals = [ '{', '}' ] + + def t_lbrace(t): + r'\{' + # Some action + t.type = '{' + return t + + def t_rbrace(t): + r'\}' + # Some action + t.type = '}' + return t + +04/19/15: beazley + Import of the 'parsetab.py' file is now constrained to only consider the + directory specified by the outputdir argument to yacc(). If not supplied, + the import will only consider the directory in which the grammar is defined. + This should greatly reduce problems with the wrong parsetab.py file being + imported by mistake. For example, if it's found somewhere else on the path + by accident. + + *** POTENTIAL INCOMPATIBILITY *** It's possible that this might break some + packaging/deployment setup if PLY was instructed to place its parsetab.py + in a different location. You'll have to specify a proper outputdir= argument + to yacc() to fix this if needed. + +04/19/15: beazley + Changed default output directory to be the same as that in which the + yacc grammar is defined. If your grammar is in a file 'calc.py', + then the parsetab.py and parser.out files should be generated in the + same directory as that file. The destination directory can be changed + using the outputdir= argument to yacc(). + +04/19/15: beazley + Changed the parsetab.py file signature slightly so that the parsetab won't + regenerate if created on a different major version of Python (ie., a + parsetab created on Python 2 will work with Python 3). + +04/16/15: beazley + Fixed Issue #44 call_errorfunc() should return the result of errorfunc() + +04/16/15: beazley + Support for versions of Python <2.7 is officially dropped. PLY may work, but + the unit tests requires Python 2.7 or newer. + +04/16/15: beazley + Fixed bug related to calling yacc(start=...). PLY wasn't regenerating the + table file correctly for this case. + +04/16/15: beazley + Added skipped tests for PyPy and Java. Related to use of Python's -O option. + +05/29/13: beazley + Added filter to make unit tests pass under 'python -3'. + Reported by Neil Muller. + +05/29/13: beazley + Fixed CPP_INTEGER regex in ply/cpp.py (Issue 21). + Reported by @vbraun. + +05/29/13: beazley + Fixed yacc validation bugs when from __future__ import unicode_literals + is being used. Reported by Kenn Knowles. + +05/29/13: beazley + Added support for Travis-CI. Contributed by Kenn Knowles. + +05/29/13: beazley + Added a .gitignore file. Suggested by Kenn Knowles. + +05/29/13: beazley + Fixed validation problems for source files that include a + different source code encoding specifier. Fix relies on + the inspect module. Should work on Python 2.6 and newer. + Not sure about older versions of Python. + Contributed by Michael Droettboom + +05/21/13: beazley + Fixed unit tests for yacc to eliminate random failures due to dict hash value + randomization in Python 3.3 + Reported by Arfrever + +10/15/12: beazley + Fixed comment whitespace processing bugs in ply/cpp.py. + Reported by Alexei Pososin. + +10/15/12: beazley + Fixed token names in ply/ctokens.py to match rule names. + Reported by Alexei Pososin. + +04/26/12: beazley + Changes to functions available in panic mode error recover. In previous versions + of PLY, the following global functions were available for use in the p_error() rule: + + yacc.errok() # Reset error state + yacc.token() # Get the next token + yacc.restart() # Reset the parsing stack + + The use of global variables was problematic for code involving multiple parsers + and frankly was a poor design overall. These functions have been moved to methods + of the parser instance created by the yacc() function. You should write code like + this: + + def p_error(p): + ... + parser.errok() + + parser = yacc.yacc() + + *** POTENTIAL INCOMPATIBILITY *** The original global functions now issue a + DeprecationWarning. + +04/19/12: beazley + Fixed some problems with line and position tracking and the use of error + symbols. If you have a grammar rule involving an error rule like this: + + def p_assignment_bad(p): + '''assignment : location EQUALS error SEMI''' + ... + + You can now do line and position tracking on the error token. For example: + + def p_assignment_bad(p): + '''assignment : location EQUALS error SEMI''' + start_line = p.lineno(3) + start_pos = p.lexpos(3) + + If the trackng=True option is supplied to parse(), you can additionally get + spans: + + def p_assignment_bad(p): + '''assignment : location EQUALS error SEMI''' + start_line, end_line = p.linespan(3) + start_pos, end_pos = p.lexspan(3) + + Note that error handling is still a hairy thing in PLY. This won't work + unless your lexer is providing accurate information. Please report bugs. + Suggested by a bug reported by Davis Herring. + +04/18/12: beazley + Change to doc string handling in lex module. Regex patterns are now first + pulled from a function's .regex attribute. If that doesn't exist, then + .doc is checked as a fallback. The @TOKEN decorator now sets the .regex + attribute of a function instead of its doc string. + Changed suggested by Kristoffer Ellersgaard Koch. + +04/18/12: beazley + Fixed issue #1: Fixed _tabversion. It should use __tabversion__ instead of __version__ + Reported by Daniele Tricoli + +04/18/12: beazley + Fixed issue #8: Literals empty list causes IndexError + Reported by Walter Nissen. + +04/18/12: beazley + Fixed issue #12: Typo in code snippet in documentation + Reported by florianschanda. + +04/18/12: beazley + Fixed issue #10: Correctly escape t_XOREQUAL pattern. + Reported by Andy Kittner. + +Version 3.4 +--------------------- +02/17/11: beazley + Minor patch to make cpp.py compatible with Python 3. Note: This + is an experimental file not currently used by the rest of PLY. + +02/17/11: beazley + Fixed setup.py trove classifiers to properly list PLY as + Python 3 compatible. + +01/02/11: beazley + Migration of repository to github. + +Version 3.3 +----------------------------- +08/25/09: beazley + Fixed issue 15 related to the set_lineno() method in yacc. Reported by + mdsherry. + +08/25/09: beazley + Fixed a bug related to regular expression compilation flags not being + properly stored in lextab.py files created by the lexer when running + in optimize mode. Reported by Bruce Frederiksen. + + +Version 3.2 +----------------------------- +03/24/09: beazley + Added an extra check to not print duplicated warning messages + about reduce/reduce conflicts. + +03/24/09: beazley + Switched PLY over to a BSD-license. + +03/23/09: beazley + Performance optimization. Discovered a few places to make + speedups in LR table generation. + +03/23/09: beazley + New warning message. PLY now warns about rules never + reduced due to reduce/reduce conflicts. Suggested by + Bruce Frederiksen. + +03/23/09: beazley + Some clean-up of warning messages related to reduce/reduce errors. + +03/23/09: beazley + Added a new picklefile option to yacc() to write the parsing + tables to a filename using the pickle module. Here is how + it works: + + yacc(picklefile="parsetab.p") + + This option can be used if the normal parsetab.py file is + extremely large. For example, on jython, it is impossible + to read parsing tables if the parsetab.py exceeds a certain + threshold. + + The filename supplied to the picklefile option is opened + relative to the current working directory of the Python + interpreter. If you need to refer to the file elsewhere, + you will need to supply an absolute or relative path. + + For maximum portability, the pickle file is written + using protocol 0. + +03/13/09: beazley + Fixed a bug in parser.out generation where the rule numbers + where off by one. + +03/13/09: beazley + Fixed a string formatting bug with one of the error messages. + Reported by Richard Reitmeyer + +Version 3.1 +----------------------------- +02/28/09: beazley + Fixed broken start argument to yacc(). PLY-3.0 broke this + feature by accident. + +02/28/09: beazley + Fixed debugging output. yacc() no longer reports shift/reduce + or reduce/reduce conflicts if debugging is turned off. This + restores similar behavior in PLY-2.5. Reported by Andrew Waters. + +Version 3.0 +----------------------------- +02/03/09: beazley + Fixed missing lexer attribute on certain tokens when + invoking the parser p_error() function. Reported by + Bart Whiteley. + +02/02/09: beazley + The lex() command now does all error-reporting and diagonistics + using the logging module interface. Pass in a Logger object + using the errorlog parameter to specify a different logger. + +02/02/09: beazley + Refactored ply.lex to use a more object-oriented and organized + approach to collecting lexer information. + +02/01/09: beazley + Removed the nowarn option from lex(). All output is controlled + by passing in a logger object. Just pass in a logger with a high + level setting to suppress output. This argument was never + documented to begin with so hopefully no one was relying upon it. + +02/01/09: beazley + Discovered and removed a dead if-statement in the lexer. This + resulted in a 6-7% speedup in lexing when I tested it. + +01/13/09: beazley + Minor change to the procedure for signalling a syntax error in a + production rule. A normal SyntaxError exception should be raised + instead of yacc.SyntaxError. + +01/13/09: beazley + Added a new method p.set_lineno(n,lineno) that can be used to set the + line number of symbol n in grammar rules. This simplifies manual + tracking of line numbers. + +01/11/09: beazley + Vastly improved debugging support for yacc.parse(). Instead of passing + debug as an integer, you can supply a Logging object (see the logging + module). Messages will be generated at the ERROR, INFO, and DEBUG + logging levels, each level providing progressively more information. + The debugging trace also shows states, grammar rule, values passed + into grammar rules, and the result of each reduction. + +01/09/09: beazley + The yacc() command now does all error-reporting and diagnostics using + the interface of the logging module. Use the errorlog parameter to + specify a logging object for error messages. Use the debuglog parameter + to specify a logging object for the 'parser.out' output. + +01/09/09: beazley + *HUGE* refactoring of the the ply.yacc() implementation. The high-level + user interface is backwards compatible, but the internals are completely + reorganized into classes. No more global variables. The internals + are also more extensible. For example, you can use the classes to + construct a LALR(1) parser in an entirely different manner than + what is currently the case. Documentation is forthcoming. + +01/07/09: beazley + Various cleanup and refactoring of yacc internals. + +01/06/09: beazley + Fixed a bug with precedence assignment. yacc was assigning the precedence + each rule based on the left-most token, when in fact, it should have been + using the right-most token. Reported by Bruce Frederiksen. + +11/27/08: beazley + Numerous changes to support Python 3.0 including removal of deprecated + statements (e.g., has_key) and the additional of compatibility code + to emulate features from Python 2 that have been removed, but which + are needed. Fixed the unit testing suite to work with Python 3.0. + The code should be backwards compatible with Python 2. + +11/26/08: beazley + Loosened the rules on what kind of objects can be passed in as the + "module" parameter to lex() and yacc(). Previously, you could only use + a module or an instance. Now, PLY just uses dir() to get a list of + symbols on whatever the object is without regard for its type. + +11/26/08: beazley + Changed all except: statements to be compatible with Python2.x/3.x syntax. + +11/26/08: beazley + Changed all raise Exception, value statements to raise Exception(value) for + forward compatibility. + +11/26/08: beazley + Removed all print statements from lex and yacc, using sys.stdout and sys.stderr + directly. Preparation for Python 3.0 support. + +11/04/08: beazley + Fixed a bug with referring to symbols on the the parsing stack using negative + indices. + +05/29/08: beazley + Completely revamped the testing system to use the unittest module for everything. + Added additional tests to cover new errors/warnings. + +Version 2.5 +----------------------------- +05/28/08: beazley + Fixed a bug with writing lex-tables in optimized mode and start states. + Reported by Kevin Henry. + +Version 2.4 +----------------------------- +05/04/08: beazley + A version number is now embedded in the table file signature so that + yacc can more gracefully accomodate changes to the output format + in the future. + +05/04/08: beazley + Removed undocumented .pushback() method on grammar productions. I'm + not sure this ever worked and can't recall ever using it. Might have + been an abandoned idea that never really got fleshed out. This + feature was never described or tested so removing it is hopefully + harmless. + +05/04/08: beazley + Added extra error checking to yacc() to detect precedence rules defined + for undefined terminal symbols. This allows yacc() to detect a potential + problem that can be really tricky to debug if no warning message or error + message is generated about it. + +05/04/08: beazley + lex() now has an outputdir that can specify the output directory for + tables when running in optimize mode. For example: + + lexer = lex.lex(optimize=True, lextab="ltab", outputdir="foo/bar") + + The behavior of specifying a table module and output directory are + more aligned with the behavior of yacc(). + +05/04/08: beazley + [Issue 9] + Fixed filename bug in when specifying the modulename in lex() and yacc(). + If you specified options such as the following: + + parser = yacc.yacc(tabmodule="foo.bar.parsetab",outputdir="foo/bar") + + yacc would create a file "foo.bar.parsetab.py" in the given directory. + Now, it simply generates a file "parsetab.py" in that directory. + Bug reported by cptbinho. + +05/04/08: beazley + Slight modification to lex() and yacc() to allow their table files + to be loaded from a previously loaded module. This might make + it easier to load the parsing tables from a complicated package + structure. For example: + + import foo.bar.spam.parsetab as parsetab + parser = yacc.yacc(tabmodule=parsetab) + + Note: lex and yacc will never regenerate the table file if used + in the form---you will get a warning message instead. + This idea suggested by Brian Clapper. + + +04/28/08: beazley + Fixed a big with p_error() functions being picked up correctly + when running in yacc(optimize=1) mode. Patch contributed by + Bart Whiteley. + +02/28/08: beazley + Fixed a bug with 'nonassoc' precedence rules. Basically the + non-precedence was being ignored and not producing the correct + run-time behavior in the parser. + +02/16/08: beazley + Slight relaxation of what the input() method to a lexer will + accept as a string. Instead of testing the input to see + if the input is a string or unicode string, it checks to see + if the input object looks like it contains string data. + This change makes it possible to pass string-like objects + in as input. For example, the object returned by mmap. + + import mmap, os + data = mmap.mmap(os.open(filename,os.O_RDONLY), + os.path.getsize(filename), + access=mmap.ACCESS_READ) + lexer.input(data) + + +11/29/07: beazley + Modification of ply.lex to allow token functions to aliased. + This is subtle, but it makes it easier to create libraries and + to reuse token specifications. For example, suppose you defined + a function like this: + + def number(t): + r'\d+' + t.value = int(t.value) + return t + + This change would allow you to define a token rule as follows: + + t_NUMBER = number + + In this case, the token type will be set to 'NUMBER' and use + the associated number() function to process tokens. + +11/28/07: beazley + Slight modification to lex and yacc to grab symbols from both + the local and global dictionaries of the caller. This + modification allows lexers and parsers to be defined using + inner functions and closures. + +11/28/07: beazley + Performance optimization: The lexer.lexmatch and t.lexer + attributes are no longer set for lexer tokens that are not + defined by functions. The only normal use of these attributes + would be in lexer rules that need to perform some kind of + special processing. Thus, it doesn't make any sense to set + them on every token. + + *** POTENTIAL INCOMPATIBILITY *** This might break code + that is mucking around with internal lexer state in some + sort of magical way. + +11/27/07: beazley + Added the ability to put the parser into error-handling mode + from within a normal production. To do this, simply raise + a yacc.SyntaxError exception like this: + + def p_some_production(p): + 'some_production : prod1 prod2' + ... + raise yacc.SyntaxError # Signal an error + + A number of things happen after this occurs: + + - The last symbol shifted onto the symbol stack is discarded + and parser state backed up to what it was before the + the rule reduction. + + - The current lookahead symbol is saved and replaced by + the 'error' symbol. + + - The parser enters error recovery mode where it tries + to either reduce the 'error' rule or it starts + discarding items off of the stack until the parser + resets. + + When an error is manually set, the parser does *not* call + the p_error() function (if any is defined). + *** NEW FEATURE *** Suggested on the mailing list + +11/27/07: beazley + Fixed structure bug in examples/ansic. Reported by Dion Blazakis. + +11/27/07: beazley + Fixed a bug in the lexer related to start conditions and ignored + token rules. If a rule was defined that changed state, but + returned no token, the lexer could be left in an inconsistent + state. Reported by + +11/27/07: beazley + Modified setup.py to support Python Eggs. Patch contributed by + Simon Cross. + +11/09/07: beazely + Fixed a bug in error handling in yacc. If a syntax error occurred and the + parser rolled the entire parse stack back, the parser would be left in in + inconsistent state that would cause it to trigger incorrect actions on + subsequent input. Reported by Ton Biegstraaten, Justin King, and others. + +11/09/07: beazley + Fixed a bug when passing empty input strings to yacc.parse(). This + would result in an error message about "No input given". Reported + by Andrew Dalke. + +Version 2.3 +----------------------------- +02/20/07: beazley + Fixed a bug with character literals if the literal '.' appeared as the + last symbol of a grammar rule. Reported by Ales Smrcka. + +02/19/07: beazley + Warning messages are now redirected to stderr instead of being printed + to standard output. + +02/19/07: beazley + Added a warning message to lex.py if it detects a literal backslash + character inside the t_ignore declaration. This is to help + problems that might occur if someone accidentally defines t_ignore + as a Python raw string. For example: + + t_ignore = r' \t' + + The idea for this is from an email I received from David Cimimi who + reported bizarre behavior in lexing as a result of defining t_ignore + as a raw string by accident. + +02/18/07: beazley + Performance improvements. Made some changes to the internal + table organization and LR parser to improve parsing performance. + +02/18/07: beazley + Automatic tracking of line number and position information must now be + enabled by a special flag to parse(). For example: + + yacc.parse(data,tracking=True) + + In many applications, it's just not that important to have the + parser automatically track all line numbers. By making this an + optional feature, it allows the parser to run significantly faster + (more than a 20% speed increase in many cases). Note: positional + information is always available for raw tokens---this change only + applies to positional information associated with nonterminal + grammar symbols. + *** POTENTIAL INCOMPATIBILITY *** + +02/18/07: beazley + Yacc no longer supports extended slices of grammar productions. + However, it does support regular slices. For example: + + def p_foo(p): + '''foo: a b c d e''' + p[0] = p[1:3] + + This change is a performance improvement to the parser--it streamlines + normal access to the grammar values since slices are now handled in + a __getslice__() method as opposed to __getitem__(). + +02/12/07: beazley + Fixed a bug in the handling of token names when combined with + start conditions. Bug reported by Todd O'Bryan. + +Version 2.2 +------------------------------ +11/01/06: beazley + Added lexpos() and lexspan() methods to grammar symbols. These + mirror the same functionality of lineno() and linespan(). For + example: + + def p_expr(p): + 'expr : expr PLUS expr' + p.lexpos(1) # Lexing position of left-hand-expression + p.lexpos(1) # Lexing position of PLUS + start,end = p.lexspan(3) # Lexing range of right hand expression + +11/01/06: beazley + Minor change to error handling. The recommended way to skip characters + in the input is to use t.lexer.skip() as shown here: + + def t_error(t): + print "Illegal character '%s'" % t.value[0] + t.lexer.skip(1) + + The old approach of just using t.skip(1) will still work, but won't + be documented. + +10/31/06: beazley + Discarded tokens can now be specified as simple strings instead of + functions. To do this, simply include the text "ignore_" in the + token declaration. For example: + + t_ignore_cppcomment = r'//.*' + + Previously, this had to be done with a function. For example: + + def t_ignore_cppcomment(t): + r'//.*' + pass + + If start conditions/states are being used, state names should appear + before the "ignore_" text. + +10/19/06: beazley + The Lex module now provides support for flex-style start conditions + as described at http://www.gnu.org/software/flex/manual/html_chapter/flex_11.html. + Please refer to this document to understand this change note. Refer to + the PLY documentation for PLY-specific explanation of how this works. + + To use start conditions, you first need to declare a set of states in + your lexer file: + + states = ( + ('foo','exclusive'), + ('bar','inclusive') + ) + + This serves the same role as the %s and %x specifiers in flex. + + One a state has been declared, tokens for that state can be + declared by defining rules of the form t_state_TOK. For example: + + t_PLUS = '\+' # Rule defined in INITIAL state + t_foo_NUM = '\d+' # Rule defined in foo state + t_bar_NUM = '\d+' # Rule defined in bar state + + t_foo_bar_NUM = '\d+' # Rule defined in both foo and bar + t_ANY_NUM = '\d+' # Rule defined in all states + + In addition to defining tokens for each state, the t_ignore and t_error + specifications can be customized for specific states. For example: + + t_foo_ignore = " " # Ignored characters for foo state + def t_bar_error(t): + # Handle errors in bar state + + With token rules, the following methods can be used to change states + + def t_TOKNAME(t): + t.lexer.begin('foo') # Begin state 'foo' + t.lexer.push_state('foo') # Begin state 'foo', push old state + # onto a stack + t.lexer.pop_state() # Restore previous state + t.lexer.current_state() # Returns name of current state + + These methods mirror the BEGIN(), yy_push_state(), yy_pop_state(), and + yy_top_state() functions in flex. + + The use of start states can be used as one way to write sub-lexers. + For example, the lexer or parser might instruct the lexer to start + generating a different set of tokens depending on the context. + + example/yply/ylex.py shows the use of start states to grab C/C++ + code fragments out of traditional yacc specification files. + + *** NEW FEATURE *** Suggested by Daniel Larraz with whom I also + discussed various aspects of the design. + +10/19/06: beazley + Minor change to the way in which yacc.py was reporting shift/reduce + conflicts. Although the underlying LALR(1) algorithm was correct, + PLY was under-reporting the number of conflicts compared to yacc/bison + when precedence rules were in effect. This change should make PLY + report the same number of conflicts as yacc. + +10/19/06: beazley + Modified yacc so that grammar rules could also include the '-' + character. For example: + + def p_expr_list(p): + 'expression-list : expression-list expression' + + Suggested by Oldrich Jedlicka. + +10/18/06: beazley + Attribute lexer.lexmatch added so that token rules can access the re + match object that was generated. For example: + + def t_FOO(t): + r'some regex' + m = t.lexer.lexmatch + # Do something with m + + + This may be useful if you want to access named groups specified within + the regex for a specific token. Suggested by Oldrich Jedlicka. + +10/16/06: beazley + Changed the error message that results if an illegal character + is encountered and no default error function is defined in lex. + The exception is now more informative about the actual cause of + the error. + +Version 2.1 +------------------------------ +10/02/06: beazley + The last Lexer object built by lex() can be found in lex.lexer. + The last Parser object built by yacc() can be found in yacc.parser. + +10/02/06: beazley + New example added: examples/yply + + This example uses PLY to convert Unix-yacc specification files to + PLY programs with the same grammar. This may be useful if you + want to convert a grammar from bison/yacc to use with PLY. + +10/02/06: beazley + Added support for a start symbol to be specified in the yacc + input file itself. Just do this: + + start = 'name' + + where 'name' matches some grammar rule. For example: + + def p_name(p): + 'name : A B C' + ... + + This mirrors the functionality of the yacc %start specifier. + +09/30/06: beazley + Some new examples added.: + + examples/GardenSnake : A simple indentation based language similar + to Python. Shows how you might handle + whitespace. Contributed by Andrew Dalke. + + examples/BASIC : An implementation of 1964 Dartmouth BASIC. + Contributed by Dave against his better + judgement. + +09/28/06: beazley + Minor patch to allow named groups to be used in lex regular + expression rules. For example: + + t_QSTRING = r'''(?P<quote>['"]).*?(?P=quote)''' + + Patch submitted by Adam Ring. + +09/28/06: beazley + LALR(1) is now the default parsing method. To use SLR, use + yacc.yacc(method="SLR"). Note: there is no performance impact + on parsing when using LALR(1) instead of SLR. However, constructing + the parsing tables will take a little longer. + +09/26/06: beazley + Change to line number tracking. To modify line numbers, modify + the line number of the lexer itself. For example: + + def t_NEWLINE(t): + r'\n' + t.lexer.lineno += 1 + + This modification is both cleanup and a performance optimization. + In past versions, lex was monitoring every token for changes in + the line number. This extra processing is unnecessary for a vast + majority of tokens. Thus, this new approach cleans it up a bit. + + *** POTENTIAL INCOMPATIBILITY *** + You will need to change code in your lexer that updates the line + number. For example, "t.lineno += 1" becomes "t.lexer.lineno += 1" + +09/26/06: beazley + Added the lexing position to tokens as an attribute lexpos. This + is the raw index into the input text at which a token appears. + This information can be used to compute column numbers and other + details (e.g., scan backwards from lexpos to the first newline + to get a column position). + +09/25/06: beazley + Changed the name of the __copy__() method on the Lexer class + to clone(). This is used to clone a Lexer object (e.g., if + you're running different lexers at the same time). + +09/21/06: beazley + Limitations related to the use of the re module have been eliminated. + Several users reported problems with regular expressions exceeding + more than 100 named groups. To solve this, lex.py is now capable + of automatically splitting its master regular regular expression into + smaller expressions as needed. This should, in theory, make it + possible to specify an arbitrarily large number of tokens. + +09/21/06: beazley + Improved error checking in lex.py. Rules that match the empty string + are now rejected (otherwise they cause the lexer to enter an infinite + loop). An extra check for rules containing '#' has also been added. + Since lex compiles regular expressions in verbose mode, '#' is interpreted + as a regex comment, it is critical to use '\#' instead. + +09/18/06: beazley + Added a @TOKEN decorator function to lex.py that can be used to + define token rules where the documentation string might be computed + in some way. + + digit = r'([0-9])' + nondigit = r'([_A-Za-z])' + identifier = r'(' + nondigit + r'(' + digit + r'|' + nondigit + r')*)' + + from ply.lex import TOKEN + + @TOKEN(identifier) + def t_ID(t): + # Do whatever + + The @TOKEN decorator merely sets the documentation string of the + associated token function as needed for lex to work. + + Note: An alternative solution is the following: + + def t_ID(t): + # Do whatever + + t_ID.__doc__ = identifier + + Note: Decorators require the use of Python 2.4 or later. If compatibility + with old versions is needed, use the latter solution. + + The need for this feature was suggested by Cem Karan. + +09/14/06: beazley + Support for single-character literal tokens has been added to yacc. + These literals must be enclosed in quotes. For example: + + def p_expr(p): + "expr : expr '+' expr" + ... + + def p_expr(p): + 'expr : expr "-" expr' + ... + + In addition to this, it is necessary to tell the lexer module about + literal characters. This is done by defining the variable 'literals' + as a list of characters. This should be defined in the module that + invokes the lex.lex() function. For example: + + literals = ['+','-','*','/','(',')','='] + + or simply + + literals = '+=*/()=' + + It is important to note that literals can only be a single character. + When the lexer fails to match a token using its normal regular expression + rules, it will check the current character against the literal list. + If found, it will be returned with a token type set to match the literal + character. Otherwise, an illegal character will be signalled. + + +09/14/06: beazley + Modified PLY to install itself as a proper Python package called 'ply'. + This will make it a little more friendly to other modules. This + changes the usage of PLY only slightly. Just do this to import the + modules + + import ply.lex as lex + import ply.yacc as yacc + + Alternatively, you can do this: + + from ply import * + + Which imports both the lex and yacc modules. + Change suggested by Lee June. + +09/13/06: beazley + Changed the handling of negative indices when used in production rules. + A negative production index now accesses already parsed symbols on the + parsing stack. For example, + + def p_foo(p): + "foo: A B C D" + print p[1] # Value of 'A' symbol + print p[2] # Value of 'B' symbol + print p[-1] # Value of whatever symbol appears before A + # on the parsing stack. + + p[0] = some_val # Sets the value of the 'foo' grammer symbol + + This behavior makes it easier to work with embedded actions within the + parsing rules. For example, in C-yacc, it is possible to write code like + this: + + bar: A { printf("seen an A = %d\n", $1); } B { do_stuff; } + + In this example, the printf() code executes immediately after A has been + parsed. Within the embedded action code, $1 refers to the A symbol on + the stack. + + To perform this equivalent action in PLY, you need to write a pair + of rules like this: + + def p_bar(p): + "bar : A seen_A B" + do_stuff + + def p_seen_A(p): + "seen_A :" + print "seen an A =", p[-1] + + The second rule "seen_A" is merely a empty production which should be + reduced as soon as A is parsed in the "bar" rule above. The use + of the negative index p[-1] is used to access whatever symbol appeared + before the seen_A symbol. + + This feature also makes it possible to support inherited attributes. + For example: + + def p_decl(p): + "decl : scope name" + + def p_scope(p): + """scope : GLOBAL + | LOCAL""" + p[0] = p[1] + + def p_name(p): + "name : ID" + if p[-1] == "GLOBAL": + # ... + else if p[-1] == "LOCAL": + #... + + In this case, the name rule is inheriting an attribute from the + scope declaration that precedes it. + + *** POTENTIAL INCOMPATIBILITY *** + If you are currently using negative indices within existing grammar rules, + your code will break. This should be extremely rare if non-existent in + most cases. The argument to various grammar rules is not usually not + processed in the same way as a list of items. + +Version 2.0 +------------------------------ +09/07/06: beazley + Major cleanup and refactoring of the LR table generation code. Both SLR + and LALR(1) table generation is now performed by the same code base with + only minor extensions for extra LALR(1) processing. + +09/07/06: beazley + Completely reimplemented the entire LALR(1) parsing engine to use the + DeRemer and Pennello algorithm for calculating lookahead sets. This + significantly improves the performance of generating LALR(1) tables + and has the added feature of actually working correctly! If you + experienced weird behavior with LALR(1) in prior releases, this should + hopefully resolve all of those problems. Many thanks to + Andrew Waters and Markus Schoepflin for submitting bug reports + and helping me test out the revised LALR(1) support. + +Version 1.8 +------------------------------ +08/02/06: beazley + Fixed a problem related to the handling of default actions in LALR(1) + parsing. If you experienced subtle and/or bizarre behavior when trying + to use the LALR(1) engine, this may correct those problems. Patch + contributed by Russ Cox. Note: This patch has been superceded by + revisions for LALR(1) parsing in Ply-2.0. + +08/02/06: beazley + Added support for slicing of productions in yacc. + Patch contributed by Patrick Mezard. + +Version 1.7 +------------------------------ +03/02/06: beazley + Fixed infinite recursion problem ReduceToTerminals() function that + would sometimes come up in LALR(1) table generation. Reported by + Markus Schoepflin. + +03/01/06: beazley + Added "reflags" argument to lex(). For example: + + lex.lex(reflags=re.UNICODE) + + This can be used to specify optional flags to the re.compile() function + used inside the lexer. This may be necessary for special situations such + as processing Unicode (e.g., if you want escapes like \w and \b to consult + the Unicode character property database). The need for this suggested by + Andreas Jung. + +03/01/06: beazley + Fixed a bug with an uninitialized variable on repeated instantiations of parser + objects when the write_tables=0 argument was used. Reported by Michael Brown. + +03/01/06: beazley + Modified lex.py to accept Unicode strings both as the regular expressions for + tokens and as input. Hopefully this is the only change needed for Unicode support. + Patch contributed by Johan Dahl. + +03/01/06: beazley + Modified the class-based interface to work with new-style or old-style classes. + Patch contributed by Michael Brown (although I tweaked it slightly so it would work + with older versions of Python). + +Version 1.6 +------------------------------ +05/27/05: beazley + Incorporated patch contributed by Christopher Stawarz to fix an extremely + devious bug in LALR(1) parser generation. This patch should fix problems + numerous people reported with LALR parsing. + +05/27/05: beazley + Fixed problem with lex.py copy constructor. Reported by Dave Aitel, Aaron Lav, + and Thad Austin. + +05/27/05: beazley + Added outputdir option to yacc() to control output directory. Contributed + by Christopher Stawarz. + +05/27/05: beazley + Added rununit.py test script to run tests using the Python unittest module. + Contributed by Miki Tebeka. + +Version 1.5 +------------------------------ +05/26/04: beazley + Major enhancement. LALR(1) parsing support is now working. + This feature was implemented by Elias Ioup (ezioup@alumni.uchicago.edu) + and optimized by David Beazley. To use LALR(1) parsing do + the following: + + yacc.yacc(method="LALR") + + Computing LALR(1) parsing tables takes about twice as long as + the default SLR method. However, LALR(1) allows you to handle + more complex grammars. For example, the ANSI C grammar + (in example/ansic) has 13 shift-reduce conflicts with SLR, but + only has 1 shift-reduce conflict with LALR(1). + +05/20/04: beazley + Added a __len__ method to parser production lists. Can + be used in parser rules like this: + + def p_somerule(p): + """a : B C D + | E F" + if (len(p) == 3): + # Must have been first rule + elif (len(p) == 2): + # Must be second rule + + Suggested by Joshua Gerth and others. + +Version 1.4 +------------------------------ +04/23/04: beazley + Incorporated a variety of patches contributed by Eric Raymond. + These include: + + 0. Cleans up some comments so they don't wrap on an 80-column display. + 1. Directs compiler errors to stderr where they belong. + 2. Implements and documents automatic line counting when \n is ignored. + 3. Changes the way progress messages are dumped when debugging is on. + The new format is both less verbose and conveys more information than + the old, including shift and reduce actions. + +04/23/04: beazley + Added a Python setup.py file to simply installation. Contributed + by Adam Kerrison. + +04/23/04: beazley + Added patches contributed by Adam Kerrison. + + - Some output is now only shown when debugging is enabled. This + means that PLY will be completely silent when not in debugging mode. + + - An optional parameter "write_tables" can be passed to yacc() to + control whether or not parsing tables are written. By default, + it is true, but it can be turned off if you don't want the yacc + table file. Note: disabling this will cause yacc() to regenerate + the parsing table each time. + +04/23/04: beazley + Added patches contributed by David McNab. This patch addes two + features: + + - The parser can be supplied as a class instead of a module. + For an example of this, see the example/classcalc directory. + + - Debugging output can be directed to a filename of the user's + choice. Use + + yacc(debugfile="somefile.out") + + +Version 1.3 +------------------------------ +12/10/02: jmdyck + Various minor adjustments to the code that Dave checked in today. + Updated test/yacc_{inf,unused}.exp to reflect today's changes. + +12/10/02: beazley + Incorporated a variety of minor bug fixes to empty production + handling and infinite recursion checking. Contributed by + Michael Dyck. + +12/10/02: beazley + Removed bogus recover() method call in yacc.restart() + +Version 1.2 +------------------------------ +11/27/02: beazley + Lexer and parser objects are now available as an attribute + of tokens and slices respectively. For example: + + def t_NUMBER(t): + r'\d+' + print t.lexer + + def p_expr_plus(t): + 'expr: expr PLUS expr' + print t.lexer + print t.parser + + This can be used for state management (if needed). + +10/31/02: beazley + Modified yacc.py to work with Python optimize mode. To make + this work, you need to use + + yacc.yacc(optimize=1) + + Furthermore, you need to first run Python in normal mode + to generate the necessary parsetab.py files. After that, + you can use python -O or python -OO. + + Note: optimized mode turns off a lot of error checking. + Only use when you are sure that your grammar is working. + Make sure parsetab.py is up to date! + +10/30/02: beazley + Added cloning of Lexer objects. For example: + + import copy + l = lex.lex() + lc = copy.copy(l) + + l.input("Some text") + lc.input("Some other text") + ... + + This might be useful if the same "lexer" is meant to + be used in different contexts---or if multiple lexers + are running concurrently. + +10/30/02: beazley + Fixed subtle bug with first set computation and empty productions. + Patch submitted by Michael Dyck. + +10/30/02: beazley + Fixed error messages to use "filename:line: message" instead + of "filename:line. message". This makes error reporting more + friendly to emacs. Patch submitted by François Pinard. + +10/30/02: beazley + Improvements to parser.out file. Terminals and nonterminals + are sorted instead of being printed in random order. + Patch submitted by François Pinard. + +10/30/02: beazley + Improvements to parser.out file output. Rules are now printed + in a way that's easier to understand. Contributed by Russ Cox. + +10/30/02: beazley + Added 'nonassoc' associativity support. This can be used + to disable the chaining of operators like a < b < c. + To use, simply specify 'nonassoc' in the precedence table + + precedence = ( + ('nonassoc', 'LESSTHAN', 'GREATERTHAN'), # Nonassociative operators + ('left', 'PLUS', 'MINUS'), + ('left', 'TIMES', 'DIVIDE'), + ('right', 'UMINUS'), # Unary minus operator + ) + + Patch contributed by Russ Cox. + +10/30/02: beazley + Modified the lexer to provide optional support for Python -O and -OO + modes. To make this work, Python *first* needs to be run in + unoptimized mode. This reads the lexing information and creates a + file "lextab.py". Then, run lex like this: + + # module foo.py + ... + ... + lex.lex(optimize=1) + + Once the lextab file has been created, subsequent calls to + lex.lex() will read data from the lextab file instead of using + introspection. In optimized mode (-O, -OO) everything should + work normally despite the loss of doc strings. + + To change the name of the file 'lextab.py' use the following: + + lex.lex(lextab="footab") + + (this creates a file footab.py) + + +Version 1.1 October 25, 2001 +------------------------------ + +10/25/01: beazley + Modified the table generator to produce much more compact data. + This should greatly reduce the size of the parsetab.py[c] file. + Caveat: the tables still need to be constructed so a little more + work is done in parsetab on import. + +10/25/01: beazley + There may be a possible bug in the cycle detector that reports errors + about infinite recursion. I'm having a little trouble tracking it + down, but if you get this problem, you can disable the cycle + detector as follows: + + yacc.yacc(check_recursion = 0) + +10/25/01: beazley + Fixed a bug in lex.py that sometimes caused illegal characters to be + reported incorrectly. Reported by Sverre Jørgensen. + +7/8/01 : beazley + Added a reference to the underlying lexer object when tokens are handled by + functions. The lexer is available as the 'lexer' attribute. This + was added to provide better lexing support for languages such as Fortran + where certain types of tokens can't be conveniently expressed as regular + expressions (and where the tokenizing function may want to perform a + little backtracking). Suggested by Pearu Peterson. + +6/20/01 : beazley + Modified yacc() function so that an optional starting symbol can be specified. + For example: + + yacc.yacc(start="statement") + + Normally yacc always treats the first production rule as the starting symbol. + However, if you are debugging your grammar it may be useful to specify + an alternative starting symbol. Idea suggested by Rich Salz. + +Version 1.0 June 18, 2001 +-------------------------- +Initial public offering + diff --git a/third_party/python/ply/MANIFEST.in b/third_party/python/ply/MANIFEST.in new file mode 100644 index 0000000000..0d37431b0b --- /dev/null +++ b/third_party/python/ply/MANIFEST.in @@ -0,0 +1,8 @@ +recursive-include example * +recursive-include doc * +recursive-include test * +include ANNOUNCE +include README.md +include CHANGES +include TODO +global-exclude *.pyc diff --git a/third_party/python/ply/PKG-INFO b/third_party/python/ply/PKG-INFO new file mode 100644 index 0000000000..6eedf42595 --- /dev/null +++ b/third_party/python/ply/PKG-INFO @@ -0,0 +1,22 @@ +Metadata-Version: 1.1 +Name: ply +Version: 3.10 +Summary: Python Lex & Yacc +Home-page: http://www.dabeaz.com/ply/ +Author: David Beazley +Author-email: dave@dabeaz.com +License: BSD +Description: + PLY is yet another implementation of lex and yacc for Python. Some notable + features include the fact that its implemented entirely in Python and it + uses LALR(1) parsing which is efficient and well suited for larger grammars. + + PLY provides most of the standard lex/yacc features including support for empty + productions, precedence rules, error recovery, and support for ambiguous grammars. + + PLY is extremely easy to use and provides very extensive error checking. + It is compatible with both Python 2 and Python 3. + +Platform: UNKNOWN +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 2 diff --git a/third_party/python/ply/README.md b/third_party/python/ply/README.md new file mode 100644 index 0000000000..e428f1b14a --- /dev/null +++ b/third_party/python/ply/README.md @@ -0,0 +1,273 @@ +PLY (Python Lex-Yacc) Version 3.10 + +Copyright (C) 2001-2017 +David M. Beazley (Dabeaz LLC) +All rights reserved. + +Redistribution and use in source and binary forms, with or without +modification, are permitted provided that the following conditions are +met: + +* Redistributions of source code must retain the above copyright notice, + this list of conditions and the following disclaimer. +* Redistributions in binary form must reproduce the above copyright notice, + this list of conditions and the following disclaimer in the documentation + and/or other materials provided with the distribution. +* Neither the name of the David Beazley or Dabeaz LLC may be used to + endorse or promote products derived from this software without + specific prior written permission. + +THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +Introduction +============ + +PLY is a 100% Python implementation of the common parsing tools lex +and yacc. Here are a few highlights: + + - PLY is very closely modeled after traditional lex/yacc. + If you know how to use these tools in C, you will find PLY + to be similar. + + - PLY provides *very* extensive error reporting and diagnostic + information to assist in parser construction. The original + implementation was developed for instructional purposes. As + a result, the system tries to identify the most common types + of errors made by novice users. + + - PLY provides full support for empty productions, error recovery, + precedence specifiers, and moderately ambiguous grammars. + + - Parsing is based on LR-parsing which is fast, memory efficient, + better suited to large grammars, and which has a number of nice + properties when dealing with syntax errors and other parsing problems. + Currently, PLY builds its parsing tables using the LALR(1) + algorithm used in yacc. + + - PLY uses Python introspection features to build lexers and parsers. + This greatly simplifies the task of parser construction since it reduces + the number of files and eliminates the need to run a separate lex/yacc + tool before running your program. + + - PLY can be used to build parsers for "real" programming languages. + Although it is not ultra-fast due to its Python implementation, + PLY can be used to parse grammars consisting of several hundred + rules (as might be found for a language like C). The lexer and LR + parser are also reasonably efficient when parsing typically + sized programs. People have used PLY to build parsers for + C, C++, ADA, and other real programming languages. + +How to Use +========== + +PLY consists of two files : lex.py and yacc.py. These are contained +within the 'ply' directory which may also be used as a Python package. +To use PLY, simply copy the 'ply' directory to your project and import +lex and yacc from the associated 'ply' package. For example: + + import ply.lex as lex + import ply.yacc as yacc + +Alternatively, you can copy just the files lex.py and yacc.py +individually and use them as modules. For example: + + import lex + import yacc + +The file setup.py can be used to install ply using distutils. + +The file doc/ply.html contains complete documentation on how to use +the system. + +The example directory contains several different examples including a +PLY specification for ANSI C as given in K&R 2nd Ed. + +A simple example is found at the end of this document + +Requirements +============ +PLY requires the use of Python 2.6 or greater. However, you should +use the latest Python release if possible. It should work on just +about any platform. PLY has been tested with both CPython and Jython. +It also seems to work with IronPython. + +Resources +========= +More information about PLY can be obtained on the PLY webpage at: + + http://www.dabeaz.com/ply + +For a detailed overview of parsing theory, consult the excellent +book "Compilers : Principles, Techniques, and Tools" by Aho, Sethi, and +Ullman. The topics found in "Lex & Yacc" by Levine, Mason, and Brown +may also be useful. + +The GitHub page for PLY can be found at: + + https://github.com/dabeaz/ply + +An old and relatively inactive discussion group for PLY is found at: + + http://groups.google.com/group/ply-hack + +Acknowledgments +=============== +A special thanks is in order for all of the students in CS326 who +suffered through about 25 different versions of these tools :-). + +The CHANGES file acknowledges those who have contributed patches. + +Elias Ioup did the first implementation of LALR(1) parsing in PLY-1.x. +Andrew Waters and Markus Schoepflin were instrumental in reporting bugs +and testing a revised LALR(1) implementation for PLY-2.0. + +Special Note for PLY-3.0 +======================== +PLY-3.0 the first PLY release to support Python 3. However, backwards +compatibility with Python 2.6 is still preserved. PLY provides dual +Python 2/3 compatibility by restricting its implementation to a common +subset of basic language features. You should not convert PLY using +2to3--it is not necessary and may in fact break the implementation. + +Example +======= + +Here is a simple example showing a PLY implementation of a calculator +with variables. + + # ----------------------------------------------------------------------------- + # calc.py + # + # A simple calculator with variables. + # ----------------------------------------------------------------------------- + + tokens = ( + 'NAME','NUMBER', + 'PLUS','MINUS','TIMES','DIVIDE','EQUALS', + 'LPAREN','RPAREN', + ) + + # Tokens + + t_PLUS = r'\+' + t_MINUS = r'-' + t_TIMES = r'\*' + t_DIVIDE = r'/' + t_EQUALS = r'=' + t_LPAREN = r'\(' + t_RPAREN = r'\)' + t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + def t_NUMBER(t): + r'\d+' + t.value = int(t.value) + return t + + # Ignored characters + t_ignore = " \t" + + def t_newline(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + + # Build the lexer + import ply.lex as lex + lex.lex() + + # Precedence rules for the arithmetic operators + precedence = ( + ('left','PLUS','MINUS'), + ('left','TIMES','DIVIDE'), + ('right','UMINUS'), + ) + + # dictionary of names (for storing variables) + names = { } + + def p_statement_assign(p): + 'statement : NAME EQUALS expression' + names[p[1]] = p[3] + + def p_statement_expr(p): + 'statement : expression' + print(p[1]) + + def p_expression_binop(p): + '''expression : expression PLUS expression + | expression MINUS expression + | expression TIMES expression + | expression DIVIDE expression''' + if p[2] == '+' : p[0] = p[1] + p[3] + elif p[2] == '-': p[0] = p[1] - p[3] + elif p[2] == '*': p[0] = p[1] * p[3] + elif p[2] == '/': p[0] = p[1] / p[3] + + def p_expression_uminus(p): + 'expression : MINUS expression %prec UMINUS' + p[0] = -p[2] + + def p_expression_group(p): + 'expression : LPAREN expression RPAREN' + p[0] = p[2] + + def p_expression_number(p): + 'expression : NUMBER' + p[0] = p[1] + + def p_expression_name(p): + 'expression : NAME' + try: + p[0] = names[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + def p_error(p): + print("Syntax error at '%s'" % p.value) + + import ply.yacc as yacc + yacc.yacc() + + while True: + try: + s = raw_input('calc > ') # use input() on Python 3 + except EOFError: + break + yacc.parse(s) + + +Bug Reports and Patches +======================= +My goal with PLY is to simply have a decent lex/yacc implementation +for Python. As a general rule, I don't spend huge amounts of time +working on it unless I receive very specific bug reports and/or +patches to fix problems. I also try to incorporate submitted feature +requests and enhancements into each new version. Please visit the PLY +github page at https://github.com/dabeaz/ply to submit issues and pull +requests. To contact me about bugs and/or new features, please send +email to dave@dabeaz.com. + +-- Dave + + + + + + + + + diff --git a/third_party/python/ply/TODO b/third_party/python/ply/TODO new file mode 100644 index 0000000000..f4800aacf4 --- /dev/null +++ b/third_party/python/ply/TODO @@ -0,0 +1,16 @@ +The PLY to-do list: + +1. Finish writing the C Preprocessor module. Started in the + file ply/cpp.py + +2. Create and document libraries of useful tokens. + +3. Expand the examples/yply tool that parses bison/yacc + files. + +4. Think of various diabolical things to do with the + new yacc internals. For example, it is now possible + to specify grammrs using completely different schemes + than the reflection approach used by PLY. + + diff --git a/third_party/python/ply/example/BASIC/README b/third_party/python/ply/example/BASIC/README new file mode 100644 index 0000000000..be24a3005e --- /dev/null +++ b/third_party/python/ply/example/BASIC/README @@ -0,0 +1,79 @@ +Inspired by a September 14, 2006 Salon article "Why Johnny Can't Code" by +David Brin (http://www.salon.com/tech/feature/2006/09/14/basic/index.html), +I thought that a fully working BASIC interpreter might be an interesting, +if not questionable, PLY example. Uh, okay, so maybe it's just a bad idea, +but in any case, here it is. + +In this example, you'll find a rough implementation of 1964 Dartmouth BASIC +as described in the manual at: + + http://www.bitsavers.org/pdf/dartmouth/BASIC_Oct64.pdf + +See also: + + http://en.wikipedia.org/wiki/Dartmouth_BASIC + +This dialect is downright primitive---there are no string variables +and no facilities for interactive input. Moreover, subroutines and functions +are brain-dead even more than they usually are for BASIC. Of course, +the GOTO statement is provided. + +Nevertheless, there are a few interesting aspects of this example: + + - It illustrates a fully working interpreter including lexing, parsing, + and interpretation of instructions. + + - The parser shows how to catch and report various kinds of parsing + errors in a more graceful way. + + - The example both parses files (supplied on command line) and + interactive input entered line by line. + + - It shows how you might represent parsed information. In this case, + each BASIC statement is encoded into a Python tuple containing the + statement type and parameters. These tuples are then stored in + a dictionary indexed by program line numbers. + + - Even though it's just BASIC, the parser contains more than 80 + rules and 150 parsing states. Thus, it's a little more meaty than + the calculator example. + +To use the example, run it as follows: + + % python basic.py hello.bas + HELLO WORLD + % + +or use it interactively: + + % python basic.py + [BASIC] 10 PRINT "HELLO WORLD" + [BASIC] 20 END + [BASIC] RUN + HELLO WORLD + [BASIC] + +The following files are defined: + + basic.py - High level script that controls everything + basiclex.py - BASIC tokenizer + basparse.py - BASIC parser + basinterp.py - BASIC interpreter that runs parsed programs. + +In addition, a number of sample BASIC programs (.bas suffix) are +provided. These were taken out of the Dartmouth manual. + +Disclaimer: I haven't spent a ton of time testing this and it's likely that +I've skimped here and there on a few finer details (e.g., strictly enforcing +variable naming rules). However, the interpreter seems to be able to run +the examples in the BASIC manual. + +Have fun! + +-Dave + + + + + + diff --git a/third_party/python/ply/example/BASIC/basic.py b/third_party/python/ply/example/BASIC/basic.py new file mode 100644 index 0000000000..70ac9e7c74 --- /dev/null +++ b/third_party/python/ply/example/BASIC/basic.py @@ -0,0 +1,65 @@ +# An implementation of Dartmouth BASIC (1964) +# + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +import basiclex +import basparse +import basinterp + +# If a filename has been specified, we try to run it. +# If a runtime error occurs, we bail out and enter +# interactive mode below +if len(sys.argv) == 2: + data = open(sys.argv[1]).read() + prog = basparse.parse(data) + if not prog: + raise SystemExit + b = basinterp.BasicInterpreter(prog) + try: + b.run() + raise SystemExit + except RuntimeError: + pass + +else: + b = basinterp.BasicInterpreter({}) + +# Interactive mode. This incrementally adds/deletes statements +# from the program stored in the BasicInterpreter object. In +# addition, special commands 'NEW','LIST',and 'RUN' are added. +# Specifying a line number with no code deletes that line from +# the program. + +while 1: + try: + line = raw_input("[BASIC] ") + except EOFError: + raise SystemExit + if not line: + continue + line += "\n" + prog = basparse.parse(line) + if not prog: + continue + + keys = list(prog) + if keys[0] > 0: + b.add_statements(prog) + else: + stat = prog[keys[0]] + if stat[0] == 'RUN': + try: + b.run() + except RuntimeError: + pass + elif stat[0] == 'LIST': + b.list() + elif stat[0] == 'BLANK': + b.del_line(stat[1]) + elif stat[0] == 'NEW': + b.new() diff --git a/third_party/python/ply/example/BASIC/basiclex.py b/third_party/python/ply/example/BASIC/basiclex.py new file mode 100644 index 0000000000..4151f4c34f --- /dev/null +++ b/third_party/python/ply/example/BASIC/basiclex.py @@ -0,0 +1,61 @@ +# An implementation of Dartmouth BASIC (1964) + +from ply import * + +keywords = ( + 'LET', 'READ', 'DATA', 'PRINT', 'GOTO', 'IF', 'THEN', 'FOR', 'NEXT', 'TO', 'STEP', + 'END', 'STOP', 'DEF', 'GOSUB', 'DIM', 'REM', 'RETURN', 'RUN', 'LIST', 'NEW', +) + +tokens = keywords + ( + 'EQUALS', 'PLUS', 'MINUS', 'TIMES', 'DIVIDE', 'POWER', + 'LPAREN', 'RPAREN', 'LT', 'LE', 'GT', 'GE', 'NE', + 'COMMA', 'SEMI', 'INTEGER', 'FLOAT', 'STRING', + 'ID', 'NEWLINE' +) + +t_ignore = ' \t' + + +def t_REM(t): + r'REM .*' + return t + + +def t_ID(t): + r'[A-Z][A-Z0-9]*' + if t.value in keywords: + t.type = t.value + return t + +t_EQUALS = r'=' +t_PLUS = r'\+' +t_MINUS = r'-' +t_TIMES = r'\*' +t_POWER = r'\^' +t_DIVIDE = r'/' +t_LPAREN = r'\(' +t_RPAREN = r'\)' +t_LT = r'<' +t_LE = r'<=' +t_GT = r'>' +t_GE = r'>=' +t_NE = r'<>' +t_COMMA = r'\,' +t_SEMI = r';' +t_INTEGER = r'\d+' +t_FLOAT = r'((\d*\.\d+)(E[\+-]?\d+)?|([1-9]\d*E[\+-]?\d+))' +t_STRING = r'\".*?\"' + + +def t_NEWLINE(t): + r'\n' + t.lexer.lineno += 1 + return t + + +def t_error(t): + print("Illegal character %s" % t.value[0]) + t.lexer.skip(1) + +lex.lex(debug=0) diff --git a/third_party/python/ply/example/BASIC/basiclog.py b/third_party/python/ply/example/BASIC/basiclog.py new file mode 100644 index 0000000000..9dcc7feda6 --- /dev/null +++ b/third_party/python/ply/example/BASIC/basiclog.py @@ -0,0 +1,73 @@ +# An implementation of Dartmouth BASIC (1964) +# + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +import logging +logging.basicConfig( + level=logging.INFO, + filename="parselog.txt", + filemode="w" +) +log = logging.getLogger() + +import basiclex +import basparse +import basinterp + +# If a filename has been specified, we try to run it. +# If a runtime error occurs, we bail out and enter +# interactive mode below +if len(sys.argv) == 2: + data = open(sys.argv[1]).read() + prog = basparse.parse(data, debug=log) + if not prog: + raise SystemExit + b = basinterp.BasicInterpreter(prog) + try: + b.run() + raise SystemExit + except RuntimeError: + pass + +else: + b = basinterp.BasicInterpreter({}) + +# Interactive mode. This incrementally adds/deletes statements +# from the program stored in the BasicInterpreter object. In +# addition, special commands 'NEW','LIST',and 'RUN' are added. +# Specifying a line number with no code deletes that line from +# the program. + +while 1: + try: + line = raw_input("[BASIC] ") + except EOFError: + raise SystemExit + if not line: + continue + line += "\n" + prog = basparse.parse(line, debug=log) + if not prog: + continue + + keys = list(prog) + if keys[0] > 0: + b.add_statements(prog) + else: + stat = prog[keys[0]] + if stat[0] == 'RUN': + try: + b.run() + except RuntimeError: + pass + elif stat[0] == 'LIST': + b.list() + elif stat[0] == 'BLANK': + b.del_line(stat[1]) + elif stat[0] == 'NEW': + b.new() diff --git a/third_party/python/ply/example/BASIC/basinterp.py b/third_party/python/ply/example/BASIC/basinterp.py new file mode 100644 index 0000000000..67762c797b --- /dev/null +++ b/third_party/python/ply/example/BASIC/basinterp.py @@ -0,0 +1,496 @@ +# This file provides the runtime support for running a basic program +# Assumes the program has been parsed using basparse.py + +import sys +import math +import random + + +class BasicInterpreter: + + # Initialize the interpreter. prog is a dictionary + # containing (line,statement) mappings + def __init__(self, prog): + self.prog = prog + + self.functions = { # Built-in function table + 'SIN': lambda z: math.sin(self.eval(z)), + 'COS': lambda z: math.cos(self.eval(z)), + 'TAN': lambda z: math.tan(self.eval(z)), + 'ATN': lambda z: math.atan(self.eval(z)), + 'EXP': lambda z: math.exp(self.eval(z)), + 'ABS': lambda z: abs(self.eval(z)), + 'LOG': lambda z: math.log(self.eval(z)), + 'SQR': lambda z: math.sqrt(self.eval(z)), + 'INT': lambda z: int(self.eval(z)), + 'RND': lambda z: random.random() + } + + # Collect all data statements + def collect_data(self): + self.data = [] + for lineno in self.stat: + if self.prog[lineno][0] == 'DATA': + self.data = self.data + self.prog[lineno][1] + self.dc = 0 # Initialize the data counter + + # Check for end statements + def check_end(self): + has_end = 0 + for lineno in self.stat: + if self.prog[lineno][0] == 'END' and not has_end: + has_end = lineno + if not has_end: + print("NO END INSTRUCTION") + self.error = 1 + return + if has_end != lineno: + print("END IS NOT LAST") + self.error = 1 + + # Check loops + def check_loops(self): + for pc in range(len(self.stat)): + lineno = self.stat[pc] + if self.prog[lineno][0] == 'FOR': + forinst = self.prog[lineno] + loopvar = forinst[1] + for i in range(pc + 1, len(self.stat)): + if self.prog[self.stat[i]][0] == 'NEXT': + nextvar = self.prog[self.stat[i]][1] + if nextvar != loopvar: + continue + self.loopend[pc] = i + break + else: + print("FOR WITHOUT NEXT AT LINE %s" % self.stat[pc]) + self.error = 1 + + # Evaluate an expression + def eval(self, expr): + etype = expr[0] + if etype == 'NUM': + return expr[1] + elif etype == 'GROUP': + return self.eval(expr[1]) + elif etype == 'UNARY': + if expr[1] == '-': + return -self.eval(expr[2]) + elif etype == 'BINOP': + if expr[1] == '+': + return self.eval(expr[2]) + self.eval(expr[3]) + elif expr[1] == '-': + return self.eval(expr[2]) - self.eval(expr[3]) + elif expr[1] == '*': + return self.eval(expr[2]) * self.eval(expr[3]) + elif expr[1] == '/': + return float(self.eval(expr[2])) / self.eval(expr[3]) + elif expr[1] == '^': + return abs(self.eval(expr[2]))**self.eval(expr[3]) + elif etype == 'VAR': + var, dim1, dim2 = expr[1] + if not dim1 and not dim2: + if var in self.vars: + return self.vars[var] + else: + print("UNDEFINED VARIABLE %s AT LINE %s" % + (var, self.stat[self.pc])) + raise RuntimeError + # May be a list lookup or a function evaluation + if dim1 and not dim2: + if var in self.functions: + # A function + return self.functions[var](dim1) + else: + # A list evaluation + if var in self.lists: + dim1val = self.eval(dim1) + if dim1val < 1 or dim1val > len(self.lists[var]): + print("LIST INDEX OUT OF BOUNDS AT LINE %s" % + self.stat[self.pc]) + raise RuntimeError + return self.lists[var][dim1val - 1] + if dim1 and dim2: + if var in self.tables: + dim1val = self.eval(dim1) + dim2val = self.eval(dim2) + if dim1val < 1 or dim1val > len(self.tables[var]) or dim2val < 1 or dim2val > len(self.tables[var][0]): + print("TABLE INDEX OUT OUT BOUNDS AT LINE %s" % + self.stat[self.pc]) + raise RuntimeError + return self.tables[var][dim1val - 1][dim2val - 1] + print("UNDEFINED VARIABLE %s AT LINE %s" % + (var, self.stat[self.pc])) + raise RuntimeError + + # Evaluate a relational expression + def releval(self, expr): + etype = expr[1] + lhs = self.eval(expr[2]) + rhs = self.eval(expr[3]) + if etype == '<': + if lhs < rhs: + return 1 + else: + return 0 + + elif etype == '<=': + if lhs <= rhs: + return 1 + else: + return 0 + + elif etype == '>': + if lhs > rhs: + return 1 + else: + return 0 + + elif etype == '>=': + if lhs >= rhs: + return 1 + else: + return 0 + + elif etype == '=': + if lhs == rhs: + return 1 + else: + return 0 + + elif etype == '<>': + if lhs != rhs: + return 1 + else: + return 0 + + # Assignment + def assign(self, target, value): + var, dim1, dim2 = target + if not dim1 and not dim2: + self.vars[var] = self.eval(value) + elif dim1 and not dim2: + # List assignment + dim1val = self.eval(dim1) + if not var in self.lists: + self.lists[var] = [0] * 10 + + if dim1val > len(self.lists[var]): + print ("DIMENSION TOO LARGE AT LINE %s" % self.stat[self.pc]) + raise RuntimeError + self.lists[var][dim1val - 1] = self.eval(value) + elif dim1 and dim2: + dim1val = self.eval(dim1) + dim2val = self.eval(dim2) + if not var in self.tables: + temp = [0] * 10 + v = [] + for i in range(10): + v.append(temp[:]) + self.tables[var] = v + # Variable already exists + if dim1val > len(self.tables[var]) or dim2val > len(self.tables[var][0]): + print("DIMENSION TOO LARGE AT LINE %s" % self.stat[self.pc]) + raise RuntimeError + self.tables[var][dim1val - 1][dim2val - 1] = self.eval(value) + + # Change the current line number + def goto(self, linenum): + if not linenum in self.prog: + print("UNDEFINED LINE NUMBER %d AT LINE %d" % + (linenum, self.stat[self.pc])) + raise RuntimeError + self.pc = self.stat.index(linenum) + + # Run it + def run(self): + self.vars = {} # All variables + self.lists = {} # List variables + self.tables = {} # Tables + self.loops = [] # Currently active loops + self.loopend = {} # Mapping saying where loops end + self.gosub = None # Gosub return point (if any) + self.error = 0 # Indicates program error + + self.stat = list(self.prog) # Ordered list of all line numbers + self.stat.sort() + self.pc = 0 # Current program counter + + # Processing prior to running + + self.collect_data() # Collect all of the data statements + self.check_end() + self.check_loops() + + if self.error: + raise RuntimeError + + while 1: + line = self.stat[self.pc] + instr = self.prog[line] + + op = instr[0] + + # END and STOP statements + if op == 'END' or op == 'STOP': + break # We're done + + # GOTO statement + elif op == 'GOTO': + newline = instr[1] + self.goto(newline) + continue + + # PRINT statement + elif op == 'PRINT': + plist = instr[1] + out = "" + for label, val in plist: + if out: + out += ' ' * (15 - (len(out) % 15)) + out += label + if val: + if label: + out += " " + eval = self.eval(val) + out += str(eval) + sys.stdout.write(out) + end = instr[2] + if not (end == ',' or end == ';'): + sys.stdout.write("\n") + if end == ',': + sys.stdout.write(" " * (15 - (len(out) % 15))) + if end == ';': + sys.stdout.write(" " * (3 - (len(out) % 3))) + + # LET statement + elif op == 'LET': + target = instr[1] + value = instr[2] + self.assign(target, value) + + # READ statement + elif op == 'READ': + for target in instr[1]: + if self.dc < len(self.data): + value = ('NUM', self.data[self.dc]) + self.assign(target, value) + self.dc += 1 + else: + # No more data. Program ends + return + elif op == 'IF': + relop = instr[1] + newline = instr[2] + if (self.releval(relop)): + self.goto(newline) + continue + + elif op == 'FOR': + loopvar = instr[1] + initval = instr[2] + finval = instr[3] + stepval = instr[4] + + # Check to see if this is a new loop + if not self.loops or self.loops[-1][0] != self.pc: + # Looks like a new loop. Make the initial assignment + newvalue = initval + self.assign((loopvar, None, None), initval) + if not stepval: + stepval = ('NUM', 1) + stepval = self.eval(stepval) # Evaluate step here + self.loops.append((self.pc, stepval)) + else: + # It's a repeat of the previous loop + # Update the value of the loop variable according to the + # step + stepval = ('NUM', self.loops[-1][1]) + newvalue = ( + 'BINOP', '+', ('VAR', (loopvar, None, None)), stepval) + + if self.loops[-1][1] < 0: + relop = '>=' + else: + relop = '<=' + if not self.releval(('RELOP', relop, newvalue, finval)): + # Loop is done. Jump to the NEXT + self.pc = self.loopend[self.pc] + self.loops.pop() + else: + self.assign((loopvar, None, None), newvalue) + + elif op == 'NEXT': + if not self.loops: + print("NEXT WITHOUT FOR AT LINE %s" % line) + return + + nextvar = instr[1] + self.pc = self.loops[-1][0] + loopinst = self.prog[self.stat[self.pc]] + forvar = loopinst[1] + if nextvar != forvar: + print("NEXT DOESN'T MATCH FOR AT LINE %s" % line) + return + continue + elif op == 'GOSUB': + newline = instr[1] + if self.gosub: + print("ALREADY IN A SUBROUTINE AT LINE %s" % line) + return + self.gosub = self.stat[self.pc] + self.goto(newline) + continue + + elif op == 'RETURN': + if not self.gosub: + print("RETURN WITHOUT A GOSUB AT LINE %s" % line) + return + self.goto(self.gosub) + self.gosub = None + + elif op == 'FUNC': + fname = instr[1] + pname = instr[2] + expr = instr[3] + + def eval_func(pvalue, name=pname, self=self, expr=expr): + self.assign((pname, None, None), pvalue) + return self.eval(expr) + self.functions[fname] = eval_func + + elif op == 'DIM': + for vname, x, y in instr[1]: + if y == 0: + # Single dimension variable + self.lists[vname] = [0] * x + else: + # Double dimension variable + temp = [0] * y + v = [] + for i in range(x): + v.append(temp[:]) + self.tables[vname] = v + + self.pc += 1 + + # Utility functions for program listing + def expr_str(self, expr): + etype = expr[0] + if etype == 'NUM': + return str(expr[1]) + elif etype == 'GROUP': + return "(%s)" % self.expr_str(expr[1]) + elif etype == 'UNARY': + if expr[1] == '-': + return "-" + str(expr[2]) + elif etype == 'BINOP': + return "%s %s %s" % (self.expr_str(expr[2]), expr[1], self.expr_str(expr[3])) + elif etype == 'VAR': + return self.var_str(expr[1]) + + def relexpr_str(self, expr): + return "%s %s %s" % (self.expr_str(expr[2]), expr[1], self.expr_str(expr[3])) + + def var_str(self, var): + varname, dim1, dim2 = var + if not dim1 and not dim2: + return varname + if dim1 and not dim2: + return "%s(%s)" % (varname, self.expr_str(dim1)) + return "%s(%s,%s)" % (varname, self.expr_str(dim1), self.expr_str(dim2)) + + # Create a program listing + def list(self): + stat = list(self.prog) # Ordered list of all line numbers + stat.sort() + for line in stat: + instr = self.prog[line] + op = instr[0] + if op in ['END', 'STOP', 'RETURN']: + print("%s %s" % (line, op)) + continue + elif op == 'REM': + print("%s %s" % (line, instr[1])) + elif op == 'PRINT': + _out = "%s %s " % (line, op) + first = 1 + for p in instr[1]: + if not first: + _out += ", " + if p[0] and p[1]: + _out += '"%s"%s' % (p[0], self.expr_str(p[1])) + elif p[1]: + _out += self.expr_str(p[1]) + else: + _out += '"%s"' % (p[0],) + first = 0 + if instr[2]: + _out += instr[2] + print(_out) + elif op == 'LET': + print("%s LET %s = %s" % + (line, self.var_str(instr[1]), self.expr_str(instr[2]))) + elif op == 'READ': + _out = "%s READ " % line + first = 1 + for r in instr[1]: + if not first: + _out += "," + _out += self.var_str(r) + first = 0 + print(_out) + elif op == 'IF': + print("%s IF %s THEN %d" % + (line, self.relexpr_str(instr[1]), instr[2])) + elif op == 'GOTO' or op == 'GOSUB': + print("%s %s %s" % (line, op, instr[1])) + elif op == 'FOR': + _out = "%s FOR %s = %s TO %s" % ( + line, instr[1], self.expr_str(instr[2]), self.expr_str(instr[3])) + if instr[4]: + _out += " STEP %s" % (self.expr_str(instr[4])) + print(_out) + elif op == 'NEXT': + print("%s NEXT %s" % (line, instr[1])) + elif op == 'FUNC': + print("%s DEF %s(%s) = %s" % + (line, instr[1], instr[2], self.expr_str(instr[3]))) + elif op == 'DIM': + _out = "%s DIM " % line + first = 1 + for vname, x, y in instr[1]: + if not first: + _out += "," + first = 0 + if y == 0: + _out += "%s(%d)" % (vname, x) + else: + _out += "%s(%d,%d)" % (vname, x, y) + + print(_out) + elif op == 'DATA': + _out = "%s DATA " % line + first = 1 + for v in instr[1]: + if not first: + _out += "," + first = 0 + _out += v + print(_out) + + # Erase the current program + def new(self): + self.prog = {} + + # Insert statements + def add_statements(self, prog): + for line, stat in prog.items(): + self.prog[line] = stat + + # Delete a statement + def del_line(self, lineno): + try: + del self.prog[lineno] + except KeyError: + pass diff --git a/third_party/python/ply/example/BASIC/basparse.py b/third_party/python/ply/example/BASIC/basparse.py new file mode 100644 index 0000000000..d610c7d909 --- /dev/null +++ b/third_party/python/ply/example/BASIC/basparse.py @@ -0,0 +1,474 @@ +# An implementation of Dartmouth BASIC (1964) +# + +from ply import * +import basiclex + +tokens = basiclex.tokens + +precedence = ( + ('left', 'PLUS', 'MINUS'), + ('left', 'TIMES', 'DIVIDE'), + ('left', 'POWER'), + ('right', 'UMINUS') +) + +# A BASIC program is a series of statements. We represent the program as a +# dictionary of tuples indexed by line number. + + +def p_program(p): + '''program : program statement + | statement''' + + if len(p) == 2 and p[1]: + p[0] = {} + line, stat = p[1] + p[0][line] = stat + elif len(p) == 3: + p[0] = p[1] + if not p[0]: + p[0] = {} + if p[2]: + line, stat = p[2] + p[0][line] = stat + +# This catch-all rule is used for any catastrophic errors. In this case, +# we simply return nothing + + +def p_program_error(p): + '''program : error''' + p[0] = None + p.parser.error = 1 + +# Format of all BASIC statements. + + +def p_statement(p): + '''statement : INTEGER command NEWLINE''' + if isinstance(p[2], str): + print("%s %s %s" % (p[2], "AT LINE", p[1])) + p[0] = None + p.parser.error = 1 + else: + lineno = int(p[1]) + p[0] = (lineno, p[2]) + +# Interactive statements. + + +def p_statement_interactive(p): + '''statement : RUN NEWLINE + | LIST NEWLINE + | NEW NEWLINE''' + p[0] = (0, (p[1], 0)) + +# Blank line number + + +def p_statement_blank(p): + '''statement : INTEGER NEWLINE''' + p[0] = (0, ('BLANK', int(p[1]))) + +# Error handling for malformed statements + + +def p_statement_bad(p): + '''statement : INTEGER error NEWLINE''' + print("MALFORMED STATEMENT AT LINE %s" % p[1]) + p[0] = None + p.parser.error = 1 + +# Blank line + + +def p_statement_newline(p): + '''statement : NEWLINE''' + p[0] = None + +# LET statement + + +def p_command_let(p): + '''command : LET variable EQUALS expr''' + p[0] = ('LET', p[2], p[4]) + + +def p_command_let_bad(p): + '''command : LET variable EQUALS error''' + p[0] = "BAD EXPRESSION IN LET" + +# READ statement + + +def p_command_read(p): + '''command : READ varlist''' + p[0] = ('READ', p[2]) + + +def p_command_read_bad(p): + '''command : READ error''' + p[0] = "MALFORMED VARIABLE LIST IN READ" + +# DATA statement + + +def p_command_data(p): + '''command : DATA numlist''' + p[0] = ('DATA', p[2]) + + +def p_command_data_bad(p): + '''command : DATA error''' + p[0] = "MALFORMED NUMBER LIST IN DATA" + +# PRINT statement + + +def p_command_print(p): + '''command : PRINT plist optend''' + p[0] = ('PRINT', p[2], p[3]) + + +def p_command_print_bad(p): + '''command : PRINT error''' + p[0] = "MALFORMED PRINT STATEMENT" + +# Optional ending on PRINT. Either a comma (,) or semicolon (;) + + +def p_optend(p): + '''optend : COMMA + | SEMI + |''' + if len(p) == 2: + p[0] = p[1] + else: + p[0] = None + +# PRINT statement with no arguments + + +def p_command_print_empty(p): + '''command : PRINT''' + p[0] = ('PRINT', [], None) + +# GOTO statement + + +def p_command_goto(p): + '''command : GOTO INTEGER''' + p[0] = ('GOTO', int(p[2])) + + +def p_command_goto_bad(p): + '''command : GOTO error''' + p[0] = "INVALID LINE NUMBER IN GOTO" + +# IF-THEN statement + + +def p_command_if(p): + '''command : IF relexpr THEN INTEGER''' + p[0] = ('IF', p[2], int(p[4])) + + +def p_command_if_bad(p): + '''command : IF error THEN INTEGER''' + p[0] = "BAD RELATIONAL EXPRESSION" + + +def p_command_if_bad2(p): + '''command : IF relexpr THEN error''' + p[0] = "INVALID LINE NUMBER IN THEN" + +# FOR statement + + +def p_command_for(p): + '''command : FOR ID EQUALS expr TO expr optstep''' + p[0] = ('FOR', p[2], p[4], p[6], p[7]) + + +def p_command_for_bad_initial(p): + '''command : FOR ID EQUALS error TO expr optstep''' + p[0] = "BAD INITIAL VALUE IN FOR STATEMENT" + + +def p_command_for_bad_final(p): + '''command : FOR ID EQUALS expr TO error optstep''' + p[0] = "BAD FINAL VALUE IN FOR STATEMENT" + + +def p_command_for_bad_step(p): + '''command : FOR ID EQUALS expr TO expr STEP error''' + p[0] = "MALFORMED STEP IN FOR STATEMENT" + +# Optional STEP qualifier on FOR statement + + +def p_optstep(p): + '''optstep : STEP expr + | empty''' + if len(p) == 3: + p[0] = p[2] + else: + p[0] = None + +# NEXT statement + + +def p_command_next(p): + '''command : NEXT ID''' + + p[0] = ('NEXT', p[2]) + + +def p_command_next_bad(p): + '''command : NEXT error''' + p[0] = "MALFORMED NEXT" + +# END statement + + +def p_command_end(p): + '''command : END''' + p[0] = ('END',) + +# REM statement + + +def p_command_rem(p): + '''command : REM''' + p[0] = ('REM', p[1]) + +# STOP statement + + +def p_command_stop(p): + '''command : STOP''' + p[0] = ('STOP',) + +# DEF statement + + +def p_command_def(p): + '''command : DEF ID LPAREN ID RPAREN EQUALS expr''' + p[0] = ('FUNC', p[2], p[4], p[7]) + + +def p_command_def_bad_rhs(p): + '''command : DEF ID LPAREN ID RPAREN EQUALS error''' + p[0] = "BAD EXPRESSION IN DEF STATEMENT" + + +def p_command_def_bad_arg(p): + '''command : DEF ID LPAREN error RPAREN EQUALS expr''' + p[0] = "BAD ARGUMENT IN DEF STATEMENT" + +# GOSUB statement + + +def p_command_gosub(p): + '''command : GOSUB INTEGER''' + p[0] = ('GOSUB', int(p[2])) + + +def p_command_gosub_bad(p): + '''command : GOSUB error''' + p[0] = "INVALID LINE NUMBER IN GOSUB" + +# RETURN statement + + +def p_command_return(p): + '''command : RETURN''' + p[0] = ('RETURN',) + +# DIM statement + + +def p_command_dim(p): + '''command : DIM dimlist''' + p[0] = ('DIM', p[2]) + + +def p_command_dim_bad(p): + '''command : DIM error''' + p[0] = "MALFORMED VARIABLE LIST IN DIM" + +# List of variables supplied to DIM statement + + +def p_dimlist(p): + '''dimlist : dimlist COMMA dimitem + | dimitem''' + if len(p) == 4: + p[0] = p[1] + p[0].append(p[3]) + else: + p[0] = [p[1]] + +# DIM items + + +def p_dimitem_single(p): + '''dimitem : ID LPAREN INTEGER RPAREN''' + p[0] = (p[1], eval(p[3]), 0) + + +def p_dimitem_double(p): + '''dimitem : ID LPAREN INTEGER COMMA INTEGER RPAREN''' + p[0] = (p[1], eval(p[3]), eval(p[5])) + +# Arithmetic expressions + + +def p_expr_binary(p): + '''expr : expr PLUS expr + | expr MINUS expr + | expr TIMES expr + | expr DIVIDE expr + | expr POWER expr''' + + p[0] = ('BINOP', p[2], p[1], p[3]) + + +def p_expr_number(p): + '''expr : INTEGER + | FLOAT''' + p[0] = ('NUM', eval(p[1])) + + +def p_expr_variable(p): + '''expr : variable''' + p[0] = ('VAR', p[1]) + + +def p_expr_group(p): + '''expr : LPAREN expr RPAREN''' + p[0] = ('GROUP', p[2]) + + +def p_expr_unary(p): + '''expr : MINUS expr %prec UMINUS''' + p[0] = ('UNARY', '-', p[2]) + +# Relational expressions + + +def p_relexpr(p): + '''relexpr : expr LT expr + | expr LE expr + | expr GT expr + | expr GE expr + | expr EQUALS expr + | expr NE expr''' + p[0] = ('RELOP', p[2], p[1], p[3]) + +# Variables + + +def p_variable(p): + '''variable : ID + | ID LPAREN expr RPAREN + | ID LPAREN expr COMMA expr RPAREN''' + if len(p) == 2: + p[0] = (p[1], None, None) + elif len(p) == 5: + p[0] = (p[1], p[3], None) + else: + p[0] = (p[1], p[3], p[5]) + +# Builds a list of variable targets as a Python list + + +def p_varlist(p): + '''varlist : varlist COMMA variable + | variable''' + if len(p) > 2: + p[0] = p[1] + p[0].append(p[3]) + else: + p[0] = [p[1]] + + +# Builds a list of numbers as a Python list + +def p_numlist(p): + '''numlist : numlist COMMA number + | number''' + + if len(p) > 2: + p[0] = p[1] + p[0].append(p[3]) + else: + p[0] = [p[1]] + +# A number. May be an integer or a float + + +def p_number(p): + '''number : INTEGER + | FLOAT''' + p[0] = eval(p[1]) + +# A signed number. + + +def p_number_signed(p): + '''number : MINUS INTEGER + | MINUS FLOAT''' + p[0] = eval("-" + p[2]) + +# List of targets for a print statement +# Returns a list of tuples (label,expr) + + +def p_plist(p): + '''plist : plist COMMA pitem + | pitem''' + if len(p) > 3: + p[0] = p[1] + p[0].append(p[3]) + else: + p[0] = [p[1]] + + +def p_item_string(p): + '''pitem : STRING''' + p[0] = (p[1][1:-1], None) + + +def p_item_string_expr(p): + '''pitem : STRING expr''' + p[0] = (p[1][1:-1], p[2]) + + +def p_item_expr(p): + '''pitem : expr''' + p[0] = ("", p[1]) + +# Empty + + +def p_empty(p): + '''empty : ''' + +# Catastrophic error handler + + +def p_error(p): + if not p: + print("SYNTAX ERROR AT EOF") + +bparser = yacc.yacc() + + +def parse(data, debug=0): + bparser.error = 0 + p = bparser.parse(data, debug=debug) + if bparser.error: + return None + return p diff --git a/third_party/python/ply/example/BASIC/dim.bas b/third_party/python/ply/example/BASIC/dim.bas new file mode 100644 index 0000000000..87bd95b32e --- /dev/null +++ b/third_party/python/ply/example/BASIC/dim.bas @@ -0,0 +1,14 @@ +5 DIM A(50,15) +10 FOR I = 1 TO 50 +20 FOR J = 1 TO 15 +30 LET A(I,J) = I + J +35 REM PRINT I,J, A(I,J) +40 NEXT J +50 NEXT I +100 FOR I = 1 TO 50 +110 FOR J = 1 TO 15 +120 PRINT A(I,J), +130 NEXT J +140 PRINT +150 NEXT I +999 END diff --git a/third_party/python/ply/example/BASIC/func.bas b/third_party/python/ply/example/BASIC/func.bas new file mode 100644 index 0000000000..447ee16a92 --- /dev/null +++ b/third_party/python/ply/example/BASIC/func.bas @@ -0,0 +1,5 @@ +10 DEF FDX(X) = 2*X +20 FOR I = 0 TO 100 +30 PRINT FDX(I) +40 NEXT I +50 END diff --git a/third_party/python/ply/example/BASIC/gcd.bas b/third_party/python/ply/example/BASIC/gcd.bas new file mode 100644 index 0000000000..d0b7746089 --- /dev/null +++ b/third_party/python/ply/example/BASIC/gcd.bas @@ -0,0 +1,22 @@ +10 PRINT "A","B","C","GCD" +20 READ A,B,C +30 LET X = A +40 LET Y = B +50 GOSUB 200 +60 LET X = G +70 LET Y = C +80 GOSUB 200 +90 PRINT A, B, C, G +100 GOTO 20 +110 DATA 60, 90, 120 +120 DATA 38456, 64872, 98765 +130 DATA 32, 384, 72 +200 LET Q = INT(X/Y) +210 LET R = X - Q*Y +220 IF R = 0 THEN 300 +230 LET X = Y +240 LET Y = R +250 GOTO 200 +300 LET G = Y +310 RETURN +999 END diff --git a/third_party/python/ply/example/BASIC/gosub.bas b/third_party/python/ply/example/BASIC/gosub.bas new file mode 100644 index 0000000000..99737b16f1 --- /dev/null +++ b/third_party/python/ply/example/BASIC/gosub.bas @@ -0,0 +1,13 @@ +100 LET X = 3 +110 GOSUB 400 +120 PRINT U, V, W +200 LET X = 5 +210 GOSUB 400 +220 LET Z = U + 2*V + 3*W +230 PRINT Z +240 GOTO 999 +400 LET U = X*X +410 LET V = X*X*X +420 LET W = X*X*X*X + X*X*X + X*X + X +430 RETURN +999 END diff --git a/third_party/python/ply/example/BASIC/hello.bas b/third_party/python/ply/example/BASIC/hello.bas new file mode 100644 index 0000000000..cc6f0b0b51 --- /dev/null +++ b/third_party/python/ply/example/BASIC/hello.bas @@ -0,0 +1,4 @@ +5 REM HELLO WORLD PROGAM +10 PRINT "HELLO WORLD" +99 END + diff --git a/third_party/python/ply/example/BASIC/linear.bas b/third_party/python/ply/example/BASIC/linear.bas new file mode 100644 index 0000000000..56c08220b3 --- /dev/null +++ b/third_party/python/ply/example/BASIC/linear.bas @@ -0,0 +1,17 @@ +1 REM ::: SOLVE A SYSTEM OF LINEAR EQUATIONS +2 REM ::: A1*X1 + A2*X2 = B1 +3 REM ::: A3*X1 + A4*X2 = B2 +4 REM -------------------------------------- +10 READ A1, A2, A3, A4 +15 LET D = A1 * A4 - A3 * A2 +20 IF D = 0 THEN 65 +30 READ B1, B2 +37 LET X1 = (B1*A4 - B2*A2) / D +42 LET X2 = (A1*B2 - A3*B1) / D +55 PRINT X1, X2 +60 GOTO 30 +65 PRINT "NO UNIQUE SOLUTION" +70 DATA 1, 2, 4 +80 DATA 2, -7, 5 +85 DATA 1, 3, 4, -7 +90 END diff --git a/third_party/python/ply/example/BASIC/maxsin.bas b/third_party/python/ply/example/BASIC/maxsin.bas new file mode 100644 index 0000000000..b96901530c --- /dev/null +++ b/third_party/python/ply/example/BASIC/maxsin.bas @@ -0,0 +1,12 @@ +5 PRINT "X VALUE", "SINE", "RESOLUTION" +10 READ D +20 LET M = -1 +30 FOR X = 0 TO 3 STEP D +40 IF SIN(X) <= M THEN 80 +50 LET X0 = X +60 LET M = SIN(X) +80 NEXT X +85 PRINT X0, M, D +90 GOTO 10 +100 DATA .1, .01, .001 +110 END diff --git a/third_party/python/ply/example/BASIC/powers.bas b/third_party/python/ply/example/BASIC/powers.bas new file mode 100644 index 0000000000..a454dc3e21 --- /dev/null +++ b/third_party/python/ply/example/BASIC/powers.bas @@ -0,0 +1,13 @@ +5 PRINT "THIS PROGRAM COMPUTES AND PRINTS THE NTH POWERS" +6 PRINT "OF THE NUMBERS LESS THAN OR EQUAL TO N FOR VARIOUS" +7 PRINT "N FROM 1 THROUGH 7" +8 PRINT +10 FOR N = 1 TO 7 +15 PRINT "N = "N +20 FOR I = 1 TO N +30 PRINT I^N, +40 NEXT I +50 PRINT +60 PRINT +70 NEXT N +80 END diff --git a/third_party/python/ply/example/BASIC/rand.bas b/third_party/python/ply/example/BASIC/rand.bas new file mode 100644 index 0000000000..4ff7a14670 --- /dev/null +++ b/third_party/python/ply/example/BASIC/rand.bas @@ -0,0 +1,4 @@ +10 FOR I = 1 TO 20 +20 PRINT INT(10*RND(0)) +30 NEXT I +40 END diff --git a/third_party/python/ply/example/BASIC/sales.bas b/third_party/python/ply/example/BASIC/sales.bas new file mode 100644 index 0000000000..a39aefb762 --- /dev/null +++ b/third_party/python/ply/example/BASIC/sales.bas @@ -0,0 +1,20 @@ +10 FOR I = 1 TO 3 +20 READ P(I) +30 NEXT I +40 FOR I = 1 TO 3 +50 FOR J = 1 TO 5 +60 READ S(I,J) +70 NEXT J +80 NEXT I +90 FOR J = 1 TO 5 +100 LET S = 0 +110 FOR I = 1 TO 3 +120 LET S = S + P(I) * S(I,J) +130 NEXT I +140 PRINT "TOTAL SALES FOR SALESMAN"J, "$"S +150 NEXT J +200 DATA 1.25, 4.30, 2.50 +210 DATA 40, 20, 37, 29, 42 +220 DATA 10, 16, 3, 21, 8 +230 DATA 35, 47, 29, 16, 33 +300 END diff --git a/third_party/python/ply/example/BASIC/sears.bas b/third_party/python/ply/example/BASIC/sears.bas new file mode 100644 index 0000000000..5ced3974e2 --- /dev/null +++ b/third_party/python/ply/example/BASIC/sears.bas @@ -0,0 +1,18 @@ +1 REM :: THIS PROGRAM COMPUTES HOW MANY TIMES YOU HAVE TO FOLD +2 REM :: A PIECE OF PAPER SO THAT IT IS TALLER THAN THE +3 REM :: SEARS TOWER. +4 REM :: S = HEIGHT OF TOWER (METERS) +5 REM :: T = THICKNESS OF PAPER (MILLIMETERS) +10 LET S = 442 +20 LET T = 0.1 +30 REM CONVERT T TO METERS +40 LET T = T * .001 +50 LET F = 1 +60 LET H = T +100 IF H > S THEN 200 +120 LET H = 2 * H +125 LET F = F + 1 +130 GOTO 100 +200 PRINT "NUMBER OF FOLDS ="F +220 PRINT "FINAL HEIGHT ="H +999 END diff --git a/third_party/python/ply/example/BASIC/sqrt1.bas b/third_party/python/ply/example/BASIC/sqrt1.bas new file mode 100644 index 0000000000..6673a91524 --- /dev/null +++ b/third_party/python/ply/example/BASIC/sqrt1.bas @@ -0,0 +1,5 @@ +10 LET X = 0 +20 LET X = X + 1 +30 PRINT X, SQR(X) +40 IF X < 100 THEN 20 +50 END diff --git a/third_party/python/ply/example/BASIC/sqrt2.bas b/third_party/python/ply/example/BASIC/sqrt2.bas new file mode 100644 index 0000000000..862d85ef26 --- /dev/null +++ b/third_party/python/ply/example/BASIC/sqrt2.bas @@ -0,0 +1,4 @@ +10 FOR X = 1 TO 100 +20 PRINT X, SQR(X) +30 NEXT X +40 END diff --git a/third_party/python/ply/example/GardenSnake/GardenSnake.py b/third_party/python/ply/example/GardenSnake/GardenSnake.py new file mode 100644 index 0000000000..8b493b40dc --- /dev/null +++ b/third_party/python/ply/example/GardenSnake/GardenSnake.py @@ -0,0 +1,777 @@ +# GardenSnake - a parser generator demonstration program +# +# This implements a modified version of a subset of Python: +# - only 'def', 'return' and 'if' statements +# - 'if' only has 'then' clause (no elif nor else) +# - single-quoted strings only, content in raw format +# - numbers are decimal.Decimal instances (not integers or floats) +# - no print statment; use the built-in 'print' function +# - only < > == + - / * implemented (and unary + -) +# - assignment and tuple assignment work +# - no generators of any sort +# - no ... well, no quite a lot + +# Why? I'm thinking about a new indentation-based configuration +# language for a project and wanted to figure out how to do it. Once +# I got that working I needed a way to test it out. My original AST +# was dumb so I decided to target Python's AST and compile it into +# Python code. Plus, it's pretty cool that it only took a day or so +# from sitting down with Ply to having working code. + +# This uses David Beazley's Ply from http://www.dabeaz.com/ply/ + +# This work is hereby released into the Public Domain. To view a copy of +# the public domain dedication, visit +# http://creativecommons.org/licenses/publicdomain/ or send a letter to +# Creative Commons, 543 Howard Street, 5th Floor, San Francisco, +# California, 94105, USA. +# +# Portions of this work are derived from Python's Grammar definition +# and may be covered under the Python copyright and license +# +# Andrew Dalke / Dalke Scientific Software, LLC +# 30 August 2006 / Cape Town, South Africa + +# Changelog: +# 30 August - added link to CC license; removed the "swapcase" encoding + +# Modifications for inclusion in PLY distribution +import sys +sys.path.insert(0, "../..") +from ply import * + +##### Lexer ###### +#import lex +import decimal + +tokens = ( + 'DEF', + 'IF', + 'NAME', + 'NUMBER', # Python decimals + 'STRING', # single quoted strings only; syntax of raw strings + 'LPAR', + 'RPAR', + 'COLON', + 'EQ', + 'ASSIGN', + 'LT', + 'GT', + 'PLUS', + 'MINUS', + 'MULT', + 'DIV', + 'RETURN', + 'WS', + 'NEWLINE', + 'COMMA', + 'SEMICOLON', + 'INDENT', + 'DEDENT', + 'ENDMARKER', +) + +#t_NUMBER = r'\d+' +# taken from decmial.py but without the leading sign + + +def t_NUMBER(t): + r"""(\d+(\.\d*)?|\.\d+)([eE][-+]? \d+)?""" + t.value = decimal.Decimal(t.value) + return t + + +def t_STRING(t): + r"'([^\\']+|\\'|\\\\)*'" # I think this is right ... + t.value = t.value[1:-1].decode("string-escape") # .swapcase() # for fun + return t + +t_COLON = r':' +t_EQ = r'==' +t_ASSIGN = r'=' +t_LT = r'<' +t_GT = r'>' +t_PLUS = r'\+' +t_MINUS = r'-' +t_MULT = r'\*' +t_DIV = r'/' +t_COMMA = r',' +t_SEMICOLON = r';' + +# Ply nicely documented how to do this. + +RESERVED = { + "def": "DEF", + "if": "IF", + "return": "RETURN", +} + + +def t_NAME(t): + r'[a-zA-Z_][a-zA-Z0-9_]*' + t.type = RESERVED.get(t.value, "NAME") + return t + +# Putting this before t_WS let it consume lines with only comments in +# them so the latter code never sees the WS part. Not consuming the +# newline. Needed for "if 1: #comment" + + +def t_comment(t): + r"[ ]*\043[^\n]*" # \043 is '#' + pass + + +# Whitespace +def t_WS(t): + r' [ ]+ ' + if t.lexer.at_line_start and t.lexer.paren_count == 0: + return t + +# Don't generate newline tokens when inside of parenthesis, eg +# a = (1, +# 2, 3) + + +def t_newline(t): + r'\n+' + t.lexer.lineno += len(t.value) + t.type = "NEWLINE" + if t.lexer.paren_count == 0: + return t + + +def t_LPAR(t): + r'\(' + t.lexer.paren_count += 1 + return t + + +def t_RPAR(t): + r'\)' + # check for underflow? should be the job of the parser + t.lexer.paren_count -= 1 + return t + + +def t_error(t): + raise SyntaxError("Unknown symbol %r" % (t.value[0],)) + print "Skipping", repr(t.value[0]) + t.lexer.skip(1) + +# I implemented INDENT / DEDENT generation as a post-processing filter + +# The original lex token stream contains WS and NEWLINE characters. +# WS will only occur before any other tokens on a line. + +# I have three filters. One tags tokens by adding two attributes. +# "must_indent" is True if the token must be indented from the +# previous code. The other is "at_line_start" which is True for WS +# and the first non-WS/non-NEWLINE on a line. It flags the check so +# see if the new line has changed indication level. + +# Python's syntax has three INDENT states +# 0) no colon hence no need to indent +# 1) "if 1: go()" - simple statements have a COLON but no need for an indent +# 2) "if 1:\n go()" - complex statements have a COLON NEWLINE and must indent +NO_INDENT = 0 +MAY_INDENT = 1 +MUST_INDENT = 2 + +# only care about whitespace at the start of a line + + +def track_tokens_filter(lexer, tokens): + lexer.at_line_start = at_line_start = True + indent = NO_INDENT + saw_colon = False + for token in tokens: + token.at_line_start = at_line_start + + if token.type == "COLON": + at_line_start = False + indent = MAY_INDENT + token.must_indent = False + + elif token.type == "NEWLINE": + at_line_start = True + if indent == MAY_INDENT: + indent = MUST_INDENT + token.must_indent = False + + elif token.type == "WS": + assert token.at_line_start == True + at_line_start = True + token.must_indent = False + + else: + # A real token; only indent after COLON NEWLINE + if indent == MUST_INDENT: + token.must_indent = True + else: + token.must_indent = False + at_line_start = False + indent = NO_INDENT + + yield token + lexer.at_line_start = at_line_start + + +def _new_token(type, lineno): + tok = lex.LexToken() + tok.type = type + tok.value = None + tok.lineno = lineno + return tok + +# Synthesize a DEDENT tag + + +def DEDENT(lineno): + return _new_token("DEDENT", lineno) + +# Synthesize an INDENT tag + + +def INDENT(lineno): + return _new_token("INDENT", lineno) + + +# Track the indentation level and emit the right INDENT / DEDENT events. +def indentation_filter(tokens): + # A stack of indentation levels; will never pop item 0 + levels = [0] + token = None + depth = 0 + prev_was_ws = False + for token in tokens: + # if 1: + # print "Process", token, + # if token.at_line_start: + # print "at_line_start", + # if token.must_indent: + # print "must_indent", + # print + + # WS only occurs at the start of the line + # There may be WS followed by NEWLINE so + # only track the depth here. Don't indent/dedent + # until there's something real. + if token.type == "WS": + assert depth == 0 + depth = len(token.value) + prev_was_ws = True + # WS tokens are never passed to the parser + continue + + if token.type == "NEWLINE": + depth = 0 + if prev_was_ws or token.at_line_start: + # ignore blank lines + continue + # pass the other cases on through + yield token + continue + + # then it must be a real token (not WS, not NEWLINE) + # which can affect the indentation level + + prev_was_ws = False + if token.must_indent: + # The current depth must be larger than the previous level + if not (depth > levels[-1]): + raise IndentationError("expected an indented block") + + levels.append(depth) + yield INDENT(token.lineno) + + elif token.at_line_start: + # Must be on the same level or one of the previous levels + if depth == levels[-1]: + # At the same level + pass + elif depth > levels[-1]: + raise IndentationError( + "indentation increase but not in new block") + else: + # Back up; but only if it matches a previous level + try: + i = levels.index(depth) + except ValueError: + raise IndentationError("inconsistent indentation") + for _ in range(i + 1, len(levels)): + yield DEDENT(token.lineno) + levels.pop() + + yield token + + ### Finished processing ### + + # Must dedent any remaining levels + if len(levels) > 1: + assert token is not None + for _ in range(1, len(levels)): + yield DEDENT(token.lineno) + + +# The top-level filter adds an ENDMARKER, if requested. +# Python's grammar uses it. +def filter(lexer, add_endmarker=True): + token = None + tokens = iter(lexer.token, None) + tokens = track_tokens_filter(lexer, tokens) + for token in indentation_filter(tokens): + yield token + + if add_endmarker: + lineno = 1 + if token is not None: + lineno = token.lineno + yield _new_token("ENDMARKER", lineno) + +# Combine Ply and my filters into a new lexer + + +class IndentLexer(object): + + def __init__(self, debug=0, optimize=0, lextab='lextab', reflags=0): + self.lexer = lex.lex(debug=debug, optimize=optimize, + lextab=lextab, reflags=reflags) + self.token_stream = None + + def input(self, s, add_endmarker=True): + self.lexer.paren_count = 0 + self.lexer.input(s) + self.token_stream = filter(self.lexer, add_endmarker) + + def token(self): + try: + return self.token_stream.next() + except StopIteration: + return None + +########## Parser (tokens -> AST) ###### + +# also part of Ply +#import yacc + +# I use the Python AST +from compiler import ast + +# Helper function + + +def Assign(left, right): + names = [] + if isinstance(left, ast.Name): + # Single assignment on left + return ast.Assign([ast.AssName(left.name, 'OP_ASSIGN')], right) + elif isinstance(left, ast.Tuple): + # List of things - make sure they are Name nodes + names = [] + for child in left.getChildren(): + if not isinstance(child, ast.Name): + raise SyntaxError("that assignment not supported") + names.append(child.name) + ass_list = [ast.AssName(name, 'OP_ASSIGN') for name in names] + return ast.Assign([ast.AssTuple(ass_list)], right) + else: + raise SyntaxError("Can't do that yet") + + +# The grammar comments come from Python's Grammar/Grammar file + +# NB: compound_stmt in single_input is followed by extra NEWLINE! +# file_input: (NEWLINE | stmt)* ENDMARKER +def p_file_input_end(p): + """file_input_end : file_input ENDMARKER""" + p[0] = ast.Stmt(p[1]) + + +def p_file_input(p): + """file_input : file_input NEWLINE + | file_input stmt + | NEWLINE + | stmt""" + if isinstance(p[len(p) - 1], basestring): + if len(p) == 3: + p[0] = p[1] + else: + p[0] = [] # p == 2 --> only a blank line + else: + if len(p) == 3: + p[0] = p[1] + p[2] + else: + p[0] = p[1] + + +# funcdef: [decorators] 'def' NAME parameters ':' suite +# ignoring decorators +def p_funcdef(p): + "funcdef : DEF NAME parameters COLON suite" + p[0] = ast.Function(None, p[2], tuple(p[3]), (), 0, None, p[5]) + +# parameters: '(' [varargslist] ')' + + +def p_parameters(p): + """parameters : LPAR RPAR + | LPAR varargslist RPAR""" + if len(p) == 3: + p[0] = [] + else: + p[0] = p[2] + + +# varargslist: (fpdef ['=' test] ',')* ('*' NAME [',' '**' NAME] | '**' NAME) | +# highly simplified +def p_varargslist(p): + """varargslist : varargslist COMMA NAME + | NAME""" + if len(p) == 4: + p[0] = p[1] + p[3] + else: + p[0] = [p[1]] + +# stmt: simple_stmt | compound_stmt + + +def p_stmt_simple(p): + """stmt : simple_stmt""" + # simple_stmt is a list + p[0] = p[1] + + +def p_stmt_compound(p): + """stmt : compound_stmt""" + p[0] = [p[1]] + +# simple_stmt: small_stmt (';' small_stmt)* [';'] NEWLINE + + +def p_simple_stmt(p): + """simple_stmt : small_stmts NEWLINE + | small_stmts SEMICOLON NEWLINE""" + p[0] = p[1] + + +def p_small_stmts(p): + """small_stmts : small_stmts SEMICOLON small_stmt + | small_stmt""" + if len(p) == 4: + p[0] = p[1] + [p[3]] + else: + p[0] = [p[1]] + +# small_stmt: expr_stmt | print_stmt | del_stmt | pass_stmt | flow_stmt | +# import_stmt | global_stmt | exec_stmt | assert_stmt + + +def p_small_stmt(p): + """small_stmt : flow_stmt + | expr_stmt""" + p[0] = p[1] + +# expr_stmt: testlist (augassign (yield_expr|testlist) | +# ('=' (yield_expr|testlist))*) +# augassign: ('+=' | '-=' | '*=' | '/=' | '%=' | '&=' | '|=' | '^=' | +# '<<=' | '>>=' | '**=' | '//=') + + +def p_expr_stmt(p): + """expr_stmt : testlist ASSIGN testlist + | testlist """ + if len(p) == 2: + # a list of expressions + p[0] = ast.Discard(p[1]) + else: + p[0] = Assign(p[1], p[3]) + + +def p_flow_stmt(p): + "flow_stmt : return_stmt" + p[0] = p[1] + +# return_stmt: 'return' [testlist] + + +def p_return_stmt(p): + "return_stmt : RETURN testlist" + p[0] = ast.Return(p[2]) + + +def p_compound_stmt(p): + """compound_stmt : if_stmt + | funcdef""" + p[0] = p[1] + + +def p_if_stmt(p): + 'if_stmt : IF test COLON suite' + p[0] = ast.If([(p[2], p[4])], None) + + +def p_suite(p): + """suite : simple_stmt + | NEWLINE INDENT stmts DEDENT""" + if len(p) == 2: + p[0] = ast.Stmt(p[1]) + else: + p[0] = ast.Stmt(p[3]) + + +def p_stmts(p): + """stmts : stmts stmt + | stmt""" + if len(p) == 3: + p[0] = p[1] + p[2] + else: + p[0] = p[1] + +# No using Python's approach because Ply supports precedence + +# comparison: expr (comp_op expr)* +# arith_expr: term (('+'|'-') term)* +# term: factor (('*'|'/'|'%'|'//') factor)* +# factor: ('+'|'-'|'~') factor | power +# comp_op: '<'|'>'|'=='|'>='|'<='|'<>'|'!='|'in'|'not' 'in'|'is'|'is' 'not' + + +def make_lt_compare((left, right)): + return ast.Compare(left, [('<', right), ]) + + +def make_gt_compare((left, right)): + return ast.Compare(left, [('>', right), ]) + + +def make_eq_compare((left, right)): + return ast.Compare(left, [('==', right), ]) + + +binary_ops = { + "+": ast.Add, + "-": ast.Sub, + "*": ast.Mul, + "/": ast.Div, + "<": make_lt_compare, + ">": make_gt_compare, + "==": make_eq_compare, +} +unary_ops = { + "+": ast.UnaryAdd, + "-": ast.UnarySub, +} +precedence = ( + ("left", "EQ", "GT", "LT"), + ("left", "PLUS", "MINUS"), + ("left", "MULT", "DIV"), +) + + +def p_comparison(p): + """comparison : comparison PLUS comparison + | comparison MINUS comparison + | comparison MULT comparison + | comparison DIV comparison + | comparison LT comparison + | comparison EQ comparison + | comparison GT comparison + | PLUS comparison + | MINUS comparison + | power""" + if len(p) == 4: + p[0] = binary_ops[p[2]]((p[1], p[3])) + elif len(p) == 3: + p[0] = unary_ops[p[1]](p[2]) + else: + p[0] = p[1] + +# power: atom trailer* ['**' factor] +# trailers enables function calls. I only allow one level of calls +# so this is 'trailer' + + +def p_power(p): + """power : atom + | atom trailer""" + if len(p) == 2: + p[0] = p[1] + else: + if p[2][0] == "CALL": + p[0] = ast.CallFunc(p[1], p[2][1], None, None) + else: + raise AssertionError("not implemented") + + +def p_atom_name(p): + """atom : NAME""" + p[0] = ast.Name(p[1]) + + +def p_atom_number(p): + """atom : NUMBER + | STRING""" + p[0] = ast.Const(p[1]) + + +def p_atom_tuple(p): + """atom : LPAR testlist RPAR""" + p[0] = p[2] + +# trailer: '(' [arglist] ')' | '[' subscriptlist ']' | '.' NAME + + +def p_trailer(p): + "trailer : LPAR arglist RPAR" + p[0] = ("CALL", p[2]) + +# testlist: test (',' test)* [','] +# Contains shift/reduce error + + +def p_testlist(p): + """testlist : testlist_multi COMMA + | testlist_multi """ + if len(p) == 2: + p[0] = p[1] + else: + # May need to promote singleton to tuple + if isinstance(p[1], list): + p[0] = p[1] + else: + p[0] = [p[1]] + # Convert into a tuple? + if isinstance(p[0], list): + p[0] = ast.Tuple(p[0]) + + +def p_testlist_multi(p): + """testlist_multi : testlist_multi COMMA test + | test""" + if len(p) == 2: + # singleton + p[0] = p[1] + else: + if isinstance(p[1], list): + p[0] = p[1] + [p[3]] + else: + # singleton -> tuple + p[0] = [p[1], p[3]] + + +# test: or_test ['if' or_test 'else' test] | lambdef +# as I don't support 'and', 'or', and 'not' this works down to 'comparison' +def p_test(p): + "test : comparison" + p[0] = p[1] + + +# arglist: (argument ',')* (argument [',']| '*' test [',' '**' test] | '**' test) +# XXX INCOMPLETE: this doesn't allow the trailing comma +def p_arglist(p): + """arglist : arglist COMMA argument + | argument""" + if len(p) == 4: + p[0] = p[1] + [p[3]] + else: + p[0] = [p[1]] + +# argument: test [gen_for] | test '=' test # Really [keyword '='] test + + +def p_argument(p): + "argument : test" + p[0] = p[1] + + +def p_error(p): + # print "Error!", repr(p) + raise SyntaxError(p) + + +class GardenSnakeParser(object): + + def __init__(self, lexer=None): + if lexer is None: + lexer = IndentLexer() + self.lexer = lexer + self.parser = yacc.yacc(start="file_input_end") + + def parse(self, code): + self.lexer.input(code) + result = self.parser.parse(lexer=self.lexer) + return ast.Module(None, result) + + +###### Code generation ###### + +from compiler import misc, syntax, pycodegen + + +class GardenSnakeCompiler(object): + + def __init__(self): + self.parser = GardenSnakeParser() + + def compile(self, code, filename="<string>"): + tree = self.parser.parse(code) + # print tree + misc.set_filename(filename, tree) + syntax.check(tree) + gen = pycodegen.ModuleCodeGenerator(tree) + code = gen.getCode() + return code + +####### Test code ####### + +compile = GardenSnakeCompiler().compile + +code = r""" + +print('LET\'S TRY THIS \\OUT') + +#Comment here +def x(a): + print('called with',a) + if a == 1: + return 2 + if a*2 > 10: return 999 / 4 + # Another comment here + + return a+2*3 + +ints = (1, 2, + 3, 4, +5) +print('mutiline-expression', ints) + +t = 4+1/3*2+6*(9-5+1) +print('predence test; should be 34+2/3:', t, t==(34+2/3)) + +print('numbers', 1,2,3,4,5) +if 1: + 8 + a=9 + print(x(a)) + +print(x(1)) +print(x(2)) +print(x(8),'3') +print('this is decimal', 1/5) +print('BIG DECIMAL', 1.234567891234567e12345) + +""" + +# Set up the GardenSnake run-time environment + + +def print_(*args): + print "-->", " ".join(map(str, args)) + +globals()["print"] = print_ + +compiled_code = compile(code) + +exec compiled_code in globals() +print "Done" diff --git a/third_party/python/ply/example/GardenSnake/README b/third_party/python/ply/example/GardenSnake/README new file mode 100644 index 0000000000..4d8be2db05 --- /dev/null +++ b/third_party/python/ply/example/GardenSnake/README @@ -0,0 +1,5 @@ +This example is Andrew Dalke's GardenSnake language. It shows how to process an +indentation-like language like Python. Further details can be found here: + +http://dalkescientific.com/writings/diary/archive/2006/08/30/gardensnake_language.html + diff --git a/third_party/python/ply/example/README b/third_party/python/ply/example/README new file mode 100644 index 0000000000..63519b557f --- /dev/null +++ b/third_party/python/ply/example/README @@ -0,0 +1,10 @@ +Simple examples: + calc - Simple calculator + classcalc - Simple calculate defined as a class + +Complex examples + ansic - ANSI C grammar from K&R + BASIC - A small BASIC interpreter + GardenSnake - A simple python-like language + yply - Converts Unix yacc files to PLY programs. + diff --git a/third_party/python/ply/example/ansic/README b/third_party/python/ply/example/ansic/README new file mode 100644 index 0000000000..e049d3b4e4 --- /dev/null +++ b/third_party/python/ply/example/ansic/README @@ -0,0 +1,2 @@ +This example is incomplete. Was going to specify an ANSI C parser. +This is part of it. diff --git a/third_party/python/ply/example/ansic/clex.py b/third_party/python/ply/example/ansic/clex.py new file mode 100644 index 0000000000..4bde1d730b --- /dev/null +++ b/third_party/python/ply/example/ansic/clex.py @@ -0,0 +1,168 @@ +# ---------------------------------------------------------------------- +# clex.py +# +# A lexer for ANSI C. +# ---------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +import ply.lex as lex + +# Reserved words +reserved = ( + 'AUTO', 'BREAK', 'CASE', 'CHAR', 'CONST', 'CONTINUE', 'DEFAULT', 'DO', 'DOUBLE', + 'ELSE', 'ENUM', 'EXTERN', 'FLOAT', 'FOR', 'GOTO', 'IF', 'INT', 'LONG', 'REGISTER', + 'RETURN', 'SHORT', 'SIGNED', 'SIZEOF', 'STATIC', 'STRUCT', 'SWITCH', 'TYPEDEF', + 'UNION', 'UNSIGNED', 'VOID', 'VOLATILE', 'WHILE', +) + +tokens = reserved + ( + # Literals (identifier, integer constant, float constant, string constant, + # char const) + 'ID', 'TYPEID', 'ICONST', 'FCONST', 'SCONST', 'CCONST', + + # Operators (+,-,*,/,%,|,&,~,^,<<,>>, ||, &&, !, <, <=, >, >=, ==, !=) + 'PLUS', 'MINUS', 'TIMES', 'DIVIDE', 'MOD', + 'OR', 'AND', 'NOT', 'XOR', 'LSHIFT', 'RSHIFT', + 'LOR', 'LAND', 'LNOT', + 'LT', 'LE', 'GT', 'GE', 'EQ', 'NE', + + # Assignment (=, *=, /=, %=, +=, -=, <<=, >>=, &=, ^=, |=) + 'EQUALS', 'TIMESEQUAL', 'DIVEQUAL', 'MODEQUAL', 'PLUSEQUAL', 'MINUSEQUAL', + 'LSHIFTEQUAL', 'RSHIFTEQUAL', 'ANDEQUAL', 'XOREQUAL', 'OREQUAL', + + # Increment/decrement (++,--) + 'PLUSPLUS', 'MINUSMINUS', + + # Structure dereference (->) + 'ARROW', + + # Conditional operator (?) + 'CONDOP', + + # Delimeters ( ) [ ] { } , . ; : + 'LPAREN', 'RPAREN', + 'LBRACKET', 'RBRACKET', + 'LBRACE', 'RBRACE', + 'COMMA', 'PERIOD', 'SEMI', 'COLON', + + # Ellipsis (...) + 'ELLIPSIS', +) + +# Completely ignored characters +t_ignore = ' \t\x0c' + +# Newlines + + +def t_NEWLINE(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + +# Operators +t_PLUS = r'\+' +t_MINUS = r'-' +t_TIMES = r'\*' +t_DIVIDE = r'/' +t_MOD = r'%' +t_OR = r'\|' +t_AND = r'&' +t_NOT = r'~' +t_XOR = r'\^' +t_LSHIFT = r'<<' +t_RSHIFT = r'>>' +t_LOR = r'\|\|' +t_LAND = r'&&' +t_LNOT = r'!' +t_LT = r'<' +t_GT = r'>' +t_LE = r'<=' +t_GE = r'>=' +t_EQ = r'==' +t_NE = r'!=' + +# Assignment operators + +t_EQUALS = r'=' +t_TIMESEQUAL = r'\*=' +t_DIVEQUAL = r'/=' +t_MODEQUAL = r'%=' +t_PLUSEQUAL = r'\+=' +t_MINUSEQUAL = r'-=' +t_LSHIFTEQUAL = r'<<=' +t_RSHIFTEQUAL = r'>>=' +t_ANDEQUAL = r'&=' +t_OREQUAL = r'\|=' +t_XOREQUAL = r'\^=' + +# Increment/decrement +t_PLUSPLUS = r'\+\+' +t_MINUSMINUS = r'--' + +# -> +t_ARROW = r'->' + +# ? +t_CONDOP = r'\?' + +# Delimeters +t_LPAREN = r'\(' +t_RPAREN = r'\)' +t_LBRACKET = r'\[' +t_RBRACKET = r'\]' +t_LBRACE = r'\{' +t_RBRACE = r'\}' +t_COMMA = r',' +t_PERIOD = r'\.' +t_SEMI = r';' +t_COLON = r':' +t_ELLIPSIS = r'\.\.\.' + +# Identifiers and reserved words + +reserved_map = {} +for r in reserved: + reserved_map[r.lower()] = r + + +def t_ID(t): + r'[A-Za-z_][\w_]*' + t.type = reserved_map.get(t.value, "ID") + return t + +# Integer literal +t_ICONST = r'\d+([uU]|[lL]|[uU][lL]|[lL][uU])?' + +# Floating literal +t_FCONST = r'((\d+)(\.\d+)(e(\+|-)?(\d+))? | (\d+)e(\+|-)?(\d+))([lL]|[fF])?' + +# String literal +t_SCONST = r'\"([^\\\n]|(\\.))*?\"' + +# Character constant 'c' or L'c' +t_CCONST = r'(L)?\'([^\\\n]|(\\.))*?\'' + +# Comments + + +def t_comment(t): + r'/\*(.|\n)*?\*/' + t.lexer.lineno += t.value.count('\n') + +# Preprocessor directive (ignored) + + +def t_preprocessor(t): + r'\#(.)*?\n' + t.lexer.lineno += 1 + + +def t_error(t): + print("Illegal character %s" % repr(t.value[0])) + t.lexer.skip(1) + +lexer = lex.lex() +if __name__ == "__main__": + lex.runmain(lexer) diff --git a/third_party/python/ply/example/ansic/cparse.py b/third_party/python/ply/example/ansic/cparse.py new file mode 100644 index 0000000000..5fe9bce042 --- /dev/null +++ b/third_party/python/ply/example/ansic/cparse.py @@ -0,0 +1,1048 @@ +# ----------------------------------------------------------------------------- +# cparse.py +# +# Simple parser for ANSI C. Based on the grammar in K&R, 2nd Ed. +# ----------------------------------------------------------------------------- + +import sys +import clex +import ply.yacc as yacc + +# Get the token map +tokens = clex.tokens + +# translation-unit: + + +def p_translation_unit_1(t): + 'translation_unit : external_declaration' + pass + + +def p_translation_unit_2(t): + 'translation_unit : translation_unit external_declaration' + pass + +# external-declaration: + + +def p_external_declaration_1(t): + 'external_declaration : function_definition' + pass + + +def p_external_declaration_2(t): + 'external_declaration : declaration' + pass + +# function-definition: + + +def p_function_definition_1(t): + 'function_definition : declaration_specifiers declarator declaration_list compound_statement' + pass + + +def p_function_definition_2(t): + 'function_definition : declarator declaration_list compound_statement' + pass + + +def p_function_definition_3(t): + 'function_definition : declarator compound_statement' + pass + + +def p_function_definition_4(t): + 'function_definition : declaration_specifiers declarator compound_statement' + pass + +# declaration: + + +def p_declaration_1(t): + 'declaration : declaration_specifiers init_declarator_list SEMI' + pass + + +def p_declaration_2(t): + 'declaration : declaration_specifiers SEMI' + pass + +# declaration-list: + + +def p_declaration_list_1(t): + 'declaration_list : declaration' + pass + + +def p_declaration_list_2(t): + 'declaration_list : declaration_list declaration ' + pass + +# declaration-specifiers + + +def p_declaration_specifiers_1(t): + 'declaration_specifiers : storage_class_specifier declaration_specifiers' + pass + + +def p_declaration_specifiers_2(t): + 'declaration_specifiers : type_specifier declaration_specifiers' + pass + + +def p_declaration_specifiers_3(t): + 'declaration_specifiers : type_qualifier declaration_specifiers' + pass + + +def p_declaration_specifiers_4(t): + 'declaration_specifiers : storage_class_specifier' + pass + + +def p_declaration_specifiers_5(t): + 'declaration_specifiers : type_specifier' + pass + + +def p_declaration_specifiers_6(t): + 'declaration_specifiers : type_qualifier' + pass + +# storage-class-specifier + + +def p_storage_class_specifier(t): + '''storage_class_specifier : AUTO + | REGISTER + | STATIC + | EXTERN + | TYPEDEF + ''' + pass + +# type-specifier: + + +def p_type_specifier(t): + '''type_specifier : VOID + | CHAR + | SHORT + | INT + | LONG + | FLOAT + | DOUBLE + | SIGNED + | UNSIGNED + | struct_or_union_specifier + | enum_specifier + | TYPEID + ''' + pass + +# type-qualifier: + + +def p_type_qualifier(t): + '''type_qualifier : CONST + | VOLATILE''' + pass + +# struct-or-union-specifier + + +def p_struct_or_union_specifier_1(t): + 'struct_or_union_specifier : struct_or_union ID LBRACE struct_declaration_list RBRACE' + pass + + +def p_struct_or_union_specifier_2(t): + 'struct_or_union_specifier : struct_or_union LBRACE struct_declaration_list RBRACE' + pass + + +def p_struct_or_union_specifier_3(t): + 'struct_or_union_specifier : struct_or_union ID' + pass + +# struct-or-union: + + +def p_struct_or_union(t): + '''struct_or_union : STRUCT + | UNION + ''' + pass + +# struct-declaration-list: + + +def p_struct_declaration_list_1(t): + 'struct_declaration_list : struct_declaration' + pass + + +def p_struct_declaration_list_2(t): + 'struct_declaration_list : struct_declaration_list struct_declaration' + pass + +# init-declarator-list: + + +def p_init_declarator_list_1(t): + 'init_declarator_list : init_declarator' + pass + + +def p_init_declarator_list_2(t): + 'init_declarator_list : init_declarator_list COMMA init_declarator' + pass + +# init-declarator + + +def p_init_declarator_1(t): + 'init_declarator : declarator' + pass + + +def p_init_declarator_2(t): + 'init_declarator : declarator EQUALS initializer' + pass + +# struct-declaration: + + +def p_struct_declaration(t): + 'struct_declaration : specifier_qualifier_list struct_declarator_list SEMI' + pass + +# specifier-qualifier-list: + + +def p_specifier_qualifier_list_1(t): + 'specifier_qualifier_list : type_specifier specifier_qualifier_list' + pass + + +def p_specifier_qualifier_list_2(t): + 'specifier_qualifier_list : type_specifier' + pass + + +def p_specifier_qualifier_list_3(t): + 'specifier_qualifier_list : type_qualifier specifier_qualifier_list' + pass + + +def p_specifier_qualifier_list_4(t): + 'specifier_qualifier_list : type_qualifier' + pass + +# struct-declarator-list: + + +def p_struct_declarator_list_1(t): + 'struct_declarator_list : struct_declarator' + pass + + +def p_struct_declarator_list_2(t): + 'struct_declarator_list : struct_declarator_list COMMA struct_declarator' + pass + +# struct-declarator: + + +def p_struct_declarator_1(t): + 'struct_declarator : declarator' + pass + + +def p_struct_declarator_2(t): + 'struct_declarator : declarator COLON constant_expression' + pass + + +def p_struct_declarator_3(t): + 'struct_declarator : COLON constant_expression' + pass + +# enum-specifier: + + +def p_enum_specifier_1(t): + 'enum_specifier : ENUM ID LBRACE enumerator_list RBRACE' + pass + + +def p_enum_specifier_2(t): + 'enum_specifier : ENUM LBRACE enumerator_list RBRACE' + pass + + +def p_enum_specifier_3(t): + 'enum_specifier : ENUM ID' + pass + +# enumerator_list: + + +def p_enumerator_list_1(t): + 'enumerator_list : enumerator' + pass + + +def p_enumerator_list_2(t): + 'enumerator_list : enumerator_list COMMA enumerator' + pass + +# enumerator: + + +def p_enumerator_1(t): + 'enumerator : ID' + pass + + +def p_enumerator_2(t): + 'enumerator : ID EQUALS constant_expression' + pass + +# declarator: + + +def p_declarator_1(t): + 'declarator : pointer direct_declarator' + pass + + +def p_declarator_2(t): + 'declarator : direct_declarator' + pass + +# direct-declarator: + + +def p_direct_declarator_1(t): + 'direct_declarator : ID' + pass + + +def p_direct_declarator_2(t): + 'direct_declarator : LPAREN declarator RPAREN' + pass + + +def p_direct_declarator_3(t): + 'direct_declarator : direct_declarator LBRACKET constant_expression_opt RBRACKET' + pass + + +def p_direct_declarator_4(t): + 'direct_declarator : direct_declarator LPAREN parameter_type_list RPAREN ' + pass + + +def p_direct_declarator_5(t): + 'direct_declarator : direct_declarator LPAREN identifier_list RPAREN ' + pass + + +def p_direct_declarator_6(t): + 'direct_declarator : direct_declarator LPAREN RPAREN ' + pass + +# pointer: + + +def p_pointer_1(t): + 'pointer : TIMES type_qualifier_list' + pass + + +def p_pointer_2(t): + 'pointer : TIMES' + pass + + +def p_pointer_3(t): + 'pointer : TIMES type_qualifier_list pointer' + pass + + +def p_pointer_4(t): + 'pointer : TIMES pointer' + pass + +# type-qualifier-list: + + +def p_type_qualifier_list_1(t): + 'type_qualifier_list : type_qualifier' + pass + + +def p_type_qualifier_list_2(t): + 'type_qualifier_list : type_qualifier_list type_qualifier' + pass + +# parameter-type-list: + + +def p_parameter_type_list_1(t): + 'parameter_type_list : parameter_list' + pass + + +def p_parameter_type_list_2(t): + 'parameter_type_list : parameter_list COMMA ELLIPSIS' + pass + +# parameter-list: + + +def p_parameter_list_1(t): + 'parameter_list : parameter_declaration' + pass + + +def p_parameter_list_2(t): + 'parameter_list : parameter_list COMMA parameter_declaration' + pass + +# parameter-declaration: + + +def p_parameter_declaration_1(t): + 'parameter_declaration : declaration_specifiers declarator' + pass + + +def p_parameter_declaration_2(t): + 'parameter_declaration : declaration_specifiers abstract_declarator_opt' + pass + +# identifier-list: + + +def p_identifier_list_1(t): + 'identifier_list : ID' + pass + + +def p_identifier_list_2(t): + 'identifier_list : identifier_list COMMA ID' + pass + +# initializer: + + +def p_initializer_1(t): + 'initializer : assignment_expression' + pass + + +def p_initializer_2(t): + '''initializer : LBRACE initializer_list RBRACE + | LBRACE initializer_list COMMA RBRACE''' + pass + +# initializer-list: + + +def p_initializer_list_1(t): + 'initializer_list : initializer' + pass + + +def p_initializer_list_2(t): + 'initializer_list : initializer_list COMMA initializer' + pass + +# type-name: + + +def p_type_name(t): + 'type_name : specifier_qualifier_list abstract_declarator_opt' + pass + + +def p_abstract_declarator_opt_1(t): + 'abstract_declarator_opt : empty' + pass + + +def p_abstract_declarator_opt_2(t): + 'abstract_declarator_opt : abstract_declarator' + pass + +# abstract-declarator: + + +def p_abstract_declarator_1(t): + 'abstract_declarator : pointer ' + pass + + +def p_abstract_declarator_2(t): + 'abstract_declarator : pointer direct_abstract_declarator' + pass + + +def p_abstract_declarator_3(t): + 'abstract_declarator : direct_abstract_declarator' + pass + +# direct-abstract-declarator: + + +def p_direct_abstract_declarator_1(t): + 'direct_abstract_declarator : LPAREN abstract_declarator RPAREN' + pass + + +def p_direct_abstract_declarator_2(t): + 'direct_abstract_declarator : direct_abstract_declarator LBRACKET constant_expression_opt RBRACKET' + pass + + +def p_direct_abstract_declarator_3(t): + 'direct_abstract_declarator : LBRACKET constant_expression_opt RBRACKET' + pass + + +def p_direct_abstract_declarator_4(t): + 'direct_abstract_declarator : direct_abstract_declarator LPAREN parameter_type_list_opt RPAREN' + pass + + +def p_direct_abstract_declarator_5(t): + 'direct_abstract_declarator : LPAREN parameter_type_list_opt RPAREN' + pass + +# Optional fields in abstract declarators + + +def p_constant_expression_opt_1(t): + 'constant_expression_opt : empty' + pass + + +def p_constant_expression_opt_2(t): + 'constant_expression_opt : constant_expression' + pass + + +def p_parameter_type_list_opt_1(t): + 'parameter_type_list_opt : empty' + pass + + +def p_parameter_type_list_opt_2(t): + 'parameter_type_list_opt : parameter_type_list' + pass + +# statement: + + +def p_statement(t): + ''' + statement : labeled_statement + | expression_statement + | compound_statement + | selection_statement + | iteration_statement + | jump_statement + ''' + pass + +# labeled-statement: + + +def p_labeled_statement_1(t): + 'labeled_statement : ID COLON statement' + pass + + +def p_labeled_statement_2(t): + 'labeled_statement : CASE constant_expression COLON statement' + pass + + +def p_labeled_statement_3(t): + 'labeled_statement : DEFAULT COLON statement' + pass + +# expression-statement: + + +def p_expression_statement(t): + 'expression_statement : expression_opt SEMI' + pass + +# compound-statement: + + +def p_compound_statement_1(t): + 'compound_statement : LBRACE declaration_list statement_list RBRACE' + pass + + +def p_compound_statement_2(t): + 'compound_statement : LBRACE statement_list RBRACE' + pass + + +def p_compound_statement_3(t): + 'compound_statement : LBRACE declaration_list RBRACE' + pass + + +def p_compound_statement_4(t): + 'compound_statement : LBRACE RBRACE' + pass + +# statement-list: + + +def p_statement_list_1(t): + 'statement_list : statement' + pass + + +def p_statement_list_2(t): + 'statement_list : statement_list statement' + pass + +# selection-statement + + +def p_selection_statement_1(t): + 'selection_statement : IF LPAREN expression RPAREN statement' + pass + + +def p_selection_statement_2(t): + 'selection_statement : IF LPAREN expression RPAREN statement ELSE statement ' + pass + + +def p_selection_statement_3(t): + 'selection_statement : SWITCH LPAREN expression RPAREN statement ' + pass + +# iteration_statement: + + +def p_iteration_statement_1(t): + 'iteration_statement : WHILE LPAREN expression RPAREN statement' + pass + + +def p_iteration_statement_2(t): + 'iteration_statement : FOR LPAREN expression_opt SEMI expression_opt SEMI expression_opt RPAREN statement ' + pass + + +def p_iteration_statement_3(t): + 'iteration_statement : DO statement WHILE LPAREN expression RPAREN SEMI' + pass + +# jump_statement: + + +def p_jump_statement_1(t): + 'jump_statement : GOTO ID SEMI' + pass + + +def p_jump_statement_2(t): + 'jump_statement : CONTINUE SEMI' + pass + + +def p_jump_statement_3(t): + 'jump_statement : BREAK SEMI' + pass + + +def p_jump_statement_4(t): + 'jump_statement : RETURN expression_opt SEMI' + pass + + +def p_expression_opt_1(t): + 'expression_opt : empty' + pass + + +def p_expression_opt_2(t): + 'expression_opt : expression' + pass + +# expression: + + +def p_expression_1(t): + 'expression : assignment_expression' + pass + + +def p_expression_2(t): + 'expression : expression COMMA assignment_expression' + pass + +# assigment_expression: + + +def p_assignment_expression_1(t): + 'assignment_expression : conditional_expression' + pass + + +def p_assignment_expression_2(t): + 'assignment_expression : unary_expression assignment_operator assignment_expression' + pass + +# assignment_operator: + + +def p_assignment_operator(t): + ''' + assignment_operator : EQUALS + | TIMESEQUAL + | DIVEQUAL + | MODEQUAL + | PLUSEQUAL + | MINUSEQUAL + | LSHIFTEQUAL + | RSHIFTEQUAL + | ANDEQUAL + | OREQUAL + | XOREQUAL + ''' + pass + +# conditional-expression + + +def p_conditional_expression_1(t): + 'conditional_expression : logical_or_expression' + pass + + +def p_conditional_expression_2(t): + 'conditional_expression : logical_or_expression CONDOP expression COLON conditional_expression ' + pass + +# constant-expression + + +def p_constant_expression(t): + 'constant_expression : conditional_expression' + pass + +# logical-or-expression + + +def p_logical_or_expression_1(t): + 'logical_or_expression : logical_and_expression' + pass + + +def p_logical_or_expression_2(t): + 'logical_or_expression : logical_or_expression LOR logical_and_expression' + pass + +# logical-and-expression + + +def p_logical_and_expression_1(t): + 'logical_and_expression : inclusive_or_expression' + pass + + +def p_logical_and_expression_2(t): + 'logical_and_expression : logical_and_expression LAND inclusive_or_expression' + pass + +# inclusive-or-expression: + + +def p_inclusive_or_expression_1(t): + 'inclusive_or_expression : exclusive_or_expression' + pass + + +def p_inclusive_or_expression_2(t): + 'inclusive_or_expression : inclusive_or_expression OR exclusive_or_expression' + pass + +# exclusive-or-expression: + + +def p_exclusive_or_expression_1(t): + 'exclusive_or_expression : and_expression' + pass + + +def p_exclusive_or_expression_2(t): + 'exclusive_or_expression : exclusive_or_expression XOR and_expression' + pass + +# AND-expression + + +def p_and_expression_1(t): + 'and_expression : equality_expression' + pass + + +def p_and_expression_2(t): + 'and_expression : and_expression AND equality_expression' + pass + + +# equality-expression: +def p_equality_expression_1(t): + 'equality_expression : relational_expression' + pass + + +def p_equality_expression_2(t): + 'equality_expression : equality_expression EQ relational_expression' + pass + + +def p_equality_expression_3(t): + 'equality_expression : equality_expression NE relational_expression' + pass + + +# relational-expression: +def p_relational_expression_1(t): + 'relational_expression : shift_expression' + pass + + +def p_relational_expression_2(t): + 'relational_expression : relational_expression LT shift_expression' + pass + + +def p_relational_expression_3(t): + 'relational_expression : relational_expression GT shift_expression' + pass + + +def p_relational_expression_4(t): + 'relational_expression : relational_expression LE shift_expression' + pass + + +def p_relational_expression_5(t): + 'relational_expression : relational_expression GE shift_expression' + pass + +# shift-expression + + +def p_shift_expression_1(t): + 'shift_expression : additive_expression' + pass + + +def p_shift_expression_2(t): + 'shift_expression : shift_expression LSHIFT additive_expression' + pass + + +def p_shift_expression_3(t): + 'shift_expression : shift_expression RSHIFT additive_expression' + pass + +# additive-expression + + +def p_additive_expression_1(t): + 'additive_expression : multiplicative_expression' + pass + + +def p_additive_expression_2(t): + 'additive_expression : additive_expression PLUS multiplicative_expression' + pass + + +def p_additive_expression_3(t): + 'additive_expression : additive_expression MINUS multiplicative_expression' + pass + +# multiplicative-expression + + +def p_multiplicative_expression_1(t): + 'multiplicative_expression : cast_expression' + pass + + +def p_multiplicative_expression_2(t): + 'multiplicative_expression : multiplicative_expression TIMES cast_expression' + pass + + +def p_multiplicative_expression_3(t): + 'multiplicative_expression : multiplicative_expression DIVIDE cast_expression' + pass + + +def p_multiplicative_expression_4(t): + 'multiplicative_expression : multiplicative_expression MOD cast_expression' + pass + +# cast-expression: + + +def p_cast_expression_1(t): + 'cast_expression : unary_expression' + pass + + +def p_cast_expression_2(t): + 'cast_expression : LPAREN type_name RPAREN cast_expression' + pass + +# unary-expression: + + +def p_unary_expression_1(t): + 'unary_expression : postfix_expression' + pass + + +def p_unary_expression_2(t): + 'unary_expression : PLUSPLUS unary_expression' + pass + + +def p_unary_expression_3(t): + 'unary_expression : MINUSMINUS unary_expression' + pass + + +def p_unary_expression_4(t): + 'unary_expression : unary_operator cast_expression' + pass + + +def p_unary_expression_5(t): + 'unary_expression : SIZEOF unary_expression' + pass + + +def p_unary_expression_6(t): + 'unary_expression : SIZEOF LPAREN type_name RPAREN' + pass + +# unary-operator + + +def p_unary_operator(t): + '''unary_operator : AND + | TIMES + | PLUS + | MINUS + | NOT + | LNOT ''' + pass + +# postfix-expression: + + +def p_postfix_expression_1(t): + 'postfix_expression : primary_expression' + pass + + +def p_postfix_expression_2(t): + 'postfix_expression : postfix_expression LBRACKET expression RBRACKET' + pass + + +def p_postfix_expression_3(t): + 'postfix_expression : postfix_expression LPAREN argument_expression_list RPAREN' + pass + + +def p_postfix_expression_4(t): + 'postfix_expression : postfix_expression LPAREN RPAREN' + pass + + +def p_postfix_expression_5(t): + 'postfix_expression : postfix_expression PERIOD ID' + pass + + +def p_postfix_expression_6(t): + 'postfix_expression : postfix_expression ARROW ID' + pass + + +def p_postfix_expression_7(t): + 'postfix_expression : postfix_expression PLUSPLUS' + pass + + +def p_postfix_expression_8(t): + 'postfix_expression : postfix_expression MINUSMINUS' + pass + +# primary-expression: + + +def p_primary_expression(t): + '''primary_expression : ID + | constant + | SCONST + | LPAREN expression RPAREN''' + pass + +# argument-expression-list: + + +def p_argument_expression_list(t): + '''argument_expression_list : assignment_expression + | argument_expression_list COMMA assignment_expression''' + pass + +# constant: + + +def p_constant(t): + '''constant : ICONST + | FCONST + | CCONST''' + pass + + +def p_empty(t): + 'empty : ' + pass + + +def p_error(t): + print("Whoa. We're hosed") + +import profile +# Build the grammar + +yacc.yacc() +#yacc.yacc(method='LALR',write_tables=False,debug=False) + +#profile.run("yacc.yacc(method='LALR')") diff --git a/third_party/python/ply/example/calc/calc.py b/third_party/python/ply/example/calc/calc.py new file mode 100644 index 0000000000..824c3d7d0a --- /dev/null +++ b/third_party/python/ply/example/calc/calc.py @@ -0,0 +1,123 @@ +# ----------------------------------------------------------------------------- +# calc.py +# +# A simple calculator with variables. This is from O'Reilly's +# "Lex and Yacc", p. 63. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +tokens = ( + 'NAME', 'NUMBER', +) + +literals = ['=', '+', '-', '*', '/', '(', ')'] + +# Tokens + +t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + +def t_NUMBER(t): + r'\d+' + t.value = int(t.value) + return t + +t_ignore = " \t" + + +def t_newline(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + +def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + +# Build the lexer +import ply.lex as lex +lex.lex() + +# Parsing rules + +precedence = ( + ('left', '+', '-'), + ('left', '*', '/'), + ('right', 'UMINUS'), +) + +# dictionary of names +names = {} + + +def p_statement_assign(p): + 'statement : NAME "=" expression' + names[p[1]] = p[3] + + +def p_statement_expr(p): + 'statement : expression' + print(p[1]) + + +def p_expression_binop(p): + '''expression : expression '+' expression + | expression '-' expression + | expression '*' expression + | expression '/' expression''' + if p[2] == '+': + p[0] = p[1] + p[3] + elif p[2] == '-': + p[0] = p[1] - p[3] + elif p[2] == '*': + p[0] = p[1] * p[3] + elif p[2] == '/': + p[0] = p[1] / p[3] + + +def p_expression_uminus(p): + "expression : '-' expression %prec UMINUS" + p[0] = -p[2] + + +def p_expression_group(p): + "expression : '(' expression ')'" + p[0] = p[2] + + +def p_expression_number(p): + "expression : NUMBER" + p[0] = p[1] + + +def p_expression_name(p): + "expression : NAME" + try: + p[0] = names[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + +def p_error(p): + if p: + print("Syntax error at '%s'" % p.value) + else: + print("Syntax error at EOF") + +import ply.yacc as yacc +yacc.yacc() + +while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + if not s: + continue + yacc.parse(s) diff --git a/third_party/python/ply/example/calcdebug/calc.py b/third_party/python/ply/example/calcdebug/calc.py new file mode 100644 index 0000000000..06831e2ca5 --- /dev/null +++ b/third_party/python/ply/example/calcdebug/calc.py @@ -0,0 +1,129 @@ +# ----------------------------------------------------------------------------- +# calc.py +# +# This example shows how to run the parser in a debugging mode +# with output routed to a logging object. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +tokens = ( + 'NAME', 'NUMBER', +) + +literals = ['=', '+', '-', '*', '/', '(', ')'] + +# Tokens + +t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + +def t_NUMBER(t): + r'\d+' + t.value = int(t.value) + return t + +t_ignore = " \t" + + +def t_newline(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + +def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + +# Build the lexer +import ply.lex as lex +lex.lex() + +# Parsing rules + +precedence = ( + ('left', '+', '-'), + ('left', '*', '/'), + ('right', 'UMINUS'), +) + +# dictionary of names +names = {} + + +def p_statement_assign(p): + 'statement : NAME "=" expression' + names[p[1]] = p[3] + + +def p_statement_expr(p): + 'statement : expression' + print(p[1]) + + +def p_expression_binop(p): + '''expression : expression '+' expression + | expression '-' expression + | expression '*' expression + | expression '/' expression''' + if p[2] == '+': + p[0] = p[1] + p[3] + elif p[2] == '-': + p[0] = p[1] - p[3] + elif p[2] == '*': + p[0] = p[1] * p[3] + elif p[2] == '/': + p[0] = p[1] / p[3] + + +def p_expression_uminus(p): + "expression : '-' expression %prec UMINUS" + p[0] = -p[2] + + +def p_expression_group(p): + "expression : '(' expression ')'" + p[0] = p[2] + + +def p_expression_number(p): + "expression : NUMBER" + p[0] = p[1] + + +def p_expression_name(p): + "expression : NAME" + try: + p[0] = names[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + +def p_error(p): + if p: + print("Syntax error at '%s'" % p.value) + else: + print("Syntax error at EOF") + +import ply.yacc as yacc +yacc.yacc() + +import logging +logging.basicConfig( + level=logging.INFO, + filename="parselog.txt" +) + +while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + if not s: + continue + yacc.parse(s, debug=logging.getLogger()) diff --git a/third_party/python/ply/example/calceof/calc.py b/third_party/python/ply/example/calceof/calc.py new file mode 100644 index 0000000000..22b39a41a8 --- /dev/null +++ b/third_party/python/ply/example/calceof/calc.py @@ -0,0 +1,132 @@ +# ----------------------------------------------------------------------------- +# calc.py +# +# A simple calculator with variables. Asks the user for more input and +# demonstrates the use of the t_eof() rule. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +tokens = ( + 'NAME', 'NUMBER', +) + +literals = ['=', '+', '-', '*', '/', '(', ')'] + +# Tokens + +t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + +def t_NUMBER(t): + r'\d+' + t.value = int(t.value) + return t + +t_ignore = " \t" + + +def t_newline(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + +def t_eof(t): + more = raw_input('... ') + if more: + t.lexer.input(more + '\n') + return t.lexer.token() + else: + return None + + +def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + +# Build the lexer +import ply.lex as lex +lex.lex() + +# Parsing rules + +precedence = ( + ('left', '+', '-'), + ('left', '*', '/'), + ('right', 'UMINUS'), +) + +# dictionary of names +names = {} + + +def p_statement_assign(p): + 'statement : NAME "=" expression' + names[p[1]] = p[3] + + +def p_statement_expr(p): + 'statement : expression' + print(p[1]) + + +def p_expression_binop(p): + '''expression : expression '+' expression + | expression '-' expression + | expression '*' expression + | expression '/' expression''' + if p[2] == '+': + p[0] = p[1] + p[3] + elif p[2] == '-': + p[0] = p[1] - p[3] + elif p[2] == '*': + p[0] = p[1] * p[3] + elif p[2] == '/': + p[0] = p[1] / p[3] + + +def p_expression_uminus(p): + "expression : '-' expression %prec UMINUS" + p[0] = -p[2] + + +def p_expression_group(p): + "expression : '(' expression ')'" + p[0] = p[2] + + +def p_expression_number(p): + "expression : NUMBER" + p[0] = p[1] + + +def p_expression_name(p): + "expression : NAME" + try: + p[0] = names[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + +def p_error(p): + if p: + print("Syntax error at '%s'" % p.value) + else: + print("Syntax error at EOF") + +import ply.yacc as yacc +yacc.yacc() + +while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + if not s: + continue + yacc.parse(s + '\n') diff --git a/third_party/python/ply/example/classcalc/calc.py b/third_party/python/ply/example/classcalc/calc.py new file mode 100755 index 0000000000..ada4afd426 --- /dev/null +++ b/third_party/python/ply/example/classcalc/calc.py @@ -0,0 +1,165 @@ +#!/usr/bin/env python + +# ----------------------------------------------------------------------------- +# calc.py +# +# A simple calculator with variables. This is from O'Reilly's +# "Lex and Yacc", p. 63. +# +# Class-based example contributed to PLY by David McNab +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +import ply.lex as lex +import ply.yacc as yacc +import os + + +class Parser: + """ + Base class for a lexer/parser that has the rules defined as methods + """ + tokens = () + precedence = () + + def __init__(self, **kw): + self.debug = kw.get('debug', 0) + self.names = {} + try: + modname = os.path.split(os.path.splitext(__file__)[0])[ + 1] + "_" + self.__class__.__name__ + except: + modname = "parser" + "_" + self.__class__.__name__ + self.debugfile = modname + ".dbg" + self.tabmodule = modname + "_" + "parsetab" + # print self.debugfile, self.tabmodule + + # Build the lexer and parser + lex.lex(module=self, debug=self.debug) + yacc.yacc(module=self, + debug=self.debug, + debugfile=self.debugfile, + tabmodule=self.tabmodule) + + def run(self): + while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + if not s: + continue + yacc.parse(s) + + +class Calc(Parser): + + tokens = ( + 'NAME', 'NUMBER', + 'PLUS', 'MINUS', 'EXP', 'TIMES', 'DIVIDE', 'EQUALS', + 'LPAREN', 'RPAREN', + ) + + # Tokens + + t_PLUS = r'\+' + t_MINUS = r'-' + t_EXP = r'\*\*' + t_TIMES = r'\*' + t_DIVIDE = r'/' + t_EQUALS = r'=' + t_LPAREN = r'\(' + t_RPAREN = r'\)' + t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + def t_NUMBER(self, t): + r'\d+' + try: + t.value = int(t.value) + except ValueError: + print("Integer value too large %s" % t.value) + t.value = 0 + # print "parsed number %s" % repr(t.value) + return t + + t_ignore = " \t" + + def t_newline(self, t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + def t_error(self, t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + + # Parsing rules + + precedence = ( + ('left', 'PLUS', 'MINUS'), + ('left', 'TIMES', 'DIVIDE'), + ('left', 'EXP'), + ('right', 'UMINUS'), + ) + + def p_statement_assign(self, p): + 'statement : NAME EQUALS expression' + self.names[p[1]] = p[3] + + def p_statement_expr(self, p): + 'statement : expression' + print(p[1]) + + def p_expression_binop(self, p): + """ + expression : expression PLUS expression + | expression MINUS expression + | expression TIMES expression + | expression DIVIDE expression + | expression EXP expression + """ + # print [repr(p[i]) for i in range(0,4)] + if p[2] == '+': + p[0] = p[1] + p[3] + elif p[2] == '-': + p[0] = p[1] - p[3] + elif p[2] == '*': + p[0] = p[1] * p[3] + elif p[2] == '/': + p[0] = p[1] / p[3] + elif p[2] == '**': + p[0] = p[1] ** p[3] + + def p_expression_uminus(self, p): + 'expression : MINUS expression %prec UMINUS' + p[0] = -p[2] + + def p_expression_group(self, p): + 'expression : LPAREN expression RPAREN' + p[0] = p[2] + + def p_expression_number(self, p): + 'expression : NUMBER' + p[0] = p[1] + + def p_expression_name(self, p): + 'expression : NAME' + try: + p[0] = self.names[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + def p_error(self, p): + if p: + print("Syntax error at '%s'" % p.value) + else: + print("Syntax error at EOF") + +if __name__ == '__main__': + calc = Calc() + calc.run() diff --git a/third_party/python/ply/example/cleanup.sh b/third_party/python/ply/example/cleanup.sh new file mode 100755 index 0000000000..3e115f41c4 --- /dev/null +++ b/third_party/python/ply/example/cleanup.sh @@ -0,0 +1,2 @@ +#!/bin/sh +rm -f */*.pyc */parsetab.py */parser.out */*~ */*.class diff --git a/third_party/python/ply/example/closurecalc/calc.py b/third_party/python/ply/example/closurecalc/calc.py new file mode 100644 index 0000000000..6031b05813 --- /dev/null +++ b/third_party/python/ply/example/closurecalc/calc.py @@ -0,0 +1,132 @@ +# ----------------------------------------------------------------------------- +# calc.py +# +# A calculator parser that makes use of closures. The function make_calculator() +# returns a function that accepts an input string and returns a result. All +# lexing rules, parsing rules, and internal state are held inside the function. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +# Make a calculator function + + +def make_calculator(): + import ply.lex as lex + import ply.yacc as yacc + + # ------- Internal calculator state + + variables = {} # Dictionary of stored variables + + # ------- Calculator tokenizing rules + + tokens = ( + 'NAME', 'NUMBER', + ) + + literals = ['=', '+', '-', '*', '/', '(', ')'] + + t_ignore = " \t" + + t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + def t_NUMBER(t): + r'\d+' + t.value = int(t.value) + return t + + def t_newline(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + + # Build the lexer + lexer = lex.lex() + + # ------- Calculator parsing rules + + precedence = ( + ('left', '+', '-'), + ('left', '*', '/'), + ('right', 'UMINUS'), + ) + + def p_statement_assign(p): + 'statement : NAME "=" expression' + variables[p[1]] = p[3] + p[0] = None + + def p_statement_expr(p): + 'statement : expression' + p[0] = p[1] + + def p_expression_binop(p): + '''expression : expression '+' expression + | expression '-' expression + | expression '*' expression + | expression '/' expression''' + if p[2] == '+': + p[0] = p[1] + p[3] + elif p[2] == '-': + p[0] = p[1] - p[3] + elif p[2] == '*': + p[0] = p[1] * p[3] + elif p[2] == '/': + p[0] = p[1] / p[3] + + def p_expression_uminus(p): + "expression : '-' expression %prec UMINUS" + p[0] = -p[2] + + def p_expression_group(p): + "expression : '(' expression ')'" + p[0] = p[2] + + def p_expression_number(p): + "expression : NUMBER" + p[0] = p[1] + + def p_expression_name(p): + "expression : NAME" + try: + p[0] = variables[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + def p_error(p): + if p: + print("Syntax error at '%s'" % p.value) + else: + print("Syntax error at EOF") + + # Build the parser + parser = yacc.yacc() + + # ------- Input function + + def input(text): + result = parser.parse(text, lexer=lexer) + return result + + return input + +# Make a calculator object and use it +calc = make_calculator() + +while True: + try: + s = raw_input("calc > ") + except EOFError: + break + r = calc(s) + if r: + print(r) diff --git a/third_party/python/ply/example/hedit/hedit.py b/third_party/python/ply/example/hedit/hedit.py new file mode 100644 index 0000000000..32da745677 --- /dev/null +++ b/third_party/python/ply/example/hedit/hedit.py @@ -0,0 +1,48 @@ +# ----------------------------------------------------------------------------- +# hedit.py +# +# Paring of Fortran H Edit descriptions (Contributed by Pearu Peterson) +# +# These tokens can't be easily tokenized because they are of the following +# form: +# +# nHc1...cn +# +# where n is a positive integer and c1 ... cn are characters. +# +# This example shows how to modify the state of the lexer to parse +# such tokens +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + + +tokens = ( + 'H_EDIT_DESCRIPTOR', +) + +# Tokens +t_ignore = " \t\n" + + +def t_H_EDIT_DESCRIPTOR(t): + r"\d+H.*" # This grabs all of the remaining text + i = t.value.index('H') + n = eval(t.value[:i]) + + # Adjust the tokenizing position + t.lexer.lexpos -= len(t.value) - (i + 1 + n) + + t.value = t.value[i + 1:i + 1 + n] + return t + + +def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + +# Build the lexer +import ply.lex as lex +lex.lex() +lex.runmain() diff --git a/third_party/python/ply/example/newclasscalc/calc.py b/third_party/python/ply/example/newclasscalc/calc.py new file mode 100755 index 0000000000..43c9506a8a --- /dev/null +++ b/third_party/python/ply/example/newclasscalc/calc.py @@ -0,0 +1,167 @@ +#!/usr/bin/env python + +# ----------------------------------------------------------------------------- +# calc.py +# +# A simple calculator with variables. This is from O'Reilly's +# "Lex and Yacc", p. 63. +# +# Class-based example contributed to PLY by David McNab. +# +# Modified to use new-style classes. Test case. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +import ply.lex as lex +import ply.yacc as yacc +import os + + +class Parser(object): + """ + Base class for a lexer/parser that has the rules defined as methods + """ + tokens = () + precedence = () + + def __init__(self, **kw): + self.debug = kw.get('debug', 0) + self.names = {} + try: + modname = os.path.split(os.path.splitext(__file__)[0])[ + 1] + "_" + self.__class__.__name__ + except: + modname = "parser" + "_" + self.__class__.__name__ + self.debugfile = modname + ".dbg" + self.tabmodule = modname + "_" + "parsetab" + # print self.debugfile, self.tabmodule + + # Build the lexer and parser + lex.lex(module=self, debug=self.debug) + yacc.yacc(module=self, + debug=self.debug, + debugfile=self.debugfile, + tabmodule=self.tabmodule) + + def run(self): + while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + if not s: + continue + yacc.parse(s) + + +class Calc(Parser): + + tokens = ( + 'NAME', 'NUMBER', + 'PLUS', 'MINUS', 'EXP', 'TIMES', 'DIVIDE', 'EQUALS', + 'LPAREN', 'RPAREN', + ) + + # Tokens + + t_PLUS = r'\+' + t_MINUS = r'-' + t_EXP = r'\*\*' + t_TIMES = r'\*' + t_DIVIDE = r'/' + t_EQUALS = r'=' + t_LPAREN = r'\(' + t_RPAREN = r'\)' + t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + def t_NUMBER(self, t): + r'\d+' + try: + t.value = int(t.value) + except ValueError: + print("Integer value too large %s" % t.value) + t.value = 0 + # print "parsed number %s" % repr(t.value) + return t + + t_ignore = " \t" + + def t_newline(self, t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + def t_error(self, t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + + # Parsing rules + + precedence = ( + ('left', 'PLUS', 'MINUS'), + ('left', 'TIMES', 'DIVIDE'), + ('left', 'EXP'), + ('right', 'UMINUS'), + ) + + def p_statement_assign(self, p): + 'statement : NAME EQUALS expression' + self.names[p[1]] = p[3] + + def p_statement_expr(self, p): + 'statement : expression' + print(p[1]) + + def p_expression_binop(self, p): + """ + expression : expression PLUS expression + | expression MINUS expression + | expression TIMES expression + | expression DIVIDE expression + | expression EXP expression + """ + # print [repr(p[i]) for i in range(0,4)] + if p[2] == '+': + p[0] = p[1] + p[3] + elif p[2] == '-': + p[0] = p[1] - p[3] + elif p[2] == '*': + p[0] = p[1] * p[3] + elif p[2] == '/': + p[0] = p[1] / p[3] + elif p[2] == '**': + p[0] = p[1] ** p[3] + + def p_expression_uminus(self, p): + 'expression : MINUS expression %prec UMINUS' + p[0] = -p[2] + + def p_expression_group(self, p): + 'expression : LPAREN expression RPAREN' + p[0] = p[2] + + def p_expression_number(self, p): + 'expression : NUMBER' + p[0] = p[1] + + def p_expression_name(self, p): + 'expression : NAME' + try: + p[0] = self.names[p[1]] + except LookupError: + print("Undefined name '%s'" % p[1]) + p[0] = 0 + + def p_error(self, p): + if p: + print("Syntax error at '%s'" % p.value) + else: + print("Syntax error at EOF") + +if __name__ == '__main__': + calc = Calc() + calc.run() diff --git a/third_party/python/ply/example/optcalc/README b/third_party/python/ply/example/optcalc/README new file mode 100644 index 0000000000..53dd5fcd55 --- /dev/null +++ b/third_party/python/ply/example/optcalc/README @@ -0,0 +1,9 @@ +An example showing how to use Python optimized mode. +To run: + + - First run 'python calc.py' + + - Then run 'python -OO calc.py' + +If working correctly, the second version should run the +same way. diff --git a/third_party/python/ply/example/optcalc/calc.py b/third_party/python/ply/example/optcalc/calc.py new file mode 100644 index 0000000000..0c223e5994 --- /dev/null +++ b/third_party/python/ply/example/optcalc/calc.py @@ -0,0 +1,134 @@ +# ----------------------------------------------------------------------------- +# calc.py +# +# A simple calculator with variables. This is from O'Reilly's +# "Lex and Yacc", p. 63. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +if sys.version_info[0] >= 3: + raw_input = input + +tokens = ( + 'NAME', 'NUMBER', + 'PLUS', 'MINUS', 'TIMES', 'DIVIDE', 'EQUALS', + 'LPAREN', 'RPAREN', +) + +# Tokens + +t_PLUS = r'\+' +t_MINUS = r'-' +t_TIMES = r'\*' +t_DIVIDE = r'/' +t_EQUALS = r'=' +t_LPAREN = r'\(' +t_RPAREN = r'\)' +t_NAME = r'[a-zA-Z_][a-zA-Z0-9_]*' + + +def t_NUMBER(t): + r'\d+' + try: + t.value = int(t.value) + except ValueError: + print("Integer value too large %s" % t.value) + t.value = 0 + return t + +t_ignore = " \t" + + +def t_newline(t): + r'\n+' + t.lexer.lineno += t.value.count("\n") + + +def t_error(t): + print("Illegal character '%s'" % t.value[0]) + t.lexer.skip(1) + +# Build the lexer +import ply.lex as lex +lex.lex(optimize=1) + +# Parsing rules + +precedence = ( + ('left', 'PLUS', 'MINUS'), + ('left', 'TIMES', 'DIVIDE'), + ('right', 'UMINUS'), +) + +# dictionary of names +names = {} + + +def p_statement_assign(t): + 'statement : NAME EQUALS expression' + names[t[1]] = t[3] + + +def p_statement_expr(t): + 'statement : expression' + print(t[1]) + + +def p_expression_binop(t): + '''expression : expression PLUS expression + | expression MINUS expression + | expression TIMES expression + | expression DIVIDE expression''' + if t[2] == '+': + t[0] = t[1] + t[3] + elif t[2] == '-': + t[0] = t[1] - t[3] + elif t[2] == '*': + t[0] = t[1] * t[3] + elif t[2] == '/': + t[0] = t[1] / t[3] + elif t[2] == '<': + t[0] = t[1] < t[3] + + +def p_expression_uminus(t): + 'expression : MINUS expression %prec UMINUS' + t[0] = -t[2] + + +def p_expression_group(t): + 'expression : LPAREN expression RPAREN' + t[0] = t[2] + + +def p_expression_number(t): + 'expression : NUMBER' + t[0] = t[1] + + +def p_expression_name(t): + 'expression : NAME' + try: + t[0] = names[t[1]] + except LookupError: + print("Undefined name '%s'" % t[1]) + t[0] = 0 + + +def p_error(t): + if t: + print("Syntax error at '%s'" % t.value) + else: + print("Syntax error at EOF") + +import ply.yacc as yacc +yacc.yacc(optimize=1) + +while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + yacc.parse(s) diff --git a/third_party/python/ply/example/unicalc/calc.py b/third_party/python/ply/example/unicalc/calc.py new file mode 100644 index 0000000000..901c4b9d76 --- /dev/null +++ b/third_party/python/ply/example/unicalc/calc.py @@ -0,0 +1,133 @@ +# ----------------------------------------------------------------------------- +# calc.py +# +# A simple calculator with variables. This is from O'Reilly's +# "Lex and Yacc", p. 63. +# +# This example uses unicode strings for tokens, docstrings, and input. +# ----------------------------------------------------------------------------- + +import sys +sys.path.insert(0, "../..") + +tokens = ( + 'NAME', 'NUMBER', + 'PLUS', 'MINUS', 'TIMES', 'DIVIDE', 'EQUALS', + 'LPAREN', 'RPAREN', +) + +# Tokens + +t_PLUS = ur'\+' +t_MINUS = ur'-' +t_TIMES = ur'\*' +t_DIVIDE = ur'/' +t_EQUALS = ur'=' +t_LPAREN = ur'\(' +t_RPAREN = ur'\)' +t_NAME = ur'[a-zA-Z_][a-zA-Z0-9_]*' + + +def t_NUMBER(t): + ur'\d+' + try: + t.value = int(t.value) + except ValueError: + print "Integer value too large", t.value + t.value = 0 + return t + +t_ignore = u" \t" + + +def t_newline(t): + ur'\n+' + t.lexer.lineno += t.value.count("\n") + + +def t_error(t): + print "Illegal character '%s'" % t.value[0] + t.lexer.skip(1) + +# Build the lexer +import ply.lex as lex +lex.lex() + +# Parsing rules + +precedence = ( + ('left', 'PLUS', 'MINUS'), + ('left', 'TIMES', 'DIVIDE'), + ('right', 'UMINUS'), +) + +# dictionary of names +names = {} + + +def p_statement_assign(p): + 'statement : NAME EQUALS expression' + names[p[1]] = p[3] + + +def p_statement_expr(p): + 'statement : expression' + print p[1] + + +def p_expression_binop(p): + '''expression : expression PLUS expression + | expression MINUS expression + | expression TIMES expression + | expression DIVIDE expression''' + if p[2] == u'+': + p[0] = p[1] + p[3] + elif p[2] == u'-': + p[0] = p[1] - p[3] + elif p[2] == u'*': + p[0] = p[1] * p[3] + elif p[2] == u'/': + p[0] = p[1] / p[3] + + +def p_expression_uminus(p): + 'expression : MINUS expression %prec UMINUS' + p[0] = -p[2] + + +def p_expression_group(p): + 'expression : LPAREN expression RPAREN' + p[0] = p[2] + + +def p_expression_number(p): + 'expression : NUMBER' + p[0] = p[1] + + +def p_expression_name(p): + 'expression : NAME' + try: + p[0] = names[p[1]] + except LookupError: + print "Undefined name '%s'" % p[1] + p[0] = 0 + + +def p_error(p): + if p: + print "Syntax error at '%s'" % p.value + else: + print "Syntax error at EOF" + +import ply.yacc as yacc +yacc.yacc() + +while 1: + try: + s = raw_input('calc > ') + except EOFError: + break + if not s: + continue + yacc.parse(unicode(s)) diff --git a/third_party/python/ply/example/yply/README b/third_party/python/ply/example/yply/README new file mode 100644 index 0000000000..bfadf36436 --- /dev/null +++ b/third_party/python/ply/example/yply/README @@ -0,0 +1,41 @@ +yply.py + +This example implements a program yply.py that converts a UNIX-yacc +specification file into a PLY-compatible program. To use, simply +run it like this: + + % python yply.py [-nocode] inputfile.y >myparser.py + +The output of this program is Python code. In the output, +any C code in the original file is included, but is commented out. +If you use the -nocode option, then all of the C code in the +original file is just discarded. + +To use the resulting grammer with PLY, you'll need to edit the +myparser.py file. Within this file, some stub code is included that +can be used to test the construction of the parsing tables. However, +you'll need to do more editing to make a workable parser. + +Disclaimer: This just an example I threw together in an afternoon. +It might have some bugs. However, it worked when I tried it on +a yacc-specified C++ parser containing 442 rules and 855 parsing +states. + +Comments: + +1. This example does not parse specification files meant for lex/flex. + You'll need to specify the tokenizer on your own. + +2. This example shows a number of interesting PLY features including + + - Parsing of literal text delimited by nested parentheses + - Some interaction between the parser and the lexer. + - Use of literals in the grammar specification + - One pass compilation. The program just emits the result, + there is no intermediate parse tree. + +3. This program could probably be cleaned up and enhanced a lot. + It would be great if someone wanted to work on this (hint). + +-Dave + diff --git a/third_party/python/ply/example/yply/ylex.py b/third_party/python/ply/example/yply/ylex.py new file mode 100644 index 0000000000..16410e250e --- /dev/null +++ b/third_party/python/ply/example/yply/ylex.py @@ -0,0 +1,119 @@ +# lexer for yacc-grammars +# +# Author: David Beazley (dave@dabeaz.com) +# Date : October 2, 2006 + +import sys +sys.path.append("../..") + +from ply import * + +tokens = ( + 'LITERAL', 'SECTION', 'TOKEN', 'LEFT', 'RIGHT', 'PREC', 'START', 'TYPE', 'NONASSOC', 'UNION', 'CODE', + 'ID', 'QLITERAL', 'NUMBER', +) + +states = (('code', 'exclusive'),) + +literals = [';', ',', '<', '>', '|', ':'] +t_ignore = ' \t' + +t_TOKEN = r'%token' +t_LEFT = r'%left' +t_RIGHT = r'%right' +t_NONASSOC = r'%nonassoc' +t_PREC = r'%prec' +t_START = r'%start' +t_TYPE = r'%type' +t_UNION = r'%union' +t_ID = r'[a-zA-Z_][a-zA-Z_0-9]*' +t_QLITERAL = r'''(?P<quote>['"]).*?(?P=quote)''' +t_NUMBER = r'\d+' + + +def t_SECTION(t): + r'%%' + if getattr(t.lexer, "lastsection", 0): + t.value = t.lexer.lexdata[t.lexpos + 2:] + t.lexer.lexpos = len(t.lexer.lexdata) + else: + t.lexer.lastsection = 0 + return t + +# Comments + + +def t_ccomment(t): + r'/\*(.|\n)*?\*/' + t.lexer.lineno += t.value.count('\n') + +t_ignore_cppcomment = r'//.*' + + +def t_LITERAL(t): + r'%\{(.|\n)*?%\}' + t.lexer.lineno += t.value.count("\n") + return t + + +def t_NEWLINE(t): + r'\n' + t.lexer.lineno += 1 + + +def t_code(t): + r'\{' + t.lexer.codestart = t.lexpos + t.lexer.level = 1 + t.lexer.begin('code') + + +def t_code_ignore_string(t): + r'\"([^\\\n]|(\\.))*?\"' + + +def t_code_ignore_char(t): + r'\'([^\\\n]|(\\.))*?\'' + + +def t_code_ignore_comment(t): + r'/\*(.|\n)*?\*/' + + +def t_code_ignore_cppcom(t): + r'//.*' + + +def t_code_lbrace(t): + r'\{' + t.lexer.level += 1 + + +def t_code_rbrace(t): + r'\}' + t.lexer.level -= 1 + if t.lexer.level == 0: + t.type = 'CODE' + t.value = t.lexer.lexdata[t.lexer.codestart:t.lexpos + 1] + t.lexer.begin('INITIAL') + t.lexer.lineno += t.value.count('\n') + return t + +t_code_ignore_nonspace = r'[^\s\}\'\"\{]+' +t_code_ignore_whitespace = r'\s+' +t_code_ignore = "" + + +def t_code_error(t): + raise RuntimeError + + +def t_error(t): + print("%d: Illegal character '%s'" % (t.lexer.lineno, t.value[0])) + print(t.value) + t.lexer.skip(1) + +lex.lex() + +if __name__ == '__main__': + lex.runmain() diff --git a/third_party/python/ply/example/yply/yparse.py b/third_party/python/ply/example/yply/yparse.py new file mode 100644 index 0000000000..1f2e8d0922 --- /dev/null +++ b/third_party/python/ply/example/yply/yparse.py @@ -0,0 +1,244 @@ +# parser for Unix yacc-based grammars +# +# Author: David Beazley (dave@dabeaz.com) +# Date : October 2, 2006 + +import ylex +tokens = ylex.tokens + +from ply import * + +tokenlist = [] +preclist = [] + +emit_code = 1 + + +def p_yacc(p): + '''yacc : defsection rulesection''' + + +def p_defsection(p): + '''defsection : definitions SECTION + | SECTION''' + p.lexer.lastsection = 1 + print("tokens = ", repr(tokenlist)) + print() + print("precedence = ", repr(preclist)) + print() + print("# -------------- RULES ----------------") + print() + + +def p_rulesection(p): + '''rulesection : rules SECTION''' + + print("# -------------- RULES END ----------------") + print_code(p[2], 0) + + +def p_definitions(p): + '''definitions : definitions definition + | definition''' + + +def p_definition_literal(p): + '''definition : LITERAL''' + print_code(p[1], 0) + + +def p_definition_start(p): + '''definition : START ID''' + print("start = '%s'" % p[2]) + + +def p_definition_token(p): + '''definition : toktype opttype idlist optsemi ''' + for i in p[3]: + if i[0] not in "'\"": + tokenlist.append(i) + if p[1] == '%left': + preclist.append(('left',) + tuple(p[3])) + elif p[1] == '%right': + preclist.append(('right',) + tuple(p[3])) + elif p[1] == '%nonassoc': + preclist.append(('nonassoc',) + tuple(p[3])) + + +def p_toktype(p): + '''toktype : TOKEN + | LEFT + | RIGHT + | NONASSOC''' + p[0] = p[1] + + +def p_opttype(p): + '''opttype : '<' ID '>' + | empty''' + + +def p_idlist(p): + '''idlist : idlist optcomma tokenid + | tokenid''' + if len(p) == 2: + p[0] = [p[1]] + else: + p[0] = p[1] + p[1].append(p[3]) + + +def p_tokenid(p): + '''tokenid : ID + | ID NUMBER + | QLITERAL + | QLITERAL NUMBER''' + p[0] = p[1] + + +def p_optsemi(p): + '''optsemi : ';' + | empty''' + + +def p_optcomma(p): + '''optcomma : ',' + | empty''' + + +def p_definition_type(p): + '''definition : TYPE '<' ID '>' namelist optsemi''' + # type declarations are ignored + + +def p_namelist(p): + '''namelist : namelist optcomma ID + | ID''' + + +def p_definition_union(p): + '''definition : UNION CODE optsemi''' + # Union declarations are ignored + + +def p_rules(p): + '''rules : rules rule + | rule''' + if len(p) == 2: + rule = p[1] + else: + rule = p[2] + + # Print out a Python equivalent of this rule + + embedded = [] # Embedded actions (a mess) + embed_count = 0 + + rulename = rule[0] + rulecount = 1 + for r in rule[1]: + # r contains one of the rule possibilities + print("def p_%s_%d(p):" % (rulename, rulecount)) + prod = [] + prodcode = "" + for i in range(len(r)): + item = r[i] + if item[0] == '{': # A code block + if i == len(r) - 1: + prodcode = item + break + else: + # an embedded action + embed_name = "_embed%d_%s" % (embed_count, rulename) + prod.append(embed_name) + embedded.append((embed_name, item)) + embed_count += 1 + else: + prod.append(item) + print(" '''%s : %s'''" % (rulename, " ".join(prod))) + # Emit code + print_code(prodcode, 4) + print() + rulecount += 1 + + for e, code in embedded: + print("def p_%s(p):" % e) + print(" '''%s : '''" % e) + print_code(code, 4) + print() + + +def p_rule(p): + '''rule : ID ':' rulelist ';' ''' + p[0] = (p[1], [p[3]]) + + +def p_rule2(p): + '''rule : ID ':' rulelist morerules ';' ''' + p[4].insert(0, p[3]) + p[0] = (p[1], p[4]) + + +def p_rule_empty(p): + '''rule : ID ':' ';' ''' + p[0] = (p[1], [[]]) + + +def p_rule_empty2(p): + '''rule : ID ':' morerules ';' ''' + + p[3].insert(0, []) + p[0] = (p[1], p[3]) + + +def p_morerules(p): + '''morerules : morerules '|' rulelist + | '|' rulelist + | '|' ''' + + if len(p) == 2: + p[0] = [[]] + elif len(p) == 3: + p[0] = [p[2]] + else: + p[0] = p[1] + p[0].append(p[3]) + +# print("morerules", len(p), p[0]) + + +def p_rulelist(p): + '''rulelist : rulelist ruleitem + | ruleitem''' + + if len(p) == 2: + p[0] = [p[1]] + else: + p[0] = p[1] + p[1].append(p[2]) + + +def p_ruleitem(p): + '''ruleitem : ID + | QLITERAL + | CODE + | PREC''' + p[0] = p[1] + + +def p_empty(p): + '''empty : ''' + + +def p_error(p): + pass + +yacc.yacc(debug=0) + + +def print_code(code, indent): + if not emit_code: + return + codelines = code.splitlines() + for c in codelines: + print("%s# %s" % (" " * indent, c)) diff --git a/third_party/python/ply/example/yply/yply.py b/third_party/python/ply/example/yply/yply.py new file mode 100755 index 0000000000..e24616c831 --- /dev/null +++ b/third_party/python/ply/example/yply/yply.py @@ -0,0 +1,51 @@ +#!/usr/local/bin/python +# yply.py +# +# Author: David Beazley (dave@dabeaz.com) +# Date : October 2, 2006 +# +# Converts a UNIX-yacc specification file into a PLY-compatible +# specification. To use, simply do this: +# +# % python yply.py [-nocode] inputfile.y >myparser.py +# +# The output of this program is Python code. In the output, +# any C code in the original file is included, but is commented. +# If you use the -nocode option, then all of the C code in the +# original file is discarded. +# +# Disclaimer: This just an example I threw together in an afternoon. +# It might have some bugs. However, it worked when I tried it on +# a yacc-specified C++ parser containing 442 rules and 855 parsing +# states. +# + +import sys +sys.path.insert(0, "../..") + +import ylex +import yparse + +from ply import * + +if len(sys.argv) == 1: + print("usage : yply.py [-nocode] inputfile") + raise SystemExit + +if len(sys.argv) == 3: + if sys.argv[1] == '-nocode': + yparse.emit_code = 0 + else: + print("Unknown option '%s'" % sys.argv[1]) + raise SystemExit + filename = sys.argv[2] +else: + filename = sys.argv[1] + +yacc.parse(open(filename).read()) + +print(""" +if __name__ == '__main__': + from ply import * + yacc.yacc() +""") diff --git a/third_party/python/ply/ply.egg-info/PKG-INFO b/third_party/python/ply/ply.egg-info/PKG-INFO new file mode 100644 index 0000000000..6eedf42595 --- /dev/null +++ b/third_party/python/ply/ply.egg-info/PKG-INFO @@ -0,0 +1,22 @@ +Metadata-Version: 1.1 +Name: ply +Version: 3.10 +Summary: Python Lex & Yacc +Home-page: http://www.dabeaz.com/ply/ +Author: David Beazley +Author-email: dave@dabeaz.com +License: BSD +Description: + PLY is yet another implementation of lex and yacc for Python. Some notable + features include the fact that its implemented entirely in Python and it + uses LALR(1) parsing which is efficient and well suited for larger grammars. + + PLY provides most of the standard lex/yacc features including support for empty + productions, precedence rules, error recovery, and support for ambiguous grammars. + + PLY is extremely easy to use and provides very extensive error checking. + It is compatible with both Python 2 and Python 3. + +Platform: UNKNOWN +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 2 diff --git a/third_party/python/ply/ply.egg-info/SOURCES.txt b/third_party/python/ply/ply.egg-info/SOURCES.txt new file mode 100644 index 0000000000..2dff7dd29b --- /dev/null +++ b/third_party/python/ply/ply.egg-info/SOURCES.txt @@ -0,0 +1,172 @@ +ANNOUNCE +CHANGES +MANIFEST.in +README.md +TODO +setup.cfg +setup.py +doc/internal.html +doc/makedoc.py +doc/ply.html +example/README +example/cleanup.sh +example/BASIC/README +example/BASIC/basic.py +example/BASIC/basiclex.py +example/BASIC/basiclog.py +example/BASIC/basinterp.py +example/BASIC/basparse.py +example/BASIC/dim.bas +example/BASIC/func.bas +example/BASIC/gcd.bas +example/BASIC/gosub.bas +example/BASIC/hello.bas +example/BASIC/linear.bas +example/BASIC/maxsin.bas +example/BASIC/powers.bas +example/BASIC/rand.bas +example/BASIC/sales.bas +example/BASIC/sears.bas +example/BASIC/sqrt1.bas +example/BASIC/sqrt2.bas +example/GardenSnake/GardenSnake.py +example/GardenSnake/README +example/ansic/README +example/ansic/clex.py +example/ansic/cparse.py +example/calc/calc.py +example/calcdebug/calc.py +example/calceof/calc.py +example/classcalc/calc.py +example/closurecalc/calc.py +example/hedit/hedit.py +example/newclasscalc/calc.py +example/optcalc/README +example/optcalc/calc.py +example/unicalc/calc.py +example/yply/README +example/yply/ylex.py +example/yply/yparse.py +example/yply/yply.py +ply/__init__.py +ply/cpp.py +ply/ctokens.py +ply/lex.py +ply/yacc.py +ply/ygen.py +ply.egg-info/PKG-INFO +ply.egg-info/SOURCES.txt +ply.egg-info/dependency_links.txt +ply.egg-info/top_level.txt +test/README +test/calclex.py +test/cleanup.sh +test/lex_closure.py +test/lex_doc1.py +test/lex_dup1.py +test/lex_dup2.py +test/lex_dup3.py +test/lex_empty.py +test/lex_error1.py +test/lex_error2.py +test/lex_error3.py +test/lex_error4.py +test/lex_hedit.py +test/lex_ignore.py +test/lex_ignore2.py +test/lex_literal1.py +test/lex_literal2.py +test/lex_literal3.py +test/lex_many_tokens.py +test/lex_module.py +test/lex_module_import.py +test/lex_object.py +test/lex_opt_alias.py +test/lex_optimize.py +test/lex_optimize2.py +test/lex_optimize3.py +test/lex_re1.py +test/lex_re2.py +test/lex_re3.py +test/lex_rule1.py +test/lex_rule2.py +test/lex_rule3.py +test/lex_state1.py +test/lex_state2.py +test/lex_state3.py +test/lex_state4.py +test/lex_state5.py +test/lex_state_noerror.py +test/lex_state_norule.py +test/lex_state_try.py +test/lex_token1.py +test/lex_token2.py +test/lex_token3.py +test/lex_token4.py +test/lex_token5.py +test/lex_token_dup.py +test/testlex.py +test/testyacc.py +test/yacc_badargs.py +test/yacc_badid.py +test/yacc_badprec.py +test/yacc_badprec2.py +test/yacc_badprec3.py +test/yacc_badrule.py +test/yacc_badtok.py +test/yacc_dup.py +test/yacc_error1.py +test/yacc_error2.py +test/yacc_error3.py +test/yacc_error4.py +test/yacc_error5.py +test/yacc_error6.py +test/yacc_error7.py +test/yacc_inf.py +test/yacc_literal.py +test/yacc_misplaced.py +test/yacc_missing1.py +test/yacc_nested.py +test/yacc_nodoc.py +test/yacc_noerror.py +test/yacc_nop.py +test/yacc_notfunc.py +test/yacc_notok.py +test/yacc_prec1.py +test/yacc_rr.py +test/yacc_rr_unused.py +test/yacc_simple.py +test/yacc_sr.py +test/yacc_term1.py +test/yacc_unicode_literals.py +test/yacc_unused.py +test/yacc_unused_rule.py +test/yacc_uprec.py +test/yacc_uprec2.py +test/pkg_test1/__init__.py +test/pkg_test1/parsing/__init__.py +test/pkg_test1/parsing/calclex.py +test/pkg_test1/parsing/calcparse.py +test/pkg_test2/__init__.py +test/pkg_test2/parsing/__init__.py +test/pkg_test2/parsing/calclex.py +test/pkg_test2/parsing/calcparse.py +test/pkg_test3/__init__.py +test/pkg_test3/generated/__init__.py +test/pkg_test3/parsing/__init__.py +test/pkg_test3/parsing/calclex.py +test/pkg_test3/parsing/calcparse.py +test/pkg_test4/__init__.py +test/pkg_test4/parsing/__init__.py +test/pkg_test4/parsing/calclex.py +test/pkg_test4/parsing/calcparse.py +test/pkg_test5/__init__.py +test/pkg_test5/parsing/__init__.py +test/pkg_test5/parsing/calclex.py +test/pkg_test5/parsing/calcparse.py +test/pkg_test6/__init__.py +test/pkg_test6/parsing/__init__.py +test/pkg_test6/parsing/calclex.py +test/pkg_test6/parsing/calcparse.py +test/pkg_test6/parsing/expression.py +test/pkg_test6/parsing/statement.py
\ No newline at end of file diff --git a/third_party/python/ply/ply.egg-info/dependency_links.txt b/third_party/python/ply/ply.egg-info/dependency_links.txt new file mode 100644 index 0000000000..8b13789179 --- /dev/null +++ b/third_party/python/ply/ply.egg-info/dependency_links.txt @@ -0,0 +1 @@ + diff --git a/third_party/python/ply/ply.egg-info/top_level.txt b/third_party/python/ply/ply.egg-info/top_level.txt new file mode 100644 index 0000000000..90412f0683 --- /dev/null +++ b/third_party/python/ply/ply.egg-info/top_level.txt @@ -0,0 +1 @@ +ply diff --git a/third_party/python/ply/ply/__init__.py b/third_party/python/ply/ply/__init__.py new file mode 100644 index 0000000000..6e53cddcf6 --- /dev/null +++ b/third_party/python/ply/ply/__init__.py @@ -0,0 +1,5 @@ +# PLY package +# Author: David Beazley (dave@dabeaz.com) + +__version__ = '3.9' +__all__ = ['lex','yacc'] diff --git a/third_party/python/ply/ply/cpp.py b/third_party/python/ply/ply/cpp.py new file mode 100644 index 0000000000..b6bfc69614 --- /dev/null +++ b/third_party/python/ply/ply/cpp.py @@ -0,0 +1,918 @@ +# ----------------------------------------------------------------------------- +# cpp.py +# +# Author: David Beazley (http://www.dabeaz.com) +# Copyright (C) 2007 +# All rights reserved +# +# This module implements an ANSI-C style lexical preprocessor for PLY. +# ----------------------------------------------------------------------------- +from __future__ import generators + +import sys + +# Some Python 3 compatibility shims +if sys.version_info.major < 3: + STRING_TYPES = (str, unicode) +else: + STRING_TYPES = str + xrange = range + +# ----------------------------------------------------------------------------- +# Default preprocessor lexer definitions. These tokens are enough to get +# a basic preprocessor working. Other modules may import these if they want +# ----------------------------------------------------------------------------- + +tokens = ( + 'CPP_ID','CPP_INTEGER', 'CPP_FLOAT', 'CPP_STRING', 'CPP_CHAR', 'CPP_WS', 'CPP_COMMENT1', 'CPP_COMMENT2', 'CPP_POUND','CPP_DPOUND' +) + +literals = "+-*/%|&~^<>=!?()[]{}.,;:\\\'\"" + +# Whitespace +def t_CPP_WS(t): + r'\s+' + t.lexer.lineno += t.value.count("\n") + return t + +t_CPP_POUND = r'\#' +t_CPP_DPOUND = r'\#\#' + +# Identifier +t_CPP_ID = r'[A-Za-z_][\w_]*' + +# Integer literal +def CPP_INTEGER(t): + r'(((((0x)|(0X))[0-9a-fA-F]+)|(\d+))([uU][lL]|[lL][uU]|[uU]|[lL])?)' + return t + +t_CPP_INTEGER = CPP_INTEGER + +# Floating literal +t_CPP_FLOAT = r'((\d+)(\.\d+)(e(\+|-)?(\d+))? | (\d+)e(\+|-)?(\d+))([lL]|[fF])?' + +# String literal +def t_CPP_STRING(t): + r'\"([^\\\n]|(\\(.|\n)))*?\"' + t.lexer.lineno += t.value.count("\n") + return t + +# Character constant 'c' or L'c' +def t_CPP_CHAR(t): + r'(L)?\'([^\\\n]|(\\(.|\n)))*?\'' + t.lexer.lineno += t.value.count("\n") + return t + +# Comment +def t_CPP_COMMENT1(t): + r'(/\*(.|\n)*?\*/)' + ncr = t.value.count("\n") + t.lexer.lineno += ncr + # replace with one space or a number of '\n' + t.type = 'CPP_WS'; t.value = '\n' * ncr if ncr else ' ' + return t + +# Line comment +def t_CPP_COMMENT2(t): + r'(//.*?(\n|$))' + # replace with '/n' + t.type = 'CPP_WS'; t.value = '\n' + return t + +def t_error(t): + t.type = t.value[0] + t.value = t.value[0] + t.lexer.skip(1) + return t + +import re +import copy +import time +import os.path + +# ----------------------------------------------------------------------------- +# trigraph() +# +# Given an input string, this function replaces all trigraph sequences. +# The following mapping is used: +# +# ??= # +# ??/ \ +# ??' ^ +# ??( [ +# ??) ] +# ??! | +# ??< { +# ??> } +# ??- ~ +# ----------------------------------------------------------------------------- + +_trigraph_pat = re.compile(r'''\?\?[=/\'\(\)\!<>\-]''') +_trigraph_rep = { + '=':'#', + '/':'\\', + "'":'^', + '(':'[', + ')':']', + '!':'|', + '<':'{', + '>':'}', + '-':'~' +} + +def trigraph(input): + return _trigraph_pat.sub(lambda g: _trigraph_rep[g.group()[-1]],input) + +# ------------------------------------------------------------------ +# Macro object +# +# This object holds information about preprocessor macros +# +# .name - Macro name (string) +# .value - Macro value (a list of tokens) +# .arglist - List of argument names +# .variadic - Boolean indicating whether or not variadic macro +# .vararg - Name of the variadic parameter +# +# When a macro is created, the macro replacement token sequence is +# pre-scanned and used to create patch lists that are later used +# during macro expansion +# ------------------------------------------------------------------ + +class Macro(object): + def __init__(self,name,value,arglist=None,variadic=False): + self.name = name + self.value = value + self.arglist = arglist + self.variadic = variadic + if variadic: + self.vararg = arglist[-1] + self.source = None + +# ------------------------------------------------------------------ +# Preprocessor object +# +# Object representing a preprocessor. Contains macro definitions, +# include directories, and other information +# ------------------------------------------------------------------ + +class Preprocessor(object): + def __init__(self,lexer=None): + if lexer is None: + lexer = lex.lexer + self.lexer = lexer + self.macros = { } + self.path = [] + self.temp_path = [] + + # Probe the lexer for selected tokens + self.lexprobe() + + tm = time.localtime() + self.define("__DATE__ \"%s\"" % time.strftime("%b %d %Y",tm)) + self.define("__TIME__ \"%s\"" % time.strftime("%H:%M:%S",tm)) + self.parser = None + + # ----------------------------------------------------------------------------- + # tokenize() + # + # Utility function. Given a string of text, tokenize into a list of tokens + # ----------------------------------------------------------------------------- + + def tokenize(self,text): + tokens = [] + self.lexer.input(text) + while True: + tok = self.lexer.token() + if not tok: break + tokens.append(tok) + return tokens + + # --------------------------------------------------------------------- + # error() + # + # Report a preprocessor error/warning of some kind + # ---------------------------------------------------------------------- + + def error(self,file,line,msg): + print("%s:%d %s" % (file,line,msg)) + + # ---------------------------------------------------------------------- + # lexprobe() + # + # This method probes the preprocessor lexer object to discover + # the token types of symbols that are important to the preprocessor. + # If this works right, the preprocessor will simply "work" + # with any suitable lexer regardless of how tokens have been named. + # ---------------------------------------------------------------------- + + def lexprobe(self): + + # Determine the token type for identifiers + self.lexer.input("identifier") + tok = self.lexer.token() + if not tok or tok.value != "identifier": + print("Couldn't determine identifier type") + else: + self.t_ID = tok.type + + # Determine the token type for integers + self.lexer.input("12345") + tok = self.lexer.token() + if not tok or int(tok.value) != 12345: + print("Couldn't determine integer type") + else: + self.t_INTEGER = tok.type + self.t_INTEGER_TYPE = type(tok.value) + + # Determine the token type for strings enclosed in double quotes + self.lexer.input("\"filename\"") + tok = self.lexer.token() + if not tok or tok.value != "\"filename\"": + print("Couldn't determine string type") + else: + self.t_STRING = tok.type + + # Determine the token type for whitespace--if any + self.lexer.input(" ") + tok = self.lexer.token() + if not tok or tok.value != " ": + self.t_SPACE = None + else: + self.t_SPACE = tok.type + + # Determine the token type for newlines + self.lexer.input("\n") + tok = self.lexer.token() + if not tok or tok.value != "\n": + self.t_NEWLINE = None + print("Couldn't determine token for newlines") + else: + self.t_NEWLINE = tok.type + + self.t_WS = (self.t_SPACE, self.t_NEWLINE) + + # Check for other characters used by the preprocessor + chars = [ '<','>','#','##','\\','(',')',',','.'] + for c in chars: + self.lexer.input(c) + tok = self.lexer.token() + if not tok or tok.value != c: + print("Unable to lex '%s' required for preprocessor" % c) + + # ---------------------------------------------------------------------- + # add_path() + # + # Adds a search path to the preprocessor. + # ---------------------------------------------------------------------- + + def add_path(self,path): + self.path.append(path) + + # ---------------------------------------------------------------------- + # group_lines() + # + # Given an input string, this function splits it into lines. Trailing whitespace + # is removed. Any line ending with \ is grouped with the next line. This + # function forms the lowest level of the preprocessor---grouping into text into + # a line-by-line format. + # ---------------------------------------------------------------------- + + def group_lines(self,input): + lex = self.lexer.clone() + lines = [x.rstrip() for x in input.splitlines()] + for i in xrange(len(lines)): + j = i+1 + while lines[i].endswith('\\') and (j < len(lines)): + lines[i] = lines[i][:-1]+lines[j] + lines[j] = "" + j += 1 + + input = "\n".join(lines) + lex.input(input) + lex.lineno = 1 + + current_line = [] + while True: + tok = lex.token() + if not tok: + break + current_line.append(tok) + if tok.type in self.t_WS and '\n' in tok.value: + yield current_line + current_line = [] + + if current_line: + yield current_line + + # ---------------------------------------------------------------------- + # tokenstrip() + # + # Remove leading/trailing whitespace tokens from a token list + # ---------------------------------------------------------------------- + + def tokenstrip(self,tokens): + i = 0 + while i < len(tokens) and tokens[i].type in self.t_WS: + i += 1 + del tokens[:i] + i = len(tokens)-1 + while i >= 0 and tokens[i].type in self.t_WS: + i -= 1 + del tokens[i+1:] + return tokens + + + # ---------------------------------------------------------------------- + # collect_args() + # + # Collects comma separated arguments from a list of tokens. The arguments + # must be enclosed in parenthesis. Returns a tuple (tokencount,args,positions) + # where tokencount is the number of tokens consumed, args is a list of arguments, + # and positions is a list of integers containing the starting index of each + # argument. Each argument is represented by a list of tokens. + # + # When collecting arguments, leading and trailing whitespace is removed + # from each argument. + # + # This function properly handles nested parenthesis and commas---these do not + # define new arguments. + # ---------------------------------------------------------------------- + + def collect_args(self,tokenlist): + args = [] + positions = [] + current_arg = [] + nesting = 1 + tokenlen = len(tokenlist) + + # Search for the opening '('. + i = 0 + while (i < tokenlen) and (tokenlist[i].type in self.t_WS): + i += 1 + + if (i < tokenlen) and (tokenlist[i].value == '('): + positions.append(i+1) + else: + self.error(self.source,tokenlist[0].lineno,"Missing '(' in macro arguments") + return 0, [], [] + + i += 1 + + while i < tokenlen: + t = tokenlist[i] + if t.value == '(': + current_arg.append(t) + nesting += 1 + elif t.value == ')': + nesting -= 1 + if nesting == 0: + if current_arg: + args.append(self.tokenstrip(current_arg)) + positions.append(i) + return i+1,args,positions + current_arg.append(t) + elif t.value == ',' and nesting == 1: + args.append(self.tokenstrip(current_arg)) + positions.append(i+1) + current_arg = [] + else: + current_arg.append(t) + i += 1 + + # Missing end argument + self.error(self.source,tokenlist[-1].lineno,"Missing ')' in macro arguments") + return 0, [],[] + + # ---------------------------------------------------------------------- + # macro_prescan() + # + # Examine the macro value (token sequence) and identify patch points + # This is used to speed up macro expansion later on---we'll know + # right away where to apply patches to the value to form the expansion + # ---------------------------------------------------------------------- + + def macro_prescan(self,macro): + macro.patch = [] # Standard macro arguments + macro.str_patch = [] # String conversion expansion + macro.var_comma_patch = [] # Variadic macro comma patch + i = 0 + while i < len(macro.value): + if macro.value[i].type == self.t_ID and macro.value[i].value in macro.arglist: + argnum = macro.arglist.index(macro.value[i].value) + # Conversion of argument to a string + if i > 0 and macro.value[i-1].value == '#': + macro.value[i] = copy.copy(macro.value[i]) + macro.value[i].type = self.t_STRING + del macro.value[i-1] + macro.str_patch.append((argnum,i-1)) + continue + # Concatenation + elif (i > 0 and macro.value[i-1].value == '##'): + macro.patch.append(('c',argnum,i-1)) + del macro.value[i-1] + continue + elif ((i+1) < len(macro.value) and macro.value[i+1].value == '##'): + macro.patch.append(('c',argnum,i)) + i += 1 + continue + # Standard expansion + else: + macro.patch.append(('e',argnum,i)) + elif macro.value[i].value == '##': + if macro.variadic and (i > 0) and (macro.value[i-1].value == ',') and \ + ((i+1) < len(macro.value)) and (macro.value[i+1].type == self.t_ID) and \ + (macro.value[i+1].value == macro.vararg): + macro.var_comma_patch.append(i-1) + i += 1 + macro.patch.sort(key=lambda x: x[2],reverse=True) + + # ---------------------------------------------------------------------- + # macro_expand_args() + # + # Given a Macro and list of arguments (each a token list), this method + # returns an expanded version of a macro. The return value is a token sequence + # representing the replacement macro tokens + # ---------------------------------------------------------------------- + + def macro_expand_args(self,macro,args): + # Make a copy of the macro token sequence + rep = [copy.copy(_x) for _x in macro.value] + + # Make string expansion patches. These do not alter the length of the replacement sequence + + str_expansion = {} + for argnum, i in macro.str_patch: + if argnum not in str_expansion: + str_expansion[argnum] = ('"%s"' % "".join([x.value for x in args[argnum]])).replace("\\","\\\\") + rep[i] = copy.copy(rep[i]) + rep[i].value = str_expansion[argnum] + + # Make the variadic macro comma patch. If the variadic macro argument is empty, we get rid + comma_patch = False + if macro.variadic and not args[-1]: + for i in macro.var_comma_patch: + rep[i] = None + comma_patch = True + + # Make all other patches. The order of these matters. It is assumed that the patch list + # has been sorted in reverse order of patch location since replacements will cause the + # size of the replacement sequence to expand from the patch point. + + expanded = { } + for ptype, argnum, i in macro.patch: + # Concatenation. Argument is left unexpanded + if ptype == 'c': + rep[i:i+1] = args[argnum] + # Normal expansion. Argument is macro expanded first + elif ptype == 'e': + if argnum not in expanded: + expanded[argnum] = self.expand_macros(args[argnum]) + rep[i:i+1] = expanded[argnum] + + # Get rid of removed comma if necessary + if comma_patch: + rep = [_i for _i in rep if _i] + + return rep + + + # ---------------------------------------------------------------------- + # expand_macros() + # + # Given a list of tokens, this function performs macro expansion. + # The expanded argument is a dictionary that contains macros already + # expanded. This is used to prevent infinite recursion. + # ---------------------------------------------------------------------- + + def expand_macros(self,tokens,expanded=None): + if expanded is None: + expanded = {} + i = 0 + while i < len(tokens): + t = tokens[i] + if t.type == self.t_ID: + if t.value in self.macros and t.value not in expanded: + # Yes, we found a macro match + expanded[t.value] = True + + m = self.macros[t.value] + if not m.arglist: + # A simple macro + ex = self.expand_macros([copy.copy(_x) for _x in m.value],expanded) + for e in ex: + e.lineno = t.lineno + tokens[i:i+1] = ex + i += len(ex) + else: + # A macro with arguments + j = i + 1 + while j < len(tokens) and tokens[j].type in self.t_WS: + j += 1 + if tokens[j].value == '(': + tokcount,args,positions = self.collect_args(tokens[j:]) + if not m.variadic and len(args) != len(m.arglist): + self.error(self.source,t.lineno,"Macro %s requires %d arguments" % (t.value,len(m.arglist))) + i = j + tokcount + elif m.variadic and len(args) < len(m.arglist)-1: + if len(m.arglist) > 2: + self.error(self.source,t.lineno,"Macro %s must have at least %d arguments" % (t.value, len(m.arglist)-1)) + else: + self.error(self.source,t.lineno,"Macro %s must have at least %d argument" % (t.value, len(m.arglist)-1)) + i = j + tokcount + else: + if m.variadic: + if len(args) == len(m.arglist)-1: + args.append([]) + else: + args[len(m.arglist)-1] = tokens[j+positions[len(m.arglist)-1]:j+tokcount-1] + del args[len(m.arglist):] + + # Get macro replacement text + rep = self.macro_expand_args(m,args) + rep = self.expand_macros(rep,expanded) + for r in rep: + r.lineno = t.lineno + tokens[i:j+tokcount] = rep + i += len(rep) + del expanded[t.value] + continue + elif t.value == '__LINE__': + t.type = self.t_INTEGER + t.value = self.t_INTEGER_TYPE(t.lineno) + + i += 1 + return tokens + + # ---------------------------------------------------------------------- + # evalexpr() + # + # Evaluate an expression token sequence for the purposes of evaluating + # integral expressions. + # ---------------------------------------------------------------------- + + def evalexpr(self,tokens): + # tokens = tokenize(line) + # Search for defined macros + i = 0 + while i < len(tokens): + if tokens[i].type == self.t_ID and tokens[i].value == 'defined': + j = i + 1 + needparen = False + result = "0L" + while j < len(tokens): + if tokens[j].type in self.t_WS: + j += 1 + continue + elif tokens[j].type == self.t_ID: + if tokens[j].value in self.macros: + result = "1L" + else: + result = "0L" + if not needparen: break + elif tokens[j].value == '(': + needparen = True + elif tokens[j].value == ')': + break + else: + self.error(self.source,tokens[i].lineno,"Malformed defined()") + j += 1 + tokens[i].type = self.t_INTEGER + tokens[i].value = self.t_INTEGER_TYPE(result) + del tokens[i+1:j+1] + i += 1 + tokens = self.expand_macros(tokens) + for i,t in enumerate(tokens): + if t.type == self.t_ID: + tokens[i] = copy.copy(t) + tokens[i].type = self.t_INTEGER + tokens[i].value = self.t_INTEGER_TYPE("0L") + elif t.type == self.t_INTEGER: + tokens[i] = copy.copy(t) + # Strip off any trailing suffixes + tokens[i].value = str(tokens[i].value) + while tokens[i].value[-1] not in "0123456789abcdefABCDEF": + tokens[i].value = tokens[i].value[:-1] + + expr = "".join([str(x.value) for x in tokens]) + expr = expr.replace("&&"," and ") + expr = expr.replace("||"," or ") + expr = expr.replace("!"," not ") + try: + result = eval(expr) + except Exception: + self.error(self.source,tokens[0].lineno,"Couldn't evaluate expression") + result = 0 + return result + + # ---------------------------------------------------------------------- + # parsegen() + # + # Parse an input string/ + # ---------------------------------------------------------------------- + def parsegen(self,input,source=None): + + # Replace trigraph sequences + t = trigraph(input) + lines = self.group_lines(t) + + if not source: + source = "" + + self.define("__FILE__ \"%s\"" % source) + + self.source = source + chunk = [] + enable = True + iftrigger = False + ifstack = [] + + for x in lines: + for i,tok in enumerate(x): + if tok.type not in self.t_WS: break + if tok.value == '#': + # Preprocessor directive + + # insert necessary whitespace instead of eaten tokens + for tok in x: + if tok.type in self.t_WS and '\n' in tok.value: + chunk.append(tok) + + dirtokens = self.tokenstrip(x[i+1:]) + if dirtokens: + name = dirtokens[0].value + args = self.tokenstrip(dirtokens[1:]) + else: + name = "" + args = [] + + if name == 'define': + if enable: + for tok in self.expand_macros(chunk): + yield tok + chunk = [] + self.define(args) + elif name == 'include': + if enable: + for tok in self.expand_macros(chunk): + yield tok + chunk = [] + oldfile = self.macros['__FILE__'] + for tok in self.include(args): + yield tok + self.macros['__FILE__'] = oldfile + self.source = source + elif name == 'undef': + if enable: + for tok in self.expand_macros(chunk): + yield tok + chunk = [] + self.undef(args) + elif name == 'ifdef': + ifstack.append((enable,iftrigger)) + if enable: + if not args[0].value in self.macros: + enable = False + iftrigger = False + else: + iftrigger = True + elif name == 'ifndef': + ifstack.append((enable,iftrigger)) + if enable: + if args[0].value in self.macros: + enable = False + iftrigger = False + else: + iftrigger = True + elif name == 'if': + ifstack.append((enable,iftrigger)) + if enable: + result = self.evalexpr(args) + if not result: + enable = False + iftrigger = False + else: + iftrigger = True + elif name == 'elif': + if ifstack: + if ifstack[-1][0]: # We only pay attention if outer "if" allows this + if enable: # If already true, we flip enable False + enable = False + elif not iftrigger: # If False, but not triggered yet, we'll check expression + result = self.evalexpr(args) + if result: + enable = True + iftrigger = True + else: + self.error(self.source,dirtokens[0].lineno,"Misplaced #elif") + + elif name == 'else': + if ifstack: + if ifstack[-1][0]: + if enable: + enable = False + elif not iftrigger: + enable = True + iftrigger = True + else: + self.error(self.source,dirtokens[0].lineno,"Misplaced #else") + + elif name == 'endif': + if ifstack: + enable,iftrigger = ifstack.pop() + else: + self.error(self.source,dirtokens[0].lineno,"Misplaced #endif") + else: + # Unknown preprocessor directive + pass + + else: + # Normal text + if enable: + chunk.extend(x) + + for tok in self.expand_macros(chunk): + yield tok + chunk = [] + + # ---------------------------------------------------------------------- + # include() + # + # Implementation of file-inclusion + # ---------------------------------------------------------------------- + + def include(self,tokens): + # Try to extract the filename and then process an include file + if not tokens: + return + if tokens: + if tokens[0].value != '<' and tokens[0].type != self.t_STRING: + tokens = self.expand_macros(tokens) + + if tokens[0].value == '<': + # Include <...> + i = 1 + while i < len(tokens): + if tokens[i].value == '>': + break + i += 1 + else: + print("Malformed #include <...>") + return + filename = "".join([x.value for x in tokens[1:i]]) + path = self.path + [""] + self.temp_path + elif tokens[0].type == self.t_STRING: + filename = tokens[0].value[1:-1] + path = self.temp_path + [""] + self.path + else: + print("Malformed #include statement") + return + for p in path: + iname = os.path.join(p,filename) + try: + data = open(iname,"r").read() + dname = os.path.dirname(iname) + if dname: + self.temp_path.insert(0,dname) + for tok in self.parsegen(data,filename): + yield tok + if dname: + del self.temp_path[0] + break + except IOError: + pass + else: + print("Couldn't find '%s'" % filename) + + # ---------------------------------------------------------------------- + # define() + # + # Define a new macro + # ---------------------------------------------------------------------- + + def define(self,tokens): + if isinstance(tokens,STRING_TYPES): + tokens = self.tokenize(tokens) + + linetok = tokens + try: + name = linetok[0] + if len(linetok) > 1: + mtype = linetok[1] + else: + mtype = None + if not mtype: + m = Macro(name.value,[]) + self.macros[name.value] = m + elif mtype.type in self.t_WS: + # A normal macro + m = Macro(name.value,self.tokenstrip(linetok[2:])) + self.macros[name.value] = m + elif mtype.value == '(': + # A macro with arguments + tokcount, args, positions = self.collect_args(linetok[1:]) + variadic = False + for a in args: + if variadic: + print("No more arguments may follow a variadic argument") + break + astr = "".join([str(_i.value) for _i in a]) + if astr == "...": + variadic = True + a[0].type = self.t_ID + a[0].value = '__VA_ARGS__' + variadic = True + del a[1:] + continue + elif astr[-3:] == "..." and a[0].type == self.t_ID: + variadic = True + del a[1:] + # If, for some reason, "." is part of the identifier, strip off the name for the purposes + # of macro expansion + if a[0].value[-3:] == '...': + a[0].value = a[0].value[:-3] + continue + if len(a) > 1 or a[0].type != self.t_ID: + print("Invalid macro argument") + break + else: + mvalue = self.tokenstrip(linetok[1+tokcount:]) + i = 0 + while i < len(mvalue): + if i+1 < len(mvalue): + if mvalue[i].type in self.t_WS and mvalue[i+1].value == '##': + del mvalue[i] + continue + elif mvalue[i].value == '##' and mvalue[i+1].type in self.t_WS: + del mvalue[i+1] + i += 1 + m = Macro(name.value,mvalue,[x[0].value for x in args],variadic) + self.macro_prescan(m) + self.macros[name.value] = m + else: + print("Bad macro definition") + except LookupError: + print("Bad macro definition") + + # ---------------------------------------------------------------------- + # undef() + # + # Undefine a macro + # ---------------------------------------------------------------------- + + def undef(self,tokens): + id = tokens[0].value + try: + del self.macros[id] + except LookupError: + pass + + # ---------------------------------------------------------------------- + # parse() + # + # Parse input text. + # ---------------------------------------------------------------------- + def parse(self,input,source=None,ignore={}): + self.ignore = ignore + self.parser = self.parsegen(input,source) + + # ---------------------------------------------------------------------- + # token() + # + # Method to return individual tokens + # ---------------------------------------------------------------------- + def token(self): + try: + while True: + tok = next(self.parser) + if tok.type not in self.ignore: return tok + except StopIteration: + self.parser = None + return None + +if __name__ == '__main__': + import ply.lex as lex + lexer = lex.lex() + + # Run a preprocessor + import sys + f = open(sys.argv[1]) + input = f.read() + + p = Preprocessor(lexer) + p.parse(input,sys.argv[1]) + while True: + tok = p.token() + if not tok: break + print(p.source, tok) + + + + + + + + + + + diff --git a/third_party/python/ply/ply/ctokens.py b/third_party/python/ply/ply/ctokens.py new file mode 100644 index 0000000000..f6f6952d60 --- /dev/null +++ b/third_party/python/ply/ply/ctokens.py @@ -0,0 +1,133 @@ +# ---------------------------------------------------------------------- +# ctokens.py +# +# Token specifications for symbols in ANSI C and C++. This file is +# meant to be used as a library in other tokenizers. +# ---------------------------------------------------------------------- + +# Reserved words + +tokens = [ + # Literals (identifier, integer constant, float constant, string constant, char const) + 'ID', 'TYPEID', 'INTEGER', 'FLOAT', 'STRING', 'CHARACTER', + + # Operators (+,-,*,/,%,|,&,~,^,<<,>>, ||, &&, !, <, <=, >, >=, ==, !=) + 'PLUS', 'MINUS', 'TIMES', 'DIVIDE', 'MODULO', + 'OR', 'AND', 'NOT', 'XOR', 'LSHIFT', 'RSHIFT', + 'LOR', 'LAND', 'LNOT', + 'LT', 'LE', 'GT', 'GE', 'EQ', 'NE', + + # Assignment (=, *=, /=, %=, +=, -=, <<=, >>=, &=, ^=, |=) + 'EQUALS', 'TIMESEQUAL', 'DIVEQUAL', 'MODEQUAL', 'PLUSEQUAL', 'MINUSEQUAL', + 'LSHIFTEQUAL','RSHIFTEQUAL', 'ANDEQUAL', 'XOREQUAL', 'OREQUAL', + + # Increment/decrement (++,--) + 'INCREMENT', 'DECREMENT', + + # Structure dereference (->) + 'ARROW', + + # Ternary operator (?) + 'TERNARY', + + # Delimeters ( ) [ ] { } , . ; : + 'LPAREN', 'RPAREN', + 'LBRACKET', 'RBRACKET', + 'LBRACE', 'RBRACE', + 'COMMA', 'PERIOD', 'SEMI', 'COLON', + + # Ellipsis (...) + 'ELLIPSIS', +] + +# Operators +t_PLUS = r'\+' +t_MINUS = r'-' +t_TIMES = r'\*' +t_DIVIDE = r'/' +t_MODULO = r'%' +t_OR = r'\|' +t_AND = r'&' +t_NOT = r'~' +t_XOR = r'\^' +t_LSHIFT = r'<<' +t_RSHIFT = r'>>' +t_LOR = r'\|\|' +t_LAND = r'&&' +t_LNOT = r'!' +t_LT = r'<' +t_GT = r'>' +t_LE = r'<=' +t_GE = r'>=' +t_EQ = r'==' +t_NE = r'!=' + +# Assignment operators + +t_EQUALS = r'=' +t_TIMESEQUAL = r'\*=' +t_DIVEQUAL = r'/=' +t_MODEQUAL = r'%=' +t_PLUSEQUAL = r'\+=' +t_MINUSEQUAL = r'-=' +t_LSHIFTEQUAL = r'<<=' +t_RSHIFTEQUAL = r'>>=' +t_ANDEQUAL = r'&=' +t_OREQUAL = r'\|=' +t_XOREQUAL = r'\^=' + +# Increment/decrement +t_INCREMENT = r'\+\+' +t_DECREMENT = r'--' + +# -> +t_ARROW = r'->' + +# ? +t_TERNARY = r'\?' + +# Delimeters +t_LPAREN = r'\(' +t_RPAREN = r'\)' +t_LBRACKET = r'\[' +t_RBRACKET = r'\]' +t_LBRACE = r'\{' +t_RBRACE = r'\}' +t_COMMA = r',' +t_PERIOD = r'\.' +t_SEMI = r';' +t_COLON = r':' +t_ELLIPSIS = r'\.\.\.' + +# Identifiers +t_ID = r'[A-Za-z_][A-Za-z0-9_]*' + +# Integer literal +t_INTEGER = r'\d+([uU]|[lL]|[uU][lL]|[lL][uU])?' + +# Floating literal +t_FLOAT = r'((\d+)(\.\d+)(e(\+|-)?(\d+))? | (\d+)e(\+|-)?(\d+))([lL]|[fF])?' + +# String literal +t_STRING = r'\"([^\\\n]|(\\.))*?\"' + +# Character constant 'c' or L'c' +t_CHARACTER = r'(L)?\'([^\\\n]|(\\.))*?\'' + +# Comment (C-Style) +def t_COMMENT(t): + r'/\*(.|\n)*?\*/' + t.lexer.lineno += t.value.count('\n') + return t + +# Comment (C++-Style) +def t_CPPCOMMENT(t): + r'//.*\n' + t.lexer.lineno += 1 + return t + + + + + + diff --git a/third_party/python/ply/ply/lex.py b/third_party/python/ply/ply/lex.py new file mode 100644 index 0000000000..3e240d1aa2 --- /dev/null +++ b/third_party/python/ply/ply/lex.py @@ -0,0 +1,1100 @@ +# ----------------------------------------------------------------------------- +# ply: lex.py +# +# Copyright (C) 2001-2017 +# David M. Beazley (Dabeaz LLC) +# All rights reserved. +# +# Redistribution and use in source and binary forms, with or without +# modification, are permitted provided that the following conditions are +# met: +# +# * Redistributions of source code must retain the above copyright notice, +# this list of conditions and the following disclaimer. +# * Redistributions in binary form must reproduce the above copyright notice, +# this list of conditions and the following disclaimer in the documentation +# and/or other materials provided with the distribution. +# * Neither the name of the David Beazley or Dabeaz LLC may be used to +# endorse or promote products derived from this software without +# specific prior written permission. +# +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +# ----------------------------------------------------------------------------- + +__version__ = '3.10' +__tabversion__ = '3.10' + +import re +import sys +import types +import copy +import os +import inspect + +# This tuple contains known string types +try: + # Python 2.6 + StringTypes = (types.StringType, types.UnicodeType) +except AttributeError: + # Python 3.0 + StringTypes = (str, bytes) + +# This regular expression is used to match valid token names +_is_identifier = re.compile(r'^[a-zA-Z0-9_]+$') + +# Exception thrown when invalid token encountered and no default error +# handler is defined. +class LexError(Exception): + def __init__(self, message, s): + self.args = (message,) + self.text = s + + +# Token class. This class is used to represent the tokens produced. +class LexToken(object): + def __str__(self): + return 'LexToken(%s,%r,%d,%d)' % (self.type, self.value, self.lineno, self.lexpos) + + def __repr__(self): + return str(self) + + +# This object is a stand-in for a logging object created by the +# logging module. + +class PlyLogger(object): + def __init__(self, f): + self.f = f + + def critical(self, msg, *args, **kwargs): + self.f.write((msg % args) + '\n') + + def warning(self, msg, *args, **kwargs): + self.f.write('WARNING: ' + (msg % args) + '\n') + + def error(self, msg, *args, **kwargs): + self.f.write('ERROR: ' + (msg % args) + '\n') + + info = critical + debug = critical + + +# Null logger is used when no output is generated. Does nothing. +class NullLogger(object): + def __getattribute__(self, name): + return self + + def __call__(self, *args, **kwargs): + return self + + +# ----------------------------------------------------------------------------- +# === Lexing Engine === +# +# The following Lexer class implements the lexer runtime. There are only +# a few public methods and attributes: +# +# input() - Store a new string in the lexer +# token() - Get the next token +# clone() - Clone the lexer +# +# lineno - Current line number +# lexpos - Current position in the input string +# ----------------------------------------------------------------------------- + +class Lexer: + def __init__(self): + self.lexre = None # Master regular expression. This is a list of + # tuples (re, findex) where re is a compiled + # regular expression and findex is a list + # mapping regex group numbers to rules + self.lexretext = None # Current regular expression strings + self.lexstatere = {} # Dictionary mapping lexer states to master regexs + self.lexstateretext = {} # Dictionary mapping lexer states to regex strings + self.lexstaterenames = {} # Dictionary mapping lexer states to symbol names + self.lexstate = 'INITIAL' # Current lexer state + self.lexstatestack = [] # Stack of lexer states + self.lexstateinfo = None # State information + self.lexstateignore = {} # Dictionary of ignored characters for each state + self.lexstateerrorf = {} # Dictionary of error functions for each state + self.lexstateeoff = {} # Dictionary of eof functions for each state + self.lexreflags = 0 # Optional re compile flags + self.lexdata = None # Actual input data (as a string) + self.lexpos = 0 # Current position in input text + self.lexlen = 0 # Length of the input text + self.lexerrorf = None # Error rule (if any) + self.lexeoff = None # EOF rule (if any) + self.lextokens = None # List of valid tokens + self.lexignore = '' # Ignored characters + self.lexliterals = '' # Literal characters that can be passed through + self.lexmodule = None # Module + self.lineno = 1 # Current line number + self.lexoptimize = False # Optimized mode + + def clone(self, object=None): + c = copy.copy(self) + + # If the object parameter has been supplied, it means we are attaching the + # lexer to a new object. In this case, we have to rebind all methods in + # the lexstatere and lexstateerrorf tables. + + if object: + newtab = {} + for key, ritem in self.lexstatere.items(): + newre = [] + for cre, findex in ritem: + newfindex = [] + for f in findex: + if not f or not f[0]: + newfindex.append(f) + continue + newfindex.append((getattr(object, f[0].__name__), f[1])) + newre.append((cre, newfindex)) + newtab[key] = newre + c.lexstatere = newtab + c.lexstateerrorf = {} + for key, ef in self.lexstateerrorf.items(): + c.lexstateerrorf[key] = getattr(object, ef.__name__) + c.lexmodule = object + return c + + # ------------------------------------------------------------ + # writetab() - Write lexer information to a table file + # ------------------------------------------------------------ + def writetab(self, lextab, outputdir=''): + if isinstance(lextab, types.ModuleType): + raise IOError("Won't overwrite existing lextab module") + basetabmodule = lextab.split('.')[-1] + filename = os.path.join(outputdir, basetabmodule) + '.py' + with open(filename, 'w') as tf: + tf.write('# %s.py. This file automatically created by PLY (version %s). Don\'t edit!\n' % (basetabmodule, __version__)) + tf.write('_tabversion = %s\n' % repr(__tabversion__)) + tf.write('_lextokens = set(%s)\n' % repr(tuple(self.lextokens))) + tf.write('_lexreflags = %s\n' % repr(self.lexreflags)) + tf.write('_lexliterals = %s\n' % repr(self.lexliterals)) + tf.write('_lexstateinfo = %s\n' % repr(self.lexstateinfo)) + + # Rewrite the lexstatere table, replacing function objects with function names + tabre = {} + for statename, lre in self.lexstatere.items(): + titem = [] + for (pat, func), retext, renames in zip(lre, self.lexstateretext[statename], self.lexstaterenames[statename]): + titem.append((retext, _funcs_to_names(func, renames))) + tabre[statename] = titem + + tf.write('_lexstatere = %s\n' % repr(tabre)) + tf.write('_lexstateignore = %s\n' % repr(self.lexstateignore)) + + taberr = {} + for statename, ef in self.lexstateerrorf.items(): + taberr[statename] = ef.__name__ if ef else None + tf.write('_lexstateerrorf = %s\n' % repr(taberr)) + + tabeof = {} + for statename, ef in self.lexstateeoff.items(): + tabeof[statename] = ef.__name__ if ef else None + tf.write('_lexstateeoff = %s\n' % repr(tabeof)) + + # ------------------------------------------------------------ + # readtab() - Read lexer information from a tab file + # ------------------------------------------------------------ + def readtab(self, tabfile, fdict): + if isinstance(tabfile, types.ModuleType): + lextab = tabfile + else: + exec('import %s' % tabfile) + lextab = sys.modules[tabfile] + + if getattr(lextab, '_tabversion', '0.0') != __tabversion__: + raise ImportError('Inconsistent PLY version') + + self.lextokens = lextab._lextokens + self.lexreflags = lextab._lexreflags + self.lexliterals = lextab._lexliterals + self.lextokens_all = self.lextokens | set(self.lexliterals) + self.lexstateinfo = lextab._lexstateinfo + self.lexstateignore = lextab._lexstateignore + self.lexstatere = {} + self.lexstateretext = {} + for statename, lre in lextab._lexstatere.items(): + titem = [] + txtitem = [] + for pat, func_name in lre: + titem.append((re.compile(pat, lextab._lexreflags), _names_to_funcs(func_name, fdict))) + + self.lexstatere[statename] = titem + self.lexstateretext[statename] = txtitem + + self.lexstateerrorf = {} + for statename, ef in lextab._lexstateerrorf.items(): + self.lexstateerrorf[statename] = fdict[ef] + + self.lexstateeoff = {} + for statename, ef in lextab._lexstateeoff.items(): + self.lexstateeoff[statename] = fdict[ef] + + self.begin('INITIAL') + + # ------------------------------------------------------------ + # input() - Push a new string into the lexer + # ------------------------------------------------------------ + def input(self, s): + # Pull off the first character to see if s looks like a string + c = s[:1] + if not isinstance(c, StringTypes): + raise ValueError('Expected a string') + self.lexdata = s + self.lexpos = 0 + self.lexlen = len(s) + + # ------------------------------------------------------------ + # begin() - Changes the lexing state + # ------------------------------------------------------------ + def begin(self, state): + if state not in self.lexstatere: + raise ValueError('Undefined state') + self.lexre = self.lexstatere[state] + self.lexretext = self.lexstateretext[state] + self.lexignore = self.lexstateignore.get(state, '') + self.lexerrorf = self.lexstateerrorf.get(state, None) + self.lexeoff = self.lexstateeoff.get(state, None) + self.lexstate = state + + # ------------------------------------------------------------ + # push_state() - Changes the lexing state and saves old on stack + # ------------------------------------------------------------ + def push_state(self, state): + self.lexstatestack.append(self.lexstate) + self.begin(state) + + # ------------------------------------------------------------ + # pop_state() - Restores the previous state + # ------------------------------------------------------------ + def pop_state(self): + self.begin(self.lexstatestack.pop()) + + # ------------------------------------------------------------ + # current_state() - Returns the current lexing state + # ------------------------------------------------------------ + def current_state(self): + return self.lexstate + + # ------------------------------------------------------------ + # skip() - Skip ahead n characters + # ------------------------------------------------------------ + def skip(self, n): + self.lexpos += n + + # ------------------------------------------------------------ + # opttoken() - Return the next token from the Lexer + # + # Note: This function has been carefully implemented to be as fast + # as possible. Don't make changes unless you really know what + # you are doing + # ------------------------------------------------------------ + def token(self): + # Make local copies of frequently referenced attributes + lexpos = self.lexpos + lexlen = self.lexlen + lexignore = self.lexignore + lexdata = self.lexdata + + while lexpos < lexlen: + # This code provides some short-circuit code for whitespace, tabs, and other ignored characters + if lexdata[lexpos] in lexignore: + lexpos += 1 + continue + + # Look for a regular expression match + for lexre, lexindexfunc in self.lexre: + m = lexre.match(lexdata, lexpos) + if not m: + continue + + # Create a token for return + tok = LexToken() + tok.value = m.group() + tok.lineno = self.lineno + tok.lexpos = lexpos + + i = m.lastindex + func, tok.type = lexindexfunc[i] + + if not func: + # If no token type was set, it's an ignored token + if tok.type: + self.lexpos = m.end() + return tok + else: + lexpos = m.end() + break + + lexpos = m.end() + + # If token is processed by a function, call it + + tok.lexer = self # Set additional attributes useful in token rules + self.lexmatch = m + self.lexpos = lexpos + + newtok = func(tok) + + # Every function must return a token, if nothing, we just move to next token + if not newtok: + lexpos = self.lexpos # This is here in case user has updated lexpos. + lexignore = self.lexignore # This is here in case there was a state change + break + + # Verify type of the token. If not in the token map, raise an error + if not self.lexoptimize: + if newtok.type not in self.lextokens_all: + raise LexError("%s:%d: Rule '%s' returned an unknown token type '%s'" % ( + func.__code__.co_filename, func.__code__.co_firstlineno, + func.__name__, newtok.type), lexdata[lexpos:]) + + return newtok + else: + # No match, see if in literals + if lexdata[lexpos] in self.lexliterals: + tok = LexToken() + tok.value = lexdata[lexpos] + tok.lineno = self.lineno + tok.type = tok.value + tok.lexpos = lexpos + self.lexpos = lexpos + 1 + return tok + + # No match. Call t_error() if defined. + if self.lexerrorf: + tok = LexToken() + tok.value = self.lexdata[lexpos:] + tok.lineno = self.lineno + tok.type = 'error' + tok.lexer = self + tok.lexpos = lexpos + self.lexpos = lexpos + newtok = self.lexerrorf(tok) + if lexpos == self.lexpos: + # Error method didn't change text position at all. This is an error. + raise LexError("Scanning error. Illegal character '%s'" % (lexdata[lexpos]), lexdata[lexpos:]) + lexpos = self.lexpos + if not newtok: + continue + return newtok + + self.lexpos = lexpos + raise LexError("Illegal character '%s' at index %d" % (lexdata[lexpos], lexpos), lexdata[lexpos:]) + + if self.lexeoff: + tok = LexToken() + tok.type = 'eof' + tok.value = '' + tok.lineno = self.lineno + tok.lexpos = lexpos + tok.lexer = self + self.lexpos = lexpos + newtok = self.lexeoff(tok) + return newtok + + self.lexpos = lexpos + 1 + if self.lexdata is None: + raise RuntimeError('No input string given with input()') + return None + + # Iterator interface + def __iter__(self): + return self + + def next(self): + t = self.token() + if t is None: + raise StopIteration + return t + + __next__ = next + +# ----------------------------------------------------------------------------- +# ==== Lex Builder === +# +# The functions and classes below are used to collect lexing information +# and build a Lexer object from it. +# ----------------------------------------------------------------------------- + +# ----------------------------------------------------------------------------- +# _get_regex(func) +# +# Returns the regular expression assigned to a function either as a doc string +# or as a .regex attribute attached by the @TOKEN decorator. +# ----------------------------------------------------------------------------- +def _get_regex(func): + return getattr(func, 'regex', func.__doc__) + +# ----------------------------------------------------------------------------- +# get_caller_module_dict() +# +# This function returns a dictionary containing all of the symbols defined within +# a caller further down the call stack. This is used to get the environment +# associated with the yacc() call if none was provided. +# ----------------------------------------------------------------------------- +def get_caller_module_dict(levels): + f = sys._getframe(levels) + ldict = f.f_globals.copy() + if f.f_globals != f.f_locals: + ldict.update(f.f_locals) + return ldict + +# ----------------------------------------------------------------------------- +# _funcs_to_names() +# +# Given a list of regular expression functions, this converts it to a list +# suitable for output to a table file +# ----------------------------------------------------------------------------- +def _funcs_to_names(funclist, namelist): + result = [] + for f, name in zip(funclist, namelist): + if f and f[0]: + result.append((name, f[1])) + else: + result.append(f) + return result + +# ----------------------------------------------------------------------------- +# _names_to_funcs() +# +# Given a list of regular expression function names, this converts it back to +# functions. +# ----------------------------------------------------------------------------- +def _names_to_funcs(namelist, fdict): + result = [] + for n in namelist: + if n and n[0]: + result.append((fdict[n[0]], n[1])) + else: + result.append(n) + return result + +# ----------------------------------------------------------------------------- +# _form_master_re() +# +# This function takes a list of all of the regex components and attempts to +# form the master regular expression. Given limitations in the Python re +# module, it may be necessary to break the master regex into separate expressions. +# ----------------------------------------------------------------------------- +def _form_master_re(relist, reflags, ldict, toknames): + if not relist: + return [] + regex = '|'.join(relist) + try: + lexre = re.compile(regex, reflags) + + # Build the index to function map for the matching engine + lexindexfunc = [None] * (max(lexre.groupindex.values()) + 1) + lexindexnames = lexindexfunc[:] + + for f, i in lexre.groupindex.items(): + handle = ldict.get(f, None) + if type(handle) in (types.FunctionType, types.MethodType): + lexindexfunc[i] = (handle, toknames[f]) + lexindexnames[i] = f + elif handle is not None: + lexindexnames[i] = f + if f.find('ignore_') > 0: + lexindexfunc[i] = (None, None) + else: + lexindexfunc[i] = (None, toknames[f]) + + return [(lexre, lexindexfunc)], [regex], [lexindexnames] + except Exception: + m = int(len(relist)/2) + if m == 0: + m = 1 + llist, lre, lnames = _form_master_re(relist[:m], reflags, ldict, toknames) + rlist, rre, rnames = _form_master_re(relist[m:], reflags, ldict, toknames) + return (llist+rlist), (lre+rre), (lnames+rnames) + +# ----------------------------------------------------------------------------- +# def _statetoken(s,names) +# +# Given a declaration name s of the form "t_" and a dictionary whose keys are +# state names, this function returns a tuple (states,tokenname) where states +# is a tuple of state names and tokenname is the name of the token. For example, +# calling this with s = "t_foo_bar_SPAM" might return (('foo','bar'),'SPAM') +# ----------------------------------------------------------------------------- +def _statetoken(s, names): + nonstate = 1 + parts = s.split('_') + for i, part in enumerate(parts[1:], 1): + if part not in names and part != 'ANY': + break + + if i > 1: + states = tuple(parts[1:i]) + else: + states = ('INITIAL',) + + if 'ANY' in states: + states = tuple(names) + + tokenname = '_'.join(parts[i:]) + return (states, tokenname) + + +# ----------------------------------------------------------------------------- +# LexerReflect() +# +# This class represents information needed to build a lexer as extracted from a +# user's input file. +# ----------------------------------------------------------------------------- +class LexerReflect(object): + def __init__(self, ldict, log=None, reflags=0): + self.ldict = ldict + self.error_func = None + self.tokens = [] + self.reflags = reflags + self.stateinfo = {'INITIAL': 'inclusive'} + self.modules = set() + self.error = False + self.log = PlyLogger(sys.stderr) if log is None else log + + # Get all of the basic information + def get_all(self): + self.get_tokens() + self.get_literals() + self.get_states() + self.get_rules() + + # Validate all of the information + def validate_all(self): + self.validate_tokens() + self.validate_literals() + self.validate_rules() + return self.error + + # Get the tokens map + def get_tokens(self): + tokens = self.ldict.get('tokens', None) + if not tokens: + self.log.error('No token list is defined') + self.error = True + return + + if not isinstance(tokens, (list, tuple)): + self.log.error('tokens must be a list or tuple') + self.error = True + return + + if not tokens: + self.log.error('tokens is empty') + self.error = True + return + + self.tokens = tokens + + # Validate the tokens + def validate_tokens(self): + terminals = {} + for n in self.tokens: + if not _is_identifier.match(n): + self.log.error("Bad token name '%s'", n) + self.error = True + if n in terminals: + self.log.warning("Token '%s' multiply defined", n) + terminals[n] = 1 + + # Get the literals specifier + def get_literals(self): + self.literals = self.ldict.get('literals', '') + if not self.literals: + self.literals = '' + + # Validate literals + def validate_literals(self): + try: + for c in self.literals: + if not isinstance(c, StringTypes) or len(c) > 1: + self.log.error('Invalid literal %s. Must be a single character', repr(c)) + self.error = True + + except TypeError: + self.log.error('Invalid literals specification. literals must be a sequence of characters') + self.error = True + + def get_states(self): + self.states = self.ldict.get('states', None) + # Build statemap + if self.states: + if not isinstance(self.states, (tuple, list)): + self.log.error('states must be defined as a tuple or list') + self.error = True + else: + for s in self.states: + if not isinstance(s, tuple) or len(s) != 2: + self.log.error("Invalid state specifier %s. Must be a tuple (statename,'exclusive|inclusive')", repr(s)) + self.error = True + continue + name, statetype = s + if not isinstance(name, StringTypes): + self.log.error('State name %s must be a string', repr(name)) + self.error = True + continue + if not (statetype == 'inclusive' or statetype == 'exclusive'): + self.log.error("State type for state %s must be 'inclusive' or 'exclusive'", name) + self.error = True + continue + if name in self.stateinfo: + self.log.error("State '%s' already defined", name) + self.error = True + continue + self.stateinfo[name] = statetype + + # Get all of the symbols with a t_ prefix and sort them into various + # categories (functions, strings, error functions, and ignore characters) + + def get_rules(self): + tsymbols = [f for f in self.ldict if f[:2] == 't_'] + + # Now build up a list of functions and a list of strings + self.toknames = {} # Mapping of symbols to token names + self.funcsym = {} # Symbols defined as functions + self.strsym = {} # Symbols defined as strings + self.ignore = {} # Ignore strings by state + self.errorf = {} # Error functions by state + self.eoff = {} # EOF functions by state + + for s in self.stateinfo: + self.funcsym[s] = [] + self.strsym[s] = [] + + if len(tsymbols) == 0: + self.log.error('No rules of the form t_rulename are defined') + self.error = True + return + + for f in tsymbols: + t = self.ldict[f] + states, tokname = _statetoken(f, self.stateinfo) + self.toknames[f] = tokname + + if hasattr(t, '__call__'): + if tokname == 'error': + for s in states: + self.errorf[s] = t + elif tokname == 'eof': + for s in states: + self.eoff[s] = t + elif tokname == 'ignore': + line = t.__code__.co_firstlineno + file = t.__code__.co_filename + self.log.error("%s:%d: Rule '%s' must be defined as a string", file, line, t.__name__) + self.error = True + else: + for s in states: + self.funcsym[s].append((f, t)) + elif isinstance(t, StringTypes): + if tokname == 'ignore': + for s in states: + self.ignore[s] = t + if '\\' in t: + self.log.warning("%s contains a literal backslash '\\'", f) + + elif tokname == 'error': + self.log.error("Rule '%s' must be defined as a function", f) + self.error = True + else: + for s in states: + self.strsym[s].append((f, t)) + else: + self.log.error('%s not defined as a function or string', f) + self.error = True + + # Sort the functions by line number + for f in self.funcsym.values(): + f.sort(key=lambda x: x[1].__code__.co_firstlineno) + + # Sort the strings by regular expression length + for s in self.strsym.values(): + s.sort(key=lambda x: len(x[1]), reverse=True) + + # Validate all of the t_rules collected + def validate_rules(self): + for state in self.stateinfo: + # Validate all rules defined by functions + + for fname, f in self.funcsym[state]: + line = f.__code__.co_firstlineno + file = f.__code__.co_filename + module = inspect.getmodule(f) + self.modules.add(module) + + tokname = self.toknames[fname] + if isinstance(f, types.MethodType): + reqargs = 2 + else: + reqargs = 1 + nargs = f.__code__.co_argcount + if nargs > reqargs: + self.log.error("%s:%d: Rule '%s' has too many arguments", file, line, f.__name__) + self.error = True + continue + + if nargs < reqargs: + self.log.error("%s:%d: Rule '%s' requires an argument", file, line, f.__name__) + self.error = True + continue + + if not _get_regex(f): + self.log.error("%s:%d: No regular expression defined for rule '%s'", file, line, f.__name__) + self.error = True + continue + + try: + c = re.compile('(?P<%s>%s)' % (fname, _get_regex(f)), self.reflags) + if c.match(''): + self.log.error("%s:%d: Regular expression for rule '%s' matches empty string", file, line, f.__name__) + self.error = True + except re.error as e: + self.log.error("%s:%d: Invalid regular expression for rule '%s'. %s", file, line, f.__name__, e) + if '#' in _get_regex(f): + self.log.error("%s:%d. Make sure '#' in rule '%s' is escaped with '\\#'", file, line, f.__name__) + self.error = True + + # Validate all rules defined by strings + for name, r in self.strsym[state]: + tokname = self.toknames[name] + if tokname == 'error': + self.log.error("Rule '%s' must be defined as a function", name) + self.error = True + continue + + if tokname not in self.tokens and tokname.find('ignore_') < 0: + self.log.error("Rule '%s' defined for an unspecified token %s", name, tokname) + self.error = True + continue + + try: + c = re.compile('(?P<%s>%s)' % (name, r), self.reflags) + if (c.match('')): + self.log.error("Regular expression for rule '%s' matches empty string", name) + self.error = True + except re.error as e: + self.log.error("Invalid regular expression for rule '%s'. %s", name, e) + if '#' in r: + self.log.error("Make sure '#' in rule '%s' is escaped with '\\#'", name) + self.error = True + + if not self.funcsym[state] and not self.strsym[state]: + self.log.error("No rules defined for state '%s'", state) + self.error = True + + # Validate the error function + efunc = self.errorf.get(state, None) + if efunc: + f = efunc + line = f.__code__.co_firstlineno + file = f.__code__.co_filename + module = inspect.getmodule(f) + self.modules.add(module) + + if isinstance(f, types.MethodType): + reqargs = 2 + else: + reqargs = 1 + nargs = f.__code__.co_argcount + if nargs > reqargs: + self.log.error("%s:%d: Rule '%s' has too many arguments", file, line, f.__name__) + self.error = True + + if nargs < reqargs: + self.log.error("%s:%d: Rule '%s' requires an argument", file, line, f.__name__) + self.error = True + + for module in self.modules: + self.validate_module(module) + + # ----------------------------------------------------------------------------- + # validate_module() + # + # This checks to see if there are duplicated t_rulename() functions or strings + # in the parser input file. This is done using a simple regular expression + # match on each line in the source code of the given module. + # ----------------------------------------------------------------------------- + + def validate_module(self, module): + try: + lines, linen = inspect.getsourcelines(module) + except IOError: + return + + fre = re.compile(r'\s*def\s+(t_[a-zA-Z_0-9]*)\(') + sre = re.compile(r'\s*(t_[a-zA-Z_0-9]*)\s*=') + + counthash = {} + linen += 1 + for line in lines: + m = fre.match(line) + if not m: + m = sre.match(line) + if m: + name = m.group(1) + prev = counthash.get(name) + if not prev: + counthash[name] = linen + else: + filename = inspect.getsourcefile(module) + self.log.error('%s:%d: Rule %s redefined. Previously defined on line %d', filename, linen, name, prev) + self.error = True + linen += 1 + +# ----------------------------------------------------------------------------- +# lex(module) +# +# Build all of the regular expression rules from definitions in the supplied module +# ----------------------------------------------------------------------------- +def lex(module=None, object=None, debug=False, optimize=False, lextab='lextab', + reflags=int(re.VERBOSE), nowarn=False, outputdir=None, debuglog=None, errorlog=None): + + if lextab is None: + lextab = 'lextab' + + global lexer + + ldict = None + stateinfo = {'INITIAL': 'inclusive'} + lexobj = Lexer() + lexobj.lexoptimize = optimize + global token, input + + if errorlog is None: + errorlog = PlyLogger(sys.stderr) + + if debug: + if debuglog is None: + debuglog = PlyLogger(sys.stderr) + + # Get the module dictionary used for the lexer + if object: + module = object + + # Get the module dictionary used for the parser + if module: + _items = [(k, getattr(module, k)) for k in dir(module)] + ldict = dict(_items) + # If no __file__ attribute is available, try to obtain it from the __module__ instead + if '__file__' not in ldict: + ldict['__file__'] = sys.modules[ldict['__module__']].__file__ + else: + ldict = get_caller_module_dict(2) + + # Determine if the module is package of a package or not. + # If so, fix the tabmodule setting so that tables load correctly + pkg = ldict.get('__package__') + if pkg and isinstance(lextab, str): + if '.' not in lextab: + lextab = pkg + '.' + lextab + + # Collect parser information from the dictionary + linfo = LexerReflect(ldict, log=errorlog, reflags=reflags) + linfo.get_all() + if not optimize: + if linfo.validate_all(): + raise SyntaxError("Can't build lexer") + + if optimize and lextab: + try: + lexobj.readtab(lextab, ldict) + token = lexobj.token + input = lexobj.input + lexer = lexobj + return lexobj + + except ImportError: + pass + + # Dump some basic debugging information + if debug: + debuglog.info('lex: tokens = %r', linfo.tokens) + debuglog.info('lex: literals = %r', linfo.literals) + debuglog.info('lex: states = %r', linfo.stateinfo) + + # Build a dictionary of valid token names + lexobj.lextokens = set() + for n in linfo.tokens: + lexobj.lextokens.add(n) + + # Get literals specification + if isinstance(linfo.literals, (list, tuple)): + lexobj.lexliterals = type(linfo.literals[0])().join(linfo.literals) + else: + lexobj.lexliterals = linfo.literals + + lexobj.lextokens_all = lexobj.lextokens | set(lexobj.lexliterals) + + # Get the stateinfo dictionary + stateinfo = linfo.stateinfo + + regexs = {} + # Build the master regular expressions + for state in stateinfo: + regex_list = [] + + # Add rules defined by functions first + for fname, f in linfo.funcsym[state]: + line = f.__code__.co_firstlineno + file = f.__code__.co_filename + regex_list.append('(?P<%s>%s)' % (fname, _get_regex(f))) + if debug: + debuglog.info("lex: Adding rule %s -> '%s' (state '%s')", fname, _get_regex(f), state) + + # Now add all of the simple rules + for name, r in linfo.strsym[state]: + regex_list.append('(?P<%s>%s)' % (name, r)) + if debug: + debuglog.info("lex: Adding rule %s -> '%s' (state '%s')", name, r, state) + + regexs[state] = regex_list + + # Build the master regular expressions + + if debug: + debuglog.info('lex: ==== MASTER REGEXS FOLLOW ====') + + for state in regexs: + lexre, re_text, re_names = _form_master_re(regexs[state], reflags, ldict, linfo.toknames) + lexobj.lexstatere[state] = lexre + lexobj.lexstateretext[state] = re_text + lexobj.lexstaterenames[state] = re_names + if debug: + for i, text in enumerate(re_text): + debuglog.info("lex: state '%s' : regex[%d] = '%s'", state, i, text) + + # For inclusive states, we need to add the regular expressions from the INITIAL state + for state, stype in stateinfo.items(): + if state != 'INITIAL' and stype == 'inclusive': + lexobj.lexstatere[state].extend(lexobj.lexstatere['INITIAL']) + lexobj.lexstateretext[state].extend(lexobj.lexstateretext['INITIAL']) + lexobj.lexstaterenames[state].extend(lexobj.lexstaterenames['INITIAL']) + + lexobj.lexstateinfo = stateinfo + lexobj.lexre = lexobj.lexstatere['INITIAL'] + lexobj.lexretext = lexobj.lexstateretext['INITIAL'] + lexobj.lexreflags = reflags + + # Set up ignore variables + lexobj.lexstateignore = linfo.ignore + lexobj.lexignore = lexobj.lexstateignore.get('INITIAL', '') + + # Set up error functions + lexobj.lexstateerrorf = linfo.errorf + lexobj.lexerrorf = linfo.errorf.get('INITIAL', None) + if not lexobj.lexerrorf: + errorlog.warning('No t_error rule is defined') + + # Set up eof functions + lexobj.lexstateeoff = linfo.eoff + lexobj.lexeoff = linfo.eoff.get('INITIAL', None) + + # Check state information for ignore and error rules + for s, stype in stateinfo.items(): + if stype == 'exclusive': + if s not in linfo.errorf: + errorlog.warning("No error rule is defined for exclusive state '%s'", s) + if s not in linfo.ignore and lexobj.lexignore: + errorlog.warning("No ignore rule is defined for exclusive state '%s'", s) + elif stype == 'inclusive': + if s not in linfo.errorf: + linfo.errorf[s] = linfo.errorf.get('INITIAL', None) + if s not in linfo.ignore: + linfo.ignore[s] = linfo.ignore.get('INITIAL', '') + + # Create global versions of the token() and input() functions + token = lexobj.token + input = lexobj.input + lexer = lexobj + + # If in optimize mode, we write the lextab + if lextab and optimize: + if outputdir is None: + # If no output directory is set, the location of the output files + # is determined according to the following rules: + # - If lextab specifies a package, files go into that package directory + # - Otherwise, files go in the same directory as the specifying module + if isinstance(lextab, types.ModuleType): + srcfile = lextab.__file__ + else: + if '.' not in lextab: + srcfile = ldict['__file__'] + else: + parts = lextab.split('.') + pkgname = '.'.join(parts[:-1]) + exec('import %s' % pkgname) + srcfile = getattr(sys.modules[pkgname], '__file__', '') + outputdir = os.path.dirname(srcfile) + try: + lexobj.writetab(lextab, outputdir) + except IOError as e: + errorlog.warning("Couldn't write lextab module %r. %s" % (lextab, e)) + + return lexobj + +# ----------------------------------------------------------------------------- +# runmain() +# +# This runs the lexer as a main program +# ----------------------------------------------------------------------------- + +def runmain(lexer=None, data=None): + if not data: + try: + filename = sys.argv[1] + f = open(filename) + data = f.read() + f.close() + except IndexError: + sys.stdout.write('Reading from standard input (type EOF to end):\n') + data = sys.stdin.read() + + if lexer: + _input = lexer.input + else: + _input = input + _input(data) + if lexer: + _token = lexer.token + else: + _token = token + + while True: + tok = _token() + if not tok: + break + sys.stdout.write('(%s,%r,%d,%d)\n' % (tok.type, tok.value, tok.lineno, tok.lexpos)) + +# ----------------------------------------------------------------------------- +# @TOKEN(regex) +# +# This decorator function can be used to set the regex expression on a function +# when its docstring might need to be set in an alternative way +# ----------------------------------------------------------------------------- + +def TOKEN(r): + def set_regex(f): + if hasattr(r, '__call__'): + f.regex = _get_regex(r) + else: + f.regex = r + return f + return set_regex + +# Alternative spelling of the TOKEN decorator +Token = TOKEN + diff --git a/third_party/python/ply/ply/yacc.py b/third_party/python/ply/ply/yacc.py new file mode 100644 index 0000000000..03bd86ee07 --- /dev/null +++ b/third_party/python/ply/ply/yacc.py @@ -0,0 +1,3494 @@ +# ----------------------------------------------------------------------------- +# ply: yacc.py +# +# Copyright (C) 2001-2017 +# David M. Beazley (Dabeaz LLC) +# All rights reserved. +# +# Redistribution and use in source and binary forms, with or without +# modification, are permitted provided that the following conditions are +# met: +# +# * Redistributions of source code must retain the above copyright notice, +# this list of conditions and the following disclaimer. +# * Redistributions in binary form must reproduce the above copyright notice, +# this list of conditions and the following disclaimer in the documentation +# and/or other materials provided with the distribution. +# * Neither the name of the David Beazley or Dabeaz LLC may be used to +# endorse or promote products derived from this software without +# specific prior written permission. +# +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +# ----------------------------------------------------------------------------- +# +# This implements an LR parser that is constructed from grammar rules defined +# as Python functions. The grammer is specified by supplying the BNF inside +# Python documentation strings. The inspiration for this technique was borrowed +# from John Aycock's Spark parsing system. PLY might be viewed as cross between +# Spark and the GNU bison utility. +# +# The current implementation is only somewhat object-oriented. The +# LR parser itself is defined in terms of an object (which allows multiple +# parsers to co-exist). However, most of the variables used during table +# construction are defined in terms of global variables. Users shouldn't +# notice unless they are trying to define multiple parsers at the same +# time using threads (in which case they should have their head examined). +# +# This implementation supports both SLR and LALR(1) parsing. LALR(1) +# support was originally implemented by Elias Ioup (ezioup@alumni.uchicago.edu), +# using the algorithm found in Aho, Sethi, and Ullman "Compilers: Principles, +# Techniques, and Tools" (The Dragon Book). LALR(1) has since been replaced +# by the more efficient DeRemer and Pennello algorithm. +# +# :::::::: WARNING ::::::: +# +# Construction of LR parsing tables is fairly complicated and expensive. +# To make this module run fast, a *LOT* of work has been put into +# optimization---often at the expensive of readability and what might +# consider to be good Python "coding style." Modify the code at your +# own risk! +# ---------------------------------------------------------------------------- + +import re +import types +import sys +import os.path +import inspect +import base64 +import warnings + +__version__ = '3.10' +__tabversion__ = '3.10' + +#----------------------------------------------------------------------------- +# === User configurable parameters === +# +# Change these to modify the default behavior of yacc (if you wish) +#----------------------------------------------------------------------------- + +yaccdebug = True # Debugging mode. If set, yacc generates a + # a 'parser.out' file in the current directory + +debug_file = 'parser.out' # Default name of the debugging file +tab_module = 'parsetab' # Default name of the table module +default_lr = 'LALR' # Default LR table generation method + +error_count = 3 # Number of symbols that must be shifted to leave recovery mode + +yaccdevel = False # Set to True if developing yacc. This turns off optimized + # implementations of certain functions. + +resultlimit = 40 # Size limit of results when running in debug mode. + +pickle_protocol = 0 # Protocol to use when writing pickle files + +# String type-checking compatibility +if sys.version_info[0] < 3: + string_types = basestring +else: + string_types = str + +MAXINT = sys.maxsize + +# This object is a stand-in for a logging object created by the +# logging module. PLY will use this by default to create things +# such as the parser.out file. If a user wants more detailed +# information, they can create their own logging object and pass +# it into PLY. + +class PlyLogger(object): + def __init__(self, f): + self.f = f + + def debug(self, msg, *args, **kwargs): + self.f.write((msg % args) + '\n') + + info = debug + + def warning(self, msg, *args, **kwargs): + self.f.write('WARNING: ' + (msg % args) + '\n') + + def error(self, msg, *args, **kwargs): + self.f.write('ERROR: ' + (msg % args) + '\n') + + critical = debug + +# Null logger is used when no output is generated. Does nothing. +class NullLogger(object): + def __getattribute__(self, name): + return self + + def __call__(self, *args, **kwargs): + return self + +# Exception raised for yacc-related errors +class YaccError(Exception): + pass + +# Format the result message that the parser produces when running in debug mode. +def format_result(r): + repr_str = repr(r) + if '\n' in repr_str: + repr_str = repr(repr_str) + if len(repr_str) > resultlimit: + repr_str = repr_str[:resultlimit] + ' ...' + result = '<%s @ 0x%x> (%s)' % (type(r).__name__, id(r), repr_str) + return result + +# Format stack entries when the parser is running in debug mode +def format_stack_entry(r): + repr_str = repr(r) + if '\n' in repr_str: + repr_str = repr(repr_str) + if len(repr_str) < 16: + return repr_str + else: + return '<%s @ 0x%x>' % (type(r).__name__, id(r)) + +# Panic mode error recovery support. This feature is being reworked--much of the +# code here is to offer a deprecation/backwards compatible transition + +_errok = None +_token = None +_restart = None +_warnmsg = '''PLY: Don't use global functions errok(), token(), and restart() in p_error(). +Instead, invoke the methods on the associated parser instance: + + def p_error(p): + ... + # Use parser.errok(), parser.token(), parser.restart() + ... + + parser = yacc.yacc() +''' + +def errok(): + warnings.warn(_warnmsg) + return _errok() + +def restart(): + warnings.warn(_warnmsg) + return _restart() + +def token(): + warnings.warn(_warnmsg) + return _token() + +# Utility function to call the p_error() function with some deprecation hacks +def call_errorfunc(errorfunc, token, parser): + global _errok, _token, _restart + _errok = parser.errok + _token = parser.token + _restart = parser.restart + r = errorfunc(token) + try: + del _errok, _token, _restart + except NameError: + pass + return r + +#----------------------------------------------------------------------------- +# === LR Parsing Engine === +# +# The following classes are used for the LR parser itself. These are not +# used during table construction and are independent of the actual LR +# table generation algorithm +#----------------------------------------------------------------------------- + +# This class is used to hold non-terminal grammar symbols during parsing. +# It normally has the following attributes set: +# .type = Grammar symbol type +# .value = Symbol value +# .lineno = Starting line number +# .endlineno = Ending line number (optional, set automatically) +# .lexpos = Starting lex position +# .endlexpos = Ending lex position (optional, set automatically) + +class YaccSymbol: + def __str__(self): + return self.type + + def __repr__(self): + return str(self) + +# This class is a wrapper around the objects actually passed to each +# grammar rule. Index lookup and assignment actually assign the +# .value attribute of the underlying YaccSymbol object. +# The lineno() method returns the line number of a given +# item (or 0 if not defined). The linespan() method returns +# a tuple of (startline,endline) representing the range of lines +# for a symbol. The lexspan() method returns a tuple (lexpos,endlexpos) +# representing the range of positional information for a symbol. + +class YaccProduction: + def __init__(self, s, stack=None): + self.slice = s + self.stack = stack + self.lexer = None + self.parser = None + + def __getitem__(self, n): + if isinstance(n, slice): + return [s.value for s in self.slice[n]] + elif n >= 0: + return self.slice[n].value + else: + return self.stack[n].value + + def __setitem__(self, n, v): + self.slice[n].value = v + + def __getslice__(self, i, j): + return [s.value for s in self.slice[i:j]] + + def __len__(self): + return len(self.slice) + + def lineno(self, n): + return getattr(self.slice[n], 'lineno', 0) + + def set_lineno(self, n, lineno): + self.slice[n].lineno = lineno + + def linespan(self, n): + startline = getattr(self.slice[n], 'lineno', 0) + endline = getattr(self.slice[n], 'endlineno', startline) + return startline, endline + + def lexpos(self, n): + return getattr(self.slice[n], 'lexpos', 0) + + def lexspan(self, n): + startpos = getattr(self.slice[n], 'lexpos', 0) + endpos = getattr(self.slice[n], 'endlexpos', startpos) + return startpos, endpos + + def error(self): + raise SyntaxError + +# ----------------------------------------------------------------------------- +# == LRParser == +# +# The LR Parsing engine. +# ----------------------------------------------------------------------------- + +class LRParser: + def __init__(self, lrtab, errorf): + self.productions = lrtab.lr_productions + self.action = lrtab.lr_action + self.goto = lrtab.lr_goto + self.errorfunc = errorf + self.set_defaulted_states() + self.errorok = True + + def errok(self): + self.errorok = True + + def restart(self): + del self.statestack[:] + del self.symstack[:] + sym = YaccSymbol() + sym.type = '$end' + self.symstack.append(sym) + self.statestack.append(0) + + # Defaulted state support. + # This method identifies parser states where there is only one possible reduction action. + # For such states, the parser can make a choose to make a rule reduction without consuming + # the next look-ahead token. This delayed invocation of the tokenizer can be useful in + # certain kinds of advanced parsing situations where the lexer and parser interact with + # each other or change states (i.e., manipulation of scope, lexer states, etc.). + # + # See: http://www.gnu.org/software/bison/manual/html_node/Default-Reductions.html#Default-Reductions + def set_defaulted_states(self): + self.defaulted_states = {} + for state, actions in self.action.items(): + rules = list(actions.values()) + if len(rules) == 1 and rules[0] < 0: + self.defaulted_states[state] = rules[0] + + def disable_defaulted_states(self): + self.defaulted_states = {} + + def parse(self, input=None, lexer=None, debug=False, tracking=False, tokenfunc=None): + if debug or yaccdevel: + if isinstance(debug, int): + debug = PlyLogger(sys.stderr) + return self.parsedebug(input, lexer, debug, tracking, tokenfunc) + elif tracking: + return self.parseopt(input, lexer, debug, tracking, tokenfunc) + else: + return self.parseopt_notrack(input, lexer, debug, tracking, tokenfunc) + + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # parsedebug(). + # + # This is the debugging enabled version of parse(). All changes made to the + # parsing engine should be made here. Optimized versions of this function + # are automatically created by the ply/ygen.py script. This script cuts out + # sections enclosed in markers such as this: + # + # #--! DEBUG + # statements + # #--! DEBUG + # + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + def parsedebug(self, input=None, lexer=None, debug=False, tracking=False, tokenfunc=None): + #--! parsedebug-start + lookahead = None # Current lookahead symbol + lookaheadstack = [] # Stack of lookahead symbols + actions = self.action # Local reference to action table (to avoid lookup on self.) + goto = self.goto # Local reference to goto table (to avoid lookup on self.) + prod = self.productions # Local reference to production list (to avoid lookup on self.) + defaulted_states = self.defaulted_states # Local reference to defaulted states + pslice = YaccProduction(None) # Production object passed to grammar rules + errorcount = 0 # Used during error recovery + + #--! DEBUG + debug.info('PLY: PARSE DEBUG START') + #--! DEBUG + + # If no lexer was given, we will try to use the lex module + if not lexer: + from . import lex + lexer = lex.lexer + + # Set up the lexer and parser objects on pslice + pslice.lexer = lexer + pslice.parser = self + + # If input was supplied, pass to lexer + if input is not None: + lexer.input(input) + + if tokenfunc is None: + # Tokenize function + get_token = lexer.token + else: + get_token = tokenfunc + + # Set the parser() token method (sometimes used in error recovery) + self.token = get_token + + # Set up the state and symbol stacks + + statestack = [] # Stack of parsing states + self.statestack = statestack + symstack = [] # Stack of grammar symbols + self.symstack = symstack + + pslice.stack = symstack # Put in the production + errtoken = None # Err token + + # The start state is assumed to be (0,$end) + + statestack.append(0) + sym = YaccSymbol() + sym.type = '$end' + symstack.append(sym) + state = 0 + while True: + # Get the next symbol on the input. If a lookahead symbol + # is already set, we just use that. Otherwise, we'll pull + # the next token off of the lookaheadstack or from the lexer + + #--! DEBUG + debug.debug('') + debug.debug('State : %s', state) + #--! DEBUG + + if state not in defaulted_states: + if not lookahead: + if not lookaheadstack: + lookahead = get_token() # Get the next token + else: + lookahead = lookaheadstack.pop() + if not lookahead: + lookahead = YaccSymbol() + lookahead.type = '$end' + + # Check the action table + ltype = lookahead.type + t = actions[state].get(ltype) + else: + t = defaulted_states[state] + #--! DEBUG + debug.debug('Defaulted state %s: Reduce using %d', state, -t) + #--! DEBUG + + #--! DEBUG + debug.debug('Stack : %s', + ('%s . %s' % (' '.join([xx.type for xx in symstack][1:]), str(lookahead))).lstrip()) + #--! DEBUG + + if t is not None: + if t > 0: + # shift a symbol on the stack + statestack.append(t) + state = t + + #--! DEBUG + debug.debug('Action : Shift and goto state %s', t) + #--! DEBUG + + symstack.append(lookahead) + lookahead = None + + # Decrease error count on successful shift + if errorcount: + errorcount -= 1 + continue + + if t < 0: + # reduce a symbol on the stack, emit a production + p = prod[-t] + pname = p.name + plen = p.len + + # Get production function + sym = YaccSymbol() + sym.type = pname # Production name + sym.value = None + + #--! DEBUG + if plen: + debug.info('Action : Reduce rule [%s] with %s and goto state %d', p.str, + '['+','.join([format_stack_entry(_v.value) for _v in symstack[-plen:]])+']', + goto[statestack[-1-plen]][pname]) + else: + debug.info('Action : Reduce rule [%s] with %s and goto state %d', p.str, [], + goto[statestack[-1]][pname]) + + #--! DEBUG + + if plen: + targ = symstack[-plen-1:] + targ[0] = sym + + #--! TRACKING + if tracking: + t1 = targ[1] + sym.lineno = t1.lineno + sym.lexpos = t1.lexpos + t1 = targ[-1] + sym.endlineno = getattr(t1, 'endlineno', t1.lineno) + sym.endlexpos = getattr(t1, 'endlexpos', t1.lexpos) + #--! TRACKING + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # The code enclosed in this section is duplicated + # below as a performance optimization. Make sure + # changes get made in both locations. + + pslice.slice = targ + + try: + # Call the grammar rule with our special slice object + del symstack[-plen:] + self.state = state + p.callable(pslice) + del statestack[-plen:] + #--! DEBUG + debug.info('Result : %s', format_result(pslice[0])) + #--! DEBUG + symstack.append(sym) + state = goto[statestack[-1]][pname] + statestack.append(state) + except SyntaxError: + # If an error was set. Enter error recovery state + lookaheadstack.append(lookahead) # Save the current lookahead token + symstack.extend(targ[1:-1]) # Put the production slice back on the stack + statestack.pop() # Pop back one state (before the reduce) + state = statestack[-1] + sym.type = 'error' + sym.value = 'error' + lookahead = sym + errorcount = error_count + self.errorok = False + + continue + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + else: + + #--! TRACKING + if tracking: + sym.lineno = lexer.lineno + sym.lexpos = lexer.lexpos + #--! TRACKING + + targ = [sym] + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # The code enclosed in this section is duplicated + # above as a performance optimization. Make sure + # changes get made in both locations. + + pslice.slice = targ + + try: + # Call the grammar rule with our special slice object + self.state = state + p.callable(pslice) + #--! DEBUG + debug.info('Result : %s', format_result(pslice[0])) + #--! DEBUG + symstack.append(sym) + state = goto[statestack[-1]][pname] + statestack.append(state) + except SyntaxError: + # If an error was set. Enter error recovery state + lookaheadstack.append(lookahead) # Save the current lookahead token + statestack.pop() # Pop back one state (before the reduce) + state = statestack[-1] + sym.type = 'error' + sym.value = 'error' + lookahead = sym + errorcount = error_count + self.errorok = False + + continue + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + if t == 0: + n = symstack[-1] + result = getattr(n, 'value', None) + #--! DEBUG + debug.info('Done : Returning %s', format_result(result)) + debug.info('PLY: PARSE DEBUG END') + #--! DEBUG + return result + + if t is None: + + #--! DEBUG + debug.error('Error : %s', + ('%s . %s' % (' '.join([xx.type for xx in symstack][1:]), str(lookahead))).lstrip()) + #--! DEBUG + + # We have some kind of parsing error here. To handle + # this, we are going to push the current token onto + # the tokenstack and replace it with an 'error' token. + # If there are any synchronization rules, they may + # catch it. + # + # In addition to pushing the error token, we call call + # the user defined p_error() function if this is the + # first syntax error. This function is only called if + # errorcount == 0. + if errorcount == 0 or self.errorok: + errorcount = error_count + self.errorok = False + errtoken = lookahead + if errtoken.type == '$end': + errtoken = None # End of file! + if self.errorfunc: + if errtoken and not hasattr(errtoken, 'lexer'): + errtoken.lexer = lexer + self.state = state + tok = call_errorfunc(self.errorfunc, errtoken, self) + if self.errorok: + # User must have done some kind of panic + # mode recovery on their own. The + # returned token is the next lookahead + lookahead = tok + errtoken = None + continue + else: + if errtoken: + if hasattr(errtoken, 'lineno'): + lineno = lookahead.lineno + else: + lineno = 0 + if lineno: + sys.stderr.write('yacc: Syntax error at line %d, token=%s\n' % (lineno, errtoken.type)) + else: + sys.stderr.write('yacc: Syntax error, token=%s' % errtoken.type) + else: + sys.stderr.write('yacc: Parse error in input. EOF\n') + return + + else: + errorcount = error_count + + # case 1: the statestack only has 1 entry on it. If we're in this state, the + # entire parse has been rolled back and we're completely hosed. The token is + # discarded and we just keep going. + + if len(statestack) <= 1 and lookahead.type != '$end': + lookahead = None + errtoken = None + state = 0 + # Nuke the pushback stack + del lookaheadstack[:] + continue + + # case 2: the statestack has a couple of entries on it, but we're + # at the end of the file. nuke the top entry and generate an error token + + # Start nuking entries on the stack + if lookahead.type == '$end': + # Whoa. We're really hosed here. Bail out + return + + if lookahead.type != 'error': + sym = symstack[-1] + if sym.type == 'error': + # Hmmm. Error is on top of stack, we'll just nuke input + # symbol and continue + #--! TRACKING + if tracking: + sym.endlineno = getattr(lookahead, 'lineno', sym.lineno) + sym.endlexpos = getattr(lookahead, 'lexpos', sym.lexpos) + #--! TRACKING + lookahead = None + continue + + # Create the error symbol for the first time and make it the new lookahead symbol + t = YaccSymbol() + t.type = 'error' + + if hasattr(lookahead, 'lineno'): + t.lineno = t.endlineno = lookahead.lineno + if hasattr(lookahead, 'lexpos'): + t.lexpos = t.endlexpos = lookahead.lexpos + t.value = lookahead + lookaheadstack.append(lookahead) + lookahead = t + else: + sym = symstack.pop() + #--! TRACKING + if tracking: + lookahead.lineno = sym.lineno + lookahead.lexpos = sym.lexpos + #--! TRACKING + statestack.pop() + state = statestack[-1] + + continue + + # Call an error function here + raise RuntimeError('yacc: internal parser error!!!\n') + + #--! parsedebug-end + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # parseopt(). + # + # Optimized version of parse() method. DO NOT EDIT THIS CODE DIRECTLY! + # This code is automatically generated by the ply/ygen.py script. Make + # changes to the parsedebug() method instead. + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + def parseopt(self, input=None, lexer=None, debug=False, tracking=False, tokenfunc=None): + #--! parseopt-start + lookahead = None # Current lookahead symbol + lookaheadstack = [] # Stack of lookahead symbols + actions = self.action # Local reference to action table (to avoid lookup on self.) + goto = self.goto # Local reference to goto table (to avoid lookup on self.) + prod = self.productions # Local reference to production list (to avoid lookup on self.) + defaulted_states = self.defaulted_states # Local reference to defaulted states + pslice = YaccProduction(None) # Production object passed to grammar rules + errorcount = 0 # Used during error recovery + + + # If no lexer was given, we will try to use the lex module + if not lexer: + from . import lex + lexer = lex.lexer + + # Set up the lexer and parser objects on pslice + pslice.lexer = lexer + pslice.parser = self + + # If input was supplied, pass to lexer + if input is not None: + lexer.input(input) + + if tokenfunc is None: + # Tokenize function + get_token = lexer.token + else: + get_token = tokenfunc + + # Set the parser() token method (sometimes used in error recovery) + self.token = get_token + + # Set up the state and symbol stacks + + statestack = [] # Stack of parsing states + self.statestack = statestack + symstack = [] # Stack of grammar symbols + self.symstack = symstack + + pslice.stack = symstack # Put in the production + errtoken = None # Err token + + # The start state is assumed to be (0,$end) + + statestack.append(0) + sym = YaccSymbol() + sym.type = '$end' + symstack.append(sym) + state = 0 + while True: + # Get the next symbol on the input. If a lookahead symbol + # is already set, we just use that. Otherwise, we'll pull + # the next token off of the lookaheadstack or from the lexer + + + if state not in defaulted_states: + if not lookahead: + if not lookaheadstack: + lookahead = get_token() # Get the next token + else: + lookahead = lookaheadstack.pop() + if not lookahead: + lookahead = YaccSymbol() + lookahead.type = '$end' + + # Check the action table + ltype = lookahead.type + t = actions[state].get(ltype) + else: + t = defaulted_states[state] + + + if t is not None: + if t > 0: + # shift a symbol on the stack + statestack.append(t) + state = t + + + symstack.append(lookahead) + lookahead = None + + # Decrease error count on successful shift + if errorcount: + errorcount -= 1 + continue + + if t < 0: + # reduce a symbol on the stack, emit a production + p = prod[-t] + pname = p.name + plen = p.len + + # Get production function + sym = YaccSymbol() + sym.type = pname # Production name + sym.value = None + + + if plen: + targ = symstack[-plen-1:] + targ[0] = sym + + #--! TRACKING + if tracking: + t1 = targ[1] + sym.lineno = t1.lineno + sym.lexpos = t1.lexpos + t1 = targ[-1] + sym.endlineno = getattr(t1, 'endlineno', t1.lineno) + sym.endlexpos = getattr(t1, 'endlexpos', t1.lexpos) + #--! TRACKING + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # The code enclosed in this section is duplicated + # below as a performance optimization. Make sure + # changes get made in both locations. + + pslice.slice = targ + + try: + # Call the grammar rule with our special slice object + del symstack[-plen:] + self.state = state + p.callable(pslice) + del statestack[-plen:] + symstack.append(sym) + state = goto[statestack[-1]][pname] + statestack.append(state) + except SyntaxError: + # If an error was set. Enter error recovery state + lookaheadstack.append(lookahead) # Save the current lookahead token + symstack.extend(targ[1:-1]) # Put the production slice back on the stack + statestack.pop() # Pop back one state (before the reduce) + state = statestack[-1] + sym.type = 'error' + sym.value = 'error' + lookahead = sym + errorcount = error_count + self.errorok = False + + continue + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + else: + + #--! TRACKING + if tracking: + sym.lineno = lexer.lineno + sym.lexpos = lexer.lexpos + #--! TRACKING + + targ = [sym] + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # The code enclosed in this section is duplicated + # above as a performance optimization. Make sure + # changes get made in both locations. + + pslice.slice = targ + + try: + # Call the grammar rule with our special slice object + self.state = state + p.callable(pslice) + symstack.append(sym) + state = goto[statestack[-1]][pname] + statestack.append(state) + except SyntaxError: + # If an error was set. Enter error recovery state + lookaheadstack.append(lookahead) # Save the current lookahead token + statestack.pop() # Pop back one state (before the reduce) + state = statestack[-1] + sym.type = 'error' + sym.value = 'error' + lookahead = sym + errorcount = error_count + self.errorok = False + + continue + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + if t == 0: + n = symstack[-1] + result = getattr(n, 'value', None) + return result + + if t is None: + + + # We have some kind of parsing error here. To handle + # this, we are going to push the current token onto + # the tokenstack and replace it with an 'error' token. + # If there are any synchronization rules, they may + # catch it. + # + # In addition to pushing the error token, we call call + # the user defined p_error() function if this is the + # first syntax error. This function is only called if + # errorcount == 0. + if errorcount == 0 or self.errorok: + errorcount = error_count + self.errorok = False + errtoken = lookahead + if errtoken.type == '$end': + errtoken = None # End of file! + if self.errorfunc: + if errtoken and not hasattr(errtoken, 'lexer'): + errtoken.lexer = lexer + self.state = state + tok = call_errorfunc(self.errorfunc, errtoken, self) + if self.errorok: + # User must have done some kind of panic + # mode recovery on their own. The + # returned token is the next lookahead + lookahead = tok + errtoken = None + continue + else: + if errtoken: + if hasattr(errtoken, 'lineno'): + lineno = lookahead.lineno + else: + lineno = 0 + if lineno: + sys.stderr.write('yacc: Syntax error at line %d, token=%s\n' % (lineno, errtoken.type)) + else: + sys.stderr.write('yacc: Syntax error, token=%s' % errtoken.type) + else: + sys.stderr.write('yacc: Parse error in input. EOF\n') + return + + else: + errorcount = error_count + + # case 1: the statestack only has 1 entry on it. If we're in this state, the + # entire parse has been rolled back and we're completely hosed. The token is + # discarded and we just keep going. + + if len(statestack) <= 1 and lookahead.type != '$end': + lookahead = None + errtoken = None + state = 0 + # Nuke the pushback stack + del lookaheadstack[:] + continue + + # case 2: the statestack has a couple of entries on it, but we're + # at the end of the file. nuke the top entry and generate an error token + + # Start nuking entries on the stack + if lookahead.type == '$end': + # Whoa. We're really hosed here. Bail out + return + + if lookahead.type != 'error': + sym = symstack[-1] + if sym.type == 'error': + # Hmmm. Error is on top of stack, we'll just nuke input + # symbol and continue + #--! TRACKING + if tracking: + sym.endlineno = getattr(lookahead, 'lineno', sym.lineno) + sym.endlexpos = getattr(lookahead, 'lexpos', sym.lexpos) + #--! TRACKING + lookahead = None + continue + + # Create the error symbol for the first time and make it the new lookahead symbol + t = YaccSymbol() + t.type = 'error' + + if hasattr(lookahead, 'lineno'): + t.lineno = t.endlineno = lookahead.lineno + if hasattr(lookahead, 'lexpos'): + t.lexpos = t.endlexpos = lookahead.lexpos + t.value = lookahead + lookaheadstack.append(lookahead) + lookahead = t + else: + sym = symstack.pop() + #--! TRACKING + if tracking: + lookahead.lineno = sym.lineno + lookahead.lexpos = sym.lexpos + #--! TRACKING + statestack.pop() + state = statestack[-1] + + continue + + # Call an error function here + raise RuntimeError('yacc: internal parser error!!!\n') + + #--! parseopt-end + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # parseopt_notrack(). + # + # Optimized version of parseopt() with line number tracking removed. + # DO NOT EDIT THIS CODE DIRECTLY. This code is automatically generated + # by the ply/ygen.py script. Make changes to the parsedebug() method instead. + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + def parseopt_notrack(self, input=None, lexer=None, debug=False, tracking=False, tokenfunc=None): + #--! parseopt-notrack-start + lookahead = None # Current lookahead symbol + lookaheadstack = [] # Stack of lookahead symbols + actions = self.action # Local reference to action table (to avoid lookup on self.) + goto = self.goto # Local reference to goto table (to avoid lookup on self.) + prod = self.productions # Local reference to production list (to avoid lookup on self.) + defaulted_states = self.defaulted_states # Local reference to defaulted states + pslice = YaccProduction(None) # Production object passed to grammar rules + errorcount = 0 # Used during error recovery + + + # If no lexer was given, we will try to use the lex module + if not lexer: + from . import lex + lexer = lex.lexer + + # Set up the lexer and parser objects on pslice + pslice.lexer = lexer + pslice.parser = self + + # If input was supplied, pass to lexer + if input is not None: + lexer.input(input) + + if tokenfunc is None: + # Tokenize function + get_token = lexer.token + else: + get_token = tokenfunc + + # Set the parser() token method (sometimes used in error recovery) + self.token = get_token + + # Set up the state and symbol stacks + + statestack = [] # Stack of parsing states + self.statestack = statestack + symstack = [] # Stack of grammar symbols + self.symstack = symstack + + pslice.stack = symstack # Put in the production + errtoken = None # Err token + + # The start state is assumed to be (0,$end) + + statestack.append(0) + sym = YaccSymbol() + sym.type = '$end' + symstack.append(sym) + state = 0 + while True: + # Get the next symbol on the input. If a lookahead symbol + # is already set, we just use that. Otherwise, we'll pull + # the next token off of the lookaheadstack or from the lexer + + + if state not in defaulted_states: + if not lookahead: + if not lookaheadstack: + lookahead = get_token() # Get the next token + else: + lookahead = lookaheadstack.pop() + if not lookahead: + lookahead = YaccSymbol() + lookahead.type = '$end' + + # Check the action table + ltype = lookahead.type + t = actions[state].get(ltype) + else: + t = defaulted_states[state] + + + if t is not None: + if t > 0: + # shift a symbol on the stack + statestack.append(t) + state = t + + + symstack.append(lookahead) + lookahead = None + + # Decrease error count on successful shift + if errorcount: + errorcount -= 1 + continue + + if t < 0: + # reduce a symbol on the stack, emit a production + p = prod[-t] + pname = p.name + plen = p.len + + # Get production function + sym = YaccSymbol() + sym.type = pname # Production name + sym.value = None + + + if plen: + targ = symstack[-plen-1:] + targ[0] = sym + + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # The code enclosed in this section is duplicated + # below as a performance optimization. Make sure + # changes get made in both locations. + + pslice.slice = targ + + try: + # Call the grammar rule with our special slice object + del symstack[-plen:] + self.state = state + p.callable(pslice) + del statestack[-plen:] + symstack.append(sym) + state = goto[statestack[-1]][pname] + statestack.append(state) + except SyntaxError: + # If an error was set. Enter error recovery state + lookaheadstack.append(lookahead) # Save the current lookahead token + symstack.extend(targ[1:-1]) # Put the production slice back on the stack + statestack.pop() # Pop back one state (before the reduce) + state = statestack[-1] + sym.type = 'error' + sym.value = 'error' + lookahead = sym + errorcount = error_count + self.errorok = False + + continue + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + else: + + + targ = [sym] + + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + # The code enclosed in this section is duplicated + # above as a performance optimization. Make sure + # changes get made in both locations. + + pslice.slice = targ + + try: + # Call the grammar rule with our special slice object + self.state = state + p.callable(pslice) + symstack.append(sym) + state = goto[statestack[-1]][pname] + statestack.append(state) + except SyntaxError: + # If an error was set. Enter error recovery state + lookaheadstack.append(lookahead) # Save the current lookahead token + statestack.pop() # Pop back one state (before the reduce) + state = statestack[-1] + sym.type = 'error' + sym.value = 'error' + lookahead = sym + errorcount = error_count + self.errorok = False + + continue + # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! + + if t == 0: + n = symstack[-1] + result = getattr(n, 'value', None) + return result + + if t is None: + + + # We have some kind of parsing error here. To handle + # this, we are going to push the current token onto + # the tokenstack and replace it with an 'error' token. + # If there are any synchronization rules, they may + # catch it. + # + # In addition to pushing the error token, we call call + # the user defined p_error() function if this is the + # first syntax error. This function is only called if + # errorcount == 0. + if errorcount == 0 or self.errorok: + errorcount = error_count + self.errorok = False + errtoken = lookahead + if errtoken.type == '$end': + errtoken = None # End of file! + if self.errorfunc: + if errtoken and not hasattr(errtoken, 'lexer'): + errtoken.lexer = lexer + self.state = state + tok = call_errorfunc(self.errorfunc, errtoken, self) + if self.errorok: + # User must have done some kind of panic + # mode recovery on their own. The + # returned token is the next lookahead + lookahead = tok + errtoken = None + continue + else: + if errtoken: + if hasattr(errtoken, 'lineno'): + lineno = lookahead.lineno + else: + lineno = 0 + if lineno: + sys.stderr.write('yacc: Syntax error at line %d, token=%s\n' % (lineno, errtoken.type)) + else: + sys.stderr.write('yacc: Syntax error, token=%s' % errtoken.type) + else: + sys.stderr.write('yacc: Parse error in input. EOF\n') + return + + else: + errorcount = error_count + + # case 1: the statestack only has 1 entry on it. If we're in this state, the + # entire parse has been rolled back and we're completely hosed. The token is + # discarded and we just keep going. + + if len(statestack) <= 1 and lookahead.type != '$end': + lookahead = None + errtoken = None + state = 0 + # Nuke the pushback stack + del lookaheadstack[:] + continue + + # case 2: the statestack has a couple of entries on it, but we're + # at the end of the file. nuke the top entry and generate an error token + + # Start nuking entries on the stack + if lookahead.type == '$end': + # Whoa. We're really hosed here. Bail out + return + + if lookahead.type != 'error': + sym = symstack[-1] + if sym.type == 'error': + # Hmmm. Error is on top of stack, we'll just nuke input + # symbol and continue + lookahead = None + continue + + # Create the error symbol for the first time and make it the new lookahead symbol + t = YaccSymbol() + t.type = 'error' + + if hasattr(lookahead, 'lineno'): + t.lineno = t.endlineno = lookahead.lineno + if hasattr(lookahead, 'lexpos'): + t.lexpos = t.endlexpos = lookahead.lexpos + t.value = lookahead + lookaheadstack.append(lookahead) + lookahead = t + else: + sym = symstack.pop() + statestack.pop() + state = statestack[-1] + + continue + + # Call an error function here + raise RuntimeError('yacc: internal parser error!!!\n') + + #--! parseopt-notrack-end + +# ----------------------------------------------------------------------------- +# === Grammar Representation === +# +# The following functions, classes, and variables are used to represent and +# manipulate the rules that make up a grammar. +# ----------------------------------------------------------------------------- + +# regex matching identifiers +_is_identifier = re.compile(r'^[a-zA-Z0-9_-]+$') + +# ----------------------------------------------------------------------------- +# class Production: +# +# This class stores the raw information about a single production or grammar rule. +# A grammar rule refers to a specification such as this: +# +# expr : expr PLUS term +# +# Here are the basic attributes defined on all productions +# +# name - Name of the production. For example 'expr' +# prod - A list of symbols on the right side ['expr','PLUS','term'] +# prec - Production precedence level +# number - Production number. +# func - Function that executes on reduce +# file - File where production function is defined +# lineno - Line number where production function is defined +# +# The following attributes are defined or optional. +# +# len - Length of the production (number of symbols on right hand side) +# usyms - Set of unique symbols found in the production +# ----------------------------------------------------------------------------- + +class Production(object): + reduced = 0 + def __init__(self, number, name, prod, precedence=('right', 0), func=None, file='', line=0): + self.name = name + self.prod = tuple(prod) + self.number = number + self.func = func + self.callable = None + self.file = file + self.line = line + self.prec = precedence + + # Internal settings used during table construction + + self.len = len(self.prod) # Length of the production + + # Create a list of unique production symbols used in the production + self.usyms = [] + for s in self.prod: + if s not in self.usyms: + self.usyms.append(s) + + # List of all LR items for the production + self.lr_items = [] + self.lr_next = None + + # Create a string representation + if self.prod: + self.str = '%s -> %s' % (self.name, ' '.join(self.prod)) + else: + self.str = '%s -> <empty>' % self.name + + def __str__(self): + return self.str + + def __repr__(self): + return 'Production(' + str(self) + ')' + + def __len__(self): + return len(self.prod) + + def __nonzero__(self): + return 1 + + def __getitem__(self, index): + return self.prod[index] + + # Return the nth lr_item from the production (or None if at the end) + def lr_item(self, n): + if n > len(self.prod): + return None + p = LRItem(self, n) + # Precompute the list of productions immediately following. + try: + p.lr_after = Prodnames[p.prod[n+1]] + except (IndexError, KeyError): + p.lr_after = [] + try: + p.lr_before = p.prod[n-1] + except IndexError: + p.lr_before = None + return p + + # Bind the production function name to a callable + def bind(self, pdict): + if self.func: + self.callable = pdict[self.func] + +# This class serves as a minimal standin for Production objects when +# reading table data from files. It only contains information +# actually used by the LR parsing engine, plus some additional +# debugging information. +class MiniProduction(object): + def __init__(self, str, name, len, func, file, line): + self.name = name + self.len = len + self.func = func + self.callable = None + self.file = file + self.line = line + self.str = str + + def __str__(self): + return self.str + + def __repr__(self): + return 'MiniProduction(%s)' % self.str + + # Bind the production function name to a callable + def bind(self, pdict): + if self.func: + self.callable = pdict[self.func] + + +# ----------------------------------------------------------------------------- +# class LRItem +# +# This class represents a specific stage of parsing a production rule. For +# example: +# +# expr : expr . PLUS term +# +# In the above, the "." represents the current location of the parse. Here +# basic attributes: +# +# name - Name of the production. For example 'expr' +# prod - A list of symbols on the right side ['expr','.', 'PLUS','term'] +# number - Production number. +# +# lr_next Next LR item. Example, if we are ' expr -> expr . PLUS term' +# then lr_next refers to 'expr -> expr PLUS . term' +# lr_index - LR item index (location of the ".") in the prod list. +# lookaheads - LALR lookahead symbols for this item +# len - Length of the production (number of symbols on right hand side) +# lr_after - List of all productions that immediately follow +# lr_before - Grammar symbol immediately before +# ----------------------------------------------------------------------------- + +class LRItem(object): + def __init__(self, p, n): + self.name = p.name + self.prod = list(p.prod) + self.number = p.number + self.lr_index = n + self.lookaheads = {} + self.prod.insert(n, '.') + self.prod = tuple(self.prod) + self.len = len(self.prod) + self.usyms = p.usyms + + def __str__(self): + if self.prod: + s = '%s -> %s' % (self.name, ' '.join(self.prod)) + else: + s = '%s -> <empty>' % self.name + return s + + def __repr__(self): + return 'LRItem(' + str(self) + ')' + +# ----------------------------------------------------------------------------- +# rightmost_terminal() +# +# Return the rightmost terminal from a list of symbols. Used in add_production() +# ----------------------------------------------------------------------------- +def rightmost_terminal(symbols, terminals): + i = len(symbols) - 1 + while i >= 0: + if symbols[i] in terminals: + return symbols[i] + i -= 1 + return None + +# ----------------------------------------------------------------------------- +# === GRAMMAR CLASS === +# +# The following class represents the contents of the specified grammar along +# with various computed properties such as first sets, follow sets, LR items, etc. +# This data is used for critical parts of the table generation process later. +# ----------------------------------------------------------------------------- + +class GrammarError(YaccError): + pass + +class Grammar(object): + def __init__(self, terminals): + self.Productions = [None] # A list of all of the productions. The first + # entry is always reserved for the purpose of + # building an augmented grammar + + self.Prodnames = {} # A dictionary mapping the names of nonterminals to a list of all + # productions of that nonterminal. + + self.Prodmap = {} # A dictionary that is only used to detect duplicate + # productions. + + self.Terminals = {} # A dictionary mapping the names of terminal symbols to a + # list of the rules where they are used. + + for term in terminals: + self.Terminals[term] = [] + + self.Terminals['error'] = [] + + self.Nonterminals = {} # A dictionary mapping names of nonterminals to a list + # of rule numbers where they are used. + + self.First = {} # A dictionary of precomputed FIRST(x) symbols + + self.Follow = {} # A dictionary of precomputed FOLLOW(x) symbols + + self.Precedence = {} # Precedence rules for each terminal. Contains tuples of the + # form ('right',level) or ('nonassoc', level) or ('left',level) + + self.UsedPrecedence = set() # Precedence rules that were actually used by the grammer. + # This is only used to provide error checking and to generate + # a warning about unused precedence rules. + + self.Start = None # Starting symbol for the grammar + + + def __len__(self): + return len(self.Productions) + + def __getitem__(self, index): + return self.Productions[index] + + # ----------------------------------------------------------------------------- + # set_precedence() + # + # Sets the precedence for a given terminal. assoc is the associativity such as + # 'left','right', or 'nonassoc'. level is a numeric level. + # + # ----------------------------------------------------------------------------- + + def set_precedence(self, term, assoc, level): + assert self.Productions == [None], 'Must call set_precedence() before add_production()' + if term in self.Precedence: + raise GrammarError('Precedence already specified for terminal %r' % term) + if assoc not in ['left', 'right', 'nonassoc']: + raise GrammarError("Associativity must be one of 'left','right', or 'nonassoc'") + self.Precedence[term] = (assoc, level) + + # ----------------------------------------------------------------------------- + # add_production() + # + # Given an action function, this function assembles a production rule and + # computes its precedence level. + # + # The production rule is supplied as a list of symbols. For example, + # a rule such as 'expr : expr PLUS term' has a production name of 'expr' and + # symbols ['expr','PLUS','term']. + # + # Precedence is determined by the precedence of the right-most non-terminal + # or the precedence of a terminal specified by %prec. + # + # A variety of error checks are performed to make sure production symbols + # are valid and that %prec is used correctly. + # ----------------------------------------------------------------------------- + + def add_production(self, prodname, syms, func=None, file='', line=0): + + if prodname in self.Terminals: + raise GrammarError('%s:%d: Illegal rule name %r. Already defined as a token' % (file, line, prodname)) + if prodname == 'error': + raise GrammarError('%s:%d: Illegal rule name %r. error is a reserved word' % (file, line, prodname)) + if not _is_identifier.match(prodname): + raise GrammarError('%s:%d: Illegal rule name %r' % (file, line, prodname)) + + # Look for literal tokens + for n, s in enumerate(syms): + if s[0] in "'\"": + try: + c = eval(s) + if (len(c) > 1): + raise GrammarError('%s:%d: Literal token %s in rule %r may only be a single character' % + (file, line, s, prodname)) + if c not in self.Terminals: + self.Terminals[c] = [] + syms[n] = c + continue + except SyntaxError: + pass + if not _is_identifier.match(s) and s != '%prec': + raise GrammarError('%s:%d: Illegal name %r in rule %r' % (file, line, s, prodname)) + + # Determine the precedence level + if '%prec' in syms: + if syms[-1] == '%prec': + raise GrammarError('%s:%d: Syntax error. Nothing follows %%prec' % (file, line)) + if syms[-2] != '%prec': + raise GrammarError('%s:%d: Syntax error. %%prec can only appear at the end of a grammar rule' % + (file, line)) + precname = syms[-1] + prodprec = self.Precedence.get(precname) + if not prodprec: + raise GrammarError('%s:%d: Nothing known about the precedence of %r' % (file, line, precname)) + else: + self.UsedPrecedence.add(precname) + del syms[-2:] # Drop %prec from the rule + else: + # If no %prec, precedence is determined by the rightmost terminal symbol + precname = rightmost_terminal(syms, self.Terminals) + prodprec = self.Precedence.get(precname, ('right', 0)) + + # See if the rule is already in the rulemap + map = '%s -> %s' % (prodname, syms) + if map in self.Prodmap: + m = self.Prodmap[map] + raise GrammarError('%s:%d: Duplicate rule %s. ' % (file, line, m) + + 'Previous definition at %s:%d' % (m.file, m.line)) + + # From this point on, everything is valid. Create a new Production instance + pnumber = len(self.Productions) + if prodname not in self.Nonterminals: + self.Nonterminals[prodname] = [] + + # Add the production number to Terminals and Nonterminals + for t in syms: + if t in self.Terminals: + self.Terminals[t].append(pnumber) + else: + if t not in self.Nonterminals: + self.Nonterminals[t] = [] + self.Nonterminals[t].append(pnumber) + + # Create a production and add it to the list of productions + p = Production(pnumber, prodname, syms, prodprec, func, file, line) + self.Productions.append(p) + self.Prodmap[map] = p + + # Add to the global productions list + try: + self.Prodnames[prodname].append(p) + except KeyError: + self.Prodnames[prodname] = [p] + + # ----------------------------------------------------------------------------- + # set_start() + # + # Sets the starting symbol and creates the augmented grammar. Production + # rule 0 is S' -> start where start is the start symbol. + # ----------------------------------------------------------------------------- + + def set_start(self, start=None): + if not start: + start = self.Productions[1].name + if start not in self.Nonterminals: + raise GrammarError('start symbol %s undefined' % start) + self.Productions[0] = Production(0, "S'", [start]) + self.Nonterminals[start].append(0) + self.Start = start + + # ----------------------------------------------------------------------------- + # find_unreachable() + # + # Find all of the nonterminal symbols that can't be reached from the starting + # symbol. Returns a list of nonterminals that can't be reached. + # ----------------------------------------------------------------------------- + + def find_unreachable(self): + + # Mark all symbols that are reachable from a symbol s + def mark_reachable_from(s): + if s in reachable: + return + reachable.add(s) + for p in self.Prodnames.get(s, []): + for r in p.prod: + mark_reachable_from(r) + + reachable = set() + mark_reachable_from(self.Productions[0].prod[0]) + return [s for s in self.Nonterminals if s not in reachable] + + # ----------------------------------------------------------------------------- + # infinite_cycles() + # + # This function looks at the various parsing rules and tries to detect + # infinite recursion cycles (grammar rules where there is no possible way + # to derive a string of only terminals). + # ----------------------------------------------------------------------------- + + def infinite_cycles(self): + terminates = {} + + # Terminals: + for t in self.Terminals: + terminates[t] = True + + terminates['$end'] = True + + # Nonterminals: + + # Initialize to false: + for n in self.Nonterminals: + terminates[n] = False + + # Then propagate termination until no change: + while True: + some_change = False + for (n, pl) in self.Prodnames.items(): + # Nonterminal n terminates iff any of its productions terminates. + for p in pl: + # Production p terminates iff all of its rhs symbols terminate. + for s in p.prod: + if not terminates[s]: + # The symbol s does not terminate, + # so production p does not terminate. + p_terminates = False + break + else: + # didn't break from the loop, + # so every symbol s terminates + # so production p terminates. + p_terminates = True + + if p_terminates: + # symbol n terminates! + if not terminates[n]: + terminates[n] = True + some_change = True + # Don't need to consider any more productions for this n. + break + + if not some_change: + break + + infinite = [] + for (s, term) in terminates.items(): + if not term: + if s not in self.Prodnames and s not in self.Terminals and s != 'error': + # s is used-but-not-defined, and we've already warned of that, + # so it would be overkill to say that it's also non-terminating. + pass + else: + infinite.append(s) + + return infinite + + # ----------------------------------------------------------------------------- + # undefined_symbols() + # + # Find all symbols that were used the grammar, but not defined as tokens or + # grammar rules. Returns a list of tuples (sym, prod) where sym in the symbol + # and prod is the production where the symbol was used. + # ----------------------------------------------------------------------------- + def undefined_symbols(self): + result = [] + for p in self.Productions: + if not p: + continue + + for s in p.prod: + if s not in self.Prodnames and s not in self.Terminals and s != 'error': + result.append((s, p)) + return result + + # ----------------------------------------------------------------------------- + # unused_terminals() + # + # Find all terminals that were defined, but not used by the grammar. Returns + # a list of all symbols. + # ----------------------------------------------------------------------------- + def unused_terminals(self): + unused_tok = [] + for s, v in self.Terminals.items(): + if s != 'error' and not v: + unused_tok.append(s) + + return unused_tok + + # ------------------------------------------------------------------------------ + # unused_rules() + # + # Find all grammar rules that were defined, but not used (maybe not reachable) + # Returns a list of productions. + # ------------------------------------------------------------------------------ + + def unused_rules(self): + unused_prod = [] + for s, v in self.Nonterminals.items(): + if not v: + p = self.Prodnames[s][0] + unused_prod.append(p) + return unused_prod + + # ----------------------------------------------------------------------------- + # unused_precedence() + # + # Returns a list of tuples (term,precedence) corresponding to precedence + # rules that were never used by the grammar. term is the name of the terminal + # on which precedence was applied and precedence is a string such as 'left' or + # 'right' corresponding to the type of precedence. + # ----------------------------------------------------------------------------- + + def unused_precedence(self): + unused = [] + for termname in self.Precedence: + if not (termname in self.Terminals or termname in self.UsedPrecedence): + unused.append((termname, self.Precedence[termname][0])) + + return unused + + # ------------------------------------------------------------------------- + # _first() + # + # Compute the value of FIRST1(beta) where beta is a tuple of symbols. + # + # During execution of compute_first1, the result may be incomplete. + # Afterward (e.g., when called from compute_follow()), it will be complete. + # ------------------------------------------------------------------------- + def _first(self, beta): + + # We are computing First(x1,x2,x3,...,xn) + result = [] + for x in beta: + x_produces_empty = False + + # Add all the non-<empty> symbols of First[x] to the result. + for f in self.First[x]: + if f == '<empty>': + x_produces_empty = True + else: + if f not in result: + result.append(f) + + if x_produces_empty: + # We have to consider the next x in beta, + # i.e. stay in the loop. + pass + else: + # We don't have to consider any further symbols in beta. + break + else: + # There was no 'break' from the loop, + # so x_produces_empty was true for all x in beta, + # so beta produces empty as well. + result.append('<empty>') + + return result + + # ------------------------------------------------------------------------- + # compute_first() + # + # Compute the value of FIRST1(X) for all symbols + # ------------------------------------------------------------------------- + def compute_first(self): + if self.First: + return self.First + + # Terminals: + for t in self.Terminals: + self.First[t] = [t] + + self.First['$end'] = ['$end'] + + # Nonterminals: + + # Initialize to the empty set: + for n in self.Nonterminals: + self.First[n] = [] + + # Then propagate symbols until no change: + while True: + some_change = False + for n in self.Nonterminals: + for p in self.Prodnames[n]: + for f in self._first(p.prod): + if f not in self.First[n]: + self.First[n].append(f) + some_change = True + if not some_change: + break + + return self.First + + # --------------------------------------------------------------------- + # compute_follow() + # + # Computes all of the follow sets for every non-terminal symbol. The + # follow set is the set of all symbols that might follow a given + # non-terminal. See the Dragon book, 2nd Ed. p. 189. + # --------------------------------------------------------------------- + def compute_follow(self, start=None): + # If already computed, return the result + if self.Follow: + return self.Follow + + # If first sets not computed yet, do that first. + if not self.First: + self.compute_first() + + # Add '$end' to the follow list of the start symbol + for k in self.Nonterminals: + self.Follow[k] = [] + + if not start: + start = self.Productions[1].name + + self.Follow[start] = ['$end'] + + while True: + didadd = False + for p in self.Productions[1:]: + # Here is the production set + for i, B in enumerate(p.prod): + if B in self.Nonterminals: + # Okay. We got a non-terminal in a production + fst = self._first(p.prod[i+1:]) + hasempty = False + for f in fst: + if f != '<empty>' and f not in self.Follow[B]: + self.Follow[B].append(f) + didadd = True + if f == '<empty>': + hasempty = True + if hasempty or i == (len(p.prod)-1): + # Add elements of follow(a) to follow(b) + for f in self.Follow[p.name]: + if f not in self.Follow[B]: + self.Follow[B].append(f) + didadd = True + if not didadd: + break + return self.Follow + + + # ----------------------------------------------------------------------------- + # build_lritems() + # + # This function walks the list of productions and builds a complete set of the + # LR items. The LR items are stored in two ways: First, they are uniquely + # numbered and placed in the list _lritems. Second, a linked list of LR items + # is built for each production. For example: + # + # E -> E PLUS E + # + # Creates the list + # + # [E -> . E PLUS E, E -> E . PLUS E, E -> E PLUS . E, E -> E PLUS E . ] + # ----------------------------------------------------------------------------- + + def build_lritems(self): + for p in self.Productions: + lastlri = p + i = 0 + lr_items = [] + while True: + if i > len(p): + lri = None + else: + lri = LRItem(p, i) + # Precompute the list of productions immediately following + try: + lri.lr_after = self.Prodnames[lri.prod[i+1]] + except (IndexError, KeyError): + lri.lr_after = [] + try: + lri.lr_before = lri.prod[i-1] + except IndexError: + lri.lr_before = None + + lastlri.lr_next = lri + if not lri: + break + lr_items.append(lri) + lastlri = lri + i += 1 + p.lr_items = lr_items + +# ----------------------------------------------------------------------------- +# == Class LRTable == +# +# This basic class represents a basic table of LR parsing information. +# Methods for generating the tables are not defined here. They are defined +# in the derived class LRGeneratedTable. +# ----------------------------------------------------------------------------- + +class VersionError(YaccError): + pass + +class LRTable(object): + def __init__(self): + self.lr_action = None + self.lr_goto = None + self.lr_productions = None + self.lr_method = None + + def read_table(self, module): + if isinstance(module, types.ModuleType): + parsetab = module + else: + exec('import %s' % module) + parsetab = sys.modules[module] + + if parsetab._tabversion != __tabversion__: + raise VersionError('yacc table file version is out of date') + + self.lr_action = parsetab._lr_action + self.lr_goto = parsetab._lr_goto + + self.lr_productions = [] + for p in parsetab._lr_productions: + self.lr_productions.append(MiniProduction(*p)) + + self.lr_method = parsetab._lr_method + return parsetab._lr_signature + + def read_pickle(self, filename): + try: + import cPickle as pickle + except ImportError: + import pickle + + if not os.path.exists(filename): + raise ImportError + + in_f = open(filename, 'rb') + + tabversion = pickle.load(in_f) + if tabversion != __tabversion__: + raise VersionError('yacc table file version is out of date') + self.lr_method = pickle.load(in_f) + signature = pickle.load(in_f) + self.lr_action = pickle.load(in_f) + self.lr_goto = pickle.load(in_f) + productions = pickle.load(in_f) + + self.lr_productions = [] + for p in productions: + self.lr_productions.append(MiniProduction(*p)) + + in_f.close() + return signature + + # Bind all production function names to callable objects in pdict + def bind_callables(self, pdict): + for p in self.lr_productions: + p.bind(pdict) + + +# ----------------------------------------------------------------------------- +# === LR Generator === +# +# The following classes and functions are used to generate LR parsing tables on +# a grammar. +# ----------------------------------------------------------------------------- + +# ----------------------------------------------------------------------------- +# digraph() +# traverse() +# +# The following two functions are used to compute set valued functions +# of the form: +# +# F(x) = F'(x) U U{F(y) | x R y} +# +# This is used to compute the values of Read() sets as well as FOLLOW sets +# in LALR(1) generation. +# +# Inputs: X - An input set +# R - A relation +# FP - Set-valued function +# ------------------------------------------------------------------------------ + +def digraph(X, R, FP): + N = {} + for x in X: + N[x] = 0 + stack = [] + F = {} + for x in X: + if N[x] == 0: + traverse(x, N, stack, F, X, R, FP) + return F + +def traverse(x, N, stack, F, X, R, FP): + stack.append(x) + d = len(stack) + N[x] = d + F[x] = FP(x) # F(X) <- F'(x) + + rel = R(x) # Get y's related to x + for y in rel: + if N[y] == 0: + traverse(y, N, stack, F, X, R, FP) + N[x] = min(N[x], N[y]) + for a in F.get(y, []): + if a not in F[x]: + F[x].append(a) + if N[x] == d: + N[stack[-1]] = MAXINT + F[stack[-1]] = F[x] + element = stack.pop() + while element != x: + N[stack[-1]] = MAXINT + F[stack[-1]] = F[x] + element = stack.pop() + +class LALRError(YaccError): + pass + +# ----------------------------------------------------------------------------- +# == LRGeneratedTable == +# +# This class implements the LR table generation algorithm. There are no +# public methods except for write() +# ----------------------------------------------------------------------------- + +class LRGeneratedTable(LRTable): + def __init__(self, grammar, method='LALR', log=None): + if method not in ['SLR', 'LALR']: + raise LALRError('Unsupported method %s' % method) + + self.grammar = grammar + self.lr_method = method + + # Set up the logger + if not log: + log = NullLogger() + self.log = log + + # Internal attributes + self.lr_action = {} # Action table + self.lr_goto = {} # Goto table + self.lr_productions = grammar.Productions # Copy of grammar Production array + self.lr_goto_cache = {} # Cache of computed gotos + self.lr0_cidhash = {} # Cache of closures + + self._add_count = 0 # Internal counter used to detect cycles + + # Diagonistic information filled in by the table generator + self.sr_conflict = 0 + self.rr_conflict = 0 + self.conflicts = [] # List of conflicts + + self.sr_conflicts = [] + self.rr_conflicts = [] + + # Build the tables + self.grammar.build_lritems() + self.grammar.compute_first() + self.grammar.compute_follow() + self.lr_parse_table() + + # Compute the LR(0) closure operation on I, where I is a set of LR(0) items. + + def lr0_closure(self, I): + self._add_count += 1 + + # Add everything in I to J + J = I[:] + didadd = True + while didadd: + didadd = False + for j in J: + for x in j.lr_after: + if getattr(x, 'lr0_added', 0) == self._add_count: + continue + # Add B --> .G to J + J.append(x.lr_next) + x.lr0_added = self._add_count + didadd = True + + return J + + # Compute the LR(0) goto function goto(I,X) where I is a set + # of LR(0) items and X is a grammar symbol. This function is written + # in a way that guarantees uniqueness of the generated goto sets + # (i.e. the same goto set will never be returned as two different Python + # objects). With uniqueness, we can later do fast set comparisons using + # id(obj) instead of element-wise comparison. + + def lr0_goto(self, I, x): + # First we look for a previously cached entry + g = self.lr_goto_cache.get((id(I), x)) + if g: + return g + + # Now we generate the goto set in a way that guarantees uniqueness + # of the result + + s = self.lr_goto_cache.get(x) + if not s: + s = {} + self.lr_goto_cache[x] = s + + gs = [] + for p in I: + n = p.lr_next + if n and n.lr_before == x: + s1 = s.get(id(n)) + if not s1: + s1 = {} + s[id(n)] = s1 + gs.append(n) + s = s1 + g = s.get('$end') + if not g: + if gs: + g = self.lr0_closure(gs) + s['$end'] = g + else: + s['$end'] = gs + self.lr_goto_cache[(id(I), x)] = g + return g + + # Compute the LR(0) sets of item function + def lr0_items(self): + C = [self.lr0_closure([self.grammar.Productions[0].lr_next])] + i = 0 + for I in C: + self.lr0_cidhash[id(I)] = i + i += 1 + + # Loop over the items in C and each grammar symbols + i = 0 + while i < len(C): + I = C[i] + i += 1 + + # Collect all of the symbols that could possibly be in the goto(I,X) sets + asyms = {} + for ii in I: + for s in ii.usyms: + asyms[s] = None + + for x in asyms: + g = self.lr0_goto(I, x) + if not g or id(g) in self.lr0_cidhash: + continue + self.lr0_cidhash[id(g)] = len(C) + C.append(g) + + return C + + # ----------------------------------------------------------------------------- + # ==== LALR(1) Parsing ==== + # + # LALR(1) parsing is almost exactly the same as SLR except that instead of + # relying upon Follow() sets when performing reductions, a more selective + # lookahead set that incorporates the state of the LR(0) machine is utilized. + # Thus, we mainly just have to focus on calculating the lookahead sets. + # + # The method used here is due to DeRemer and Pennelo (1982). + # + # DeRemer, F. L., and T. J. Pennelo: "Efficient Computation of LALR(1) + # Lookahead Sets", ACM Transactions on Programming Languages and Systems, + # Vol. 4, No. 4, Oct. 1982, pp. 615-649 + # + # Further details can also be found in: + # + # J. Tremblay and P. Sorenson, "The Theory and Practice of Compiler Writing", + # McGraw-Hill Book Company, (1985). + # + # ----------------------------------------------------------------------------- + + # ----------------------------------------------------------------------------- + # compute_nullable_nonterminals() + # + # Creates a dictionary containing all of the non-terminals that might produce + # an empty production. + # ----------------------------------------------------------------------------- + + def compute_nullable_nonterminals(self): + nullable = set() + num_nullable = 0 + while True: + for p in self.grammar.Productions[1:]: + if p.len == 0: + nullable.add(p.name) + continue + for t in p.prod: + if t not in nullable: + break + else: + nullable.add(p.name) + if len(nullable) == num_nullable: + break + num_nullable = len(nullable) + return nullable + + # ----------------------------------------------------------------------------- + # find_nonterminal_trans(C) + # + # Given a set of LR(0) items, this functions finds all of the non-terminal + # transitions. These are transitions in which a dot appears immediately before + # a non-terminal. Returns a list of tuples of the form (state,N) where state + # is the state number and N is the nonterminal symbol. + # + # The input C is the set of LR(0) items. + # ----------------------------------------------------------------------------- + + def find_nonterminal_transitions(self, C): + trans = [] + for stateno, state in enumerate(C): + for p in state: + if p.lr_index < p.len - 1: + t = (stateno, p.prod[p.lr_index+1]) + if t[1] in self.grammar.Nonterminals: + if t not in trans: + trans.append(t) + return trans + + # ----------------------------------------------------------------------------- + # dr_relation() + # + # Computes the DR(p,A) relationships for non-terminal transitions. The input + # is a tuple (state,N) where state is a number and N is a nonterminal symbol. + # + # Returns a list of terminals. + # ----------------------------------------------------------------------------- + + def dr_relation(self, C, trans, nullable): + dr_set = {} + state, N = trans + terms = [] + + g = self.lr0_goto(C[state], N) + for p in g: + if p.lr_index < p.len - 1: + a = p.prod[p.lr_index+1] + if a in self.grammar.Terminals: + if a not in terms: + terms.append(a) + + # This extra bit is to handle the start state + if state == 0 and N == self.grammar.Productions[0].prod[0]: + terms.append('$end') + + return terms + + # ----------------------------------------------------------------------------- + # reads_relation() + # + # Computes the READS() relation (p,A) READS (t,C). + # ----------------------------------------------------------------------------- + + def reads_relation(self, C, trans, empty): + # Look for empty transitions + rel = [] + state, N = trans + + g = self.lr0_goto(C[state], N) + j = self.lr0_cidhash.get(id(g), -1) + for p in g: + if p.lr_index < p.len - 1: + a = p.prod[p.lr_index + 1] + if a in empty: + rel.append((j, a)) + + return rel + + # ----------------------------------------------------------------------------- + # compute_lookback_includes() + # + # Determines the lookback and includes relations + # + # LOOKBACK: + # + # This relation is determined by running the LR(0) state machine forward. + # For example, starting with a production "N : . A B C", we run it forward + # to obtain "N : A B C ." We then build a relationship between this final + # state and the starting state. These relationships are stored in a dictionary + # lookdict. + # + # INCLUDES: + # + # Computes the INCLUDE() relation (p,A) INCLUDES (p',B). + # + # This relation is used to determine non-terminal transitions that occur + # inside of other non-terminal transition states. (p,A) INCLUDES (p', B) + # if the following holds: + # + # B -> LAT, where T -> epsilon and p' -L-> p + # + # L is essentially a prefix (which may be empty), T is a suffix that must be + # able to derive an empty string. State p' must lead to state p with the string L. + # + # ----------------------------------------------------------------------------- + + def compute_lookback_includes(self, C, trans, nullable): + lookdict = {} # Dictionary of lookback relations + includedict = {} # Dictionary of include relations + + # Make a dictionary of non-terminal transitions + dtrans = {} + for t in trans: + dtrans[t] = 1 + + # Loop over all transitions and compute lookbacks and includes + for state, N in trans: + lookb = [] + includes = [] + for p in C[state]: + if p.name != N: + continue + + # Okay, we have a name match. We now follow the production all the way + # through the state machine until we get the . on the right hand side + + lr_index = p.lr_index + j = state + while lr_index < p.len - 1: + lr_index = lr_index + 1 + t = p.prod[lr_index] + + # Check to see if this symbol and state are a non-terminal transition + if (j, t) in dtrans: + # Yes. Okay, there is some chance that this is an includes relation + # the only way to know for certain is whether the rest of the + # production derives empty + + li = lr_index + 1 + while li < p.len: + if p.prod[li] in self.grammar.Terminals: + break # No forget it + if p.prod[li] not in nullable: + break + li = li + 1 + else: + # Appears to be a relation between (j,t) and (state,N) + includes.append((j, t)) + + g = self.lr0_goto(C[j], t) # Go to next set + j = self.lr0_cidhash.get(id(g), -1) # Go to next state + + # When we get here, j is the final state, now we have to locate the production + for r in C[j]: + if r.name != p.name: + continue + if r.len != p.len: + continue + i = 0 + # This look is comparing a production ". A B C" with "A B C ." + while i < r.lr_index: + if r.prod[i] != p.prod[i+1]: + break + i = i + 1 + else: + lookb.append((j, r)) + for i in includes: + if i not in includedict: + includedict[i] = [] + includedict[i].append((state, N)) + lookdict[(state, N)] = lookb + + return lookdict, includedict + + # ----------------------------------------------------------------------------- + # compute_read_sets() + # + # Given a set of LR(0) items, this function computes the read sets. + # + # Inputs: C = Set of LR(0) items + # ntrans = Set of nonterminal transitions + # nullable = Set of empty transitions + # + # Returns a set containing the read sets + # ----------------------------------------------------------------------------- + + def compute_read_sets(self, C, ntrans, nullable): + FP = lambda x: self.dr_relation(C, x, nullable) + R = lambda x: self.reads_relation(C, x, nullable) + F = digraph(ntrans, R, FP) + return F + + # ----------------------------------------------------------------------------- + # compute_follow_sets() + # + # Given a set of LR(0) items, a set of non-terminal transitions, a readset, + # and an include set, this function computes the follow sets + # + # Follow(p,A) = Read(p,A) U U {Follow(p',B) | (p,A) INCLUDES (p',B)} + # + # Inputs: + # ntrans = Set of nonterminal transitions + # readsets = Readset (previously computed) + # inclsets = Include sets (previously computed) + # + # Returns a set containing the follow sets + # ----------------------------------------------------------------------------- + + def compute_follow_sets(self, ntrans, readsets, inclsets): + FP = lambda x: readsets[x] + R = lambda x: inclsets.get(x, []) + F = digraph(ntrans, R, FP) + return F + + # ----------------------------------------------------------------------------- + # add_lookaheads() + # + # Attaches the lookahead symbols to grammar rules. + # + # Inputs: lookbacks - Set of lookback relations + # followset - Computed follow set + # + # This function directly attaches the lookaheads to productions contained + # in the lookbacks set + # ----------------------------------------------------------------------------- + + def add_lookaheads(self, lookbacks, followset): + for trans, lb in lookbacks.items(): + # Loop over productions in lookback + for state, p in lb: + if state not in p.lookaheads: + p.lookaheads[state] = [] + f = followset.get(trans, []) + for a in f: + if a not in p.lookaheads[state]: + p.lookaheads[state].append(a) + + # ----------------------------------------------------------------------------- + # add_lalr_lookaheads() + # + # This function does all of the work of adding lookahead information for use + # with LALR parsing + # ----------------------------------------------------------------------------- + + def add_lalr_lookaheads(self, C): + # Determine all of the nullable nonterminals + nullable = self.compute_nullable_nonterminals() + + # Find all non-terminal transitions + trans = self.find_nonterminal_transitions(C) + + # Compute read sets + readsets = self.compute_read_sets(C, trans, nullable) + + # Compute lookback/includes relations + lookd, included = self.compute_lookback_includes(C, trans, nullable) + + # Compute LALR FOLLOW sets + followsets = self.compute_follow_sets(trans, readsets, included) + + # Add all of the lookaheads + self.add_lookaheads(lookd, followsets) + + # ----------------------------------------------------------------------------- + # lr_parse_table() + # + # This function constructs the parse tables for SLR or LALR + # ----------------------------------------------------------------------------- + def lr_parse_table(self): + Productions = self.grammar.Productions + Precedence = self.grammar.Precedence + goto = self.lr_goto # Goto array + action = self.lr_action # Action array + log = self.log # Logger for output + + actionp = {} # Action production array (temporary) + + log.info('Parsing method: %s', self.lr_method) + + # Step 1: Construct C = { I0, I1, ... IN}, collection of LR(0) items + # This determines the number of states + + C = self.lr0_items() + + if self.lr_method == 'LALR': + self.add_lalr_lookaheads(C) + + # Build the parser table, state by state + st = 0 + for I in C: + # Loop over each production in I + actlist = [] # List of actions + st_action = {} + st_actionp = {} + st_goto = {} + log.info('') + log.info('state %d', st) + log.info('') + for p in I: + log.info(' (%d) %s', p.number, p) + log.info('') + + for p in I: + if p.len == p.lr_index + 1: + if p.name == "S'": + # Start symbol. Accept! + st_action['$end'] = 0 + st_actionp['$end'] = p + else: + # We are at the end of a production. Reduce! + if self.lr_method == 'LALR': + laheads = p.lookaheads[st] + else: + laheads = self.grammar.Follow[p.name] + for a in laheads: + actlist.append((a, p, 'reduce using rule %d (%s)' % (p.number, p))) + r = st_action.get(a) + if r is not None: + # Whoa. Have a shift/reduce or reduce/reduce conflict + if r > 0: + # Need to decide on shift or reduce here + # By default we favor shifting. Need to add + # some precedence rules here. + + # Shift precedence comes from the token + sprec, slevel = Precedence.get(a, ('right', 0)) + + # Reduce precedence comes from rule being reduced (p) + rprec, rlevel = Productions[p.number].prec + + if (slevel < rlevel) or ((slevel == rlevel) and (rprec == 'left')): + # We really need to reduce here. + st_action[a] = -p.number + st_actionp[a] = p + if not slevel and not rlevel: + log.info(' ! shift/reduce conflict for %s resolved as reduce', a) + self.sr_conflicts.append((st, a, 'reduce')) + Productions[p.number].reduced += 1 + elif (slevel == rlevel) and (rprec == 'nonassoc'): + st_action[a] = None + else: + # Hmmm. Guess we'll keep the shift + if not rlevel: + log.info(' ! shift/reduce conflict for %s resolved as shift', a) + self.sr_conflicts.append((st, a, 'shift')) + elif r < 0: + # Reduce/reduce conflict. In this case, we favor the rule + # that was defined first in the grammar file + oldp = Productions[-r] + pp = Productions[p.number] + if oldp.line > pp.line: + st_action[a] = -p.number + st_actionp[a] = p + chosenp, rejectp = pp, oldp + Productions[p.number].reduced += 1 + Productions[oldp.number].reduced -= 1 + else: + chosenp, rejectp = oldp, pp + self.rr_conflicts.append((st, chosenp, rejectp)) + log.info(' ! reduce/reduce conflict for %s resolved using rule %d (%s)', + a, st_actionp[a].number, st_actionp[a]) + else: + raise LALRError('Unknown conflict in state %d' % st) + else: + st_action[a] = -p.number + st_actionp[a] = p + Productions[p.number].reduced += 1 + else: + i = p.lr_index + a = p.prod[i+1] # Get symbol right after the "." + if a in self.grammar.Terminals: + g = self.lr0_goto(I, a) + j = self.lr0_cidhash.get(id(g), -1) + if j >= 0: + # We are in a shift state + actlist.append((a, p, 'shift and go to state %d' % j)) + r = st_action.get(a) + if r is not None: + # Whoa have a shift/reduce or shift/shift conflict + if r > 0: + if r != j: + raise LALRError('Shift/shift conflict in state %d' % st) + elif r < 0: + # Do a precedence check. + # - if precedence of reduce rule is higher, we reduce. + # - if precedence of reduce is same and left assoc, we reduce. + # - otherwise we shift + + # Shift precedence comes from the token + sprec, slevel = Precedence.get(a, ('right', 0)) + + # Reduce precedence comes from the rule that could have been reduced + rprec, rlevel = Productions[st_actionp[a].number].prec + + if (slevel > rlevel) or ((slevel == rlevel) and (rprec == 'right')): + # We decide to shift here... highest precedence to shift + Productions[st_actionp[a].number].reduced -= 1 + st_action[a] = j + st_actionp[a] = p + if not rlevel: + log.info(' ! shift/reduce conflict for %s resolved as shift', a) + self.sr_conflicts.append((st, a, 'shift')) + elif (slevel == rlevel) and (rprec == 'nonassoc'): + st_action[a] = None + else: + # Hmmm. Guess we'll keep the reduce + if not slevel and not rlevel: + log.info(' ! shift/reduce conflict for %s resolved as reduce', a) + self.sr_conflicts.append((st, a, 'reduce')) + + else: + raise LALRError('Unknown conflict in state %d' % st) + else: + st_action[a] = j + st_actionp[a] = p + + # Print the actions associated with each terminal + _actprint = {} + for a, p, m in actlist: + if a in st_action: + if p is st_actionp[a]: + log.info(' %-15s %s', a, m) + _actprint[(a, m)] = 1 + log.info('') + # Print the actions that were not used. (debugging) + not_used = 0 + for a, p, m in actlist: + if a in st_action: + if p is not st_actionp[a]: + if not (a, m) in _actprint: + log.debug(' ! %-15s [ %s ]', a, m) + not_used = 1 + _actprint[(a, m)] = 1 + if not_used: + log.debug('') + + # Construct the goto table for this state + + nkeys = {} + for ii in I: + for s in ii.usyms: + if s in self.grammar.Nonterminals: + nkeys[s] = None + for n in nkeys: + g = self.lr0_goto(I, n) + j = self.lr0_cidhash.get(id(g), -1) + if j >= 0: + st_goto[n] = j + log.info(' %-30s shift and go to state %d', n, j) + + action[st] = st_action + actionp[st] = st_actionp + goto[st] = st_goto + st += 1 + + # ----------------------------------------------------------------------------- + # write() + # + # This function writes the LR parsing tables to a file + # ----------------------------------------------------------------------------- + + def write_table(self, tabmodule, outputdir='', signature=''): + if isinstance(tabmodule, types.ModuleType): + raise IOError("Won't overwrite existing tabmodule") + + basemodulename = tabmodule.split('.')[-1] + filename = os.path.join(outputdir, basemodulename) + '.py' + try: + f = open(filename, 'w') + + f.write(''' +# %s +# This file is automatically generated. Do not edit. +_tabversion = %r + +_lr_method = %r + +_lr_signature = %r + ''' % (os.path.basename(filename), __tabversion__, self.lr_method, signature)) + + # Change smaller to 0 to go back to original tables + smaller = 1 + + # Factor out names to try and make smaller + if smaller: + items = {} + + for s, nd in self.lr_action.items(): + for name, v in nd.items(): + i = items.get(name) + if not i: + i = ([], []) + items[name] = i + i[0].append(s) + i[1].append(v) + + f.write('\n_lr_action_items = {') + for k, v in items.items(): + f.write('%r:([' % k) + for i in v[0]: + f.write('%r,' % i) + f.write('],[') + for i in v[1]: + f.write('%r,' % i) + + f.write(']),') + f.write('}\n') + + f.write(''' +_lr_action = {} +for _k, _v in _lr_action_items.items(): + for _x,_y in zip(_v[0],_v[1]): + if not _x in _lr_action: _lr_action[_x] = {} + _lr_action[_x][_k] = _y +del _lr_action_items +''') + + else: + f.write('\n_lr_action = { ') + for k, v in self.lr_action.items(): + f.write('(%r,%r):%r,' % (k[0], k[1], v)) + f.write('}\n') + + if smaller: + # Factor out names to try and make smaller + items = {} + + for s, nd in self.lr_goto.items(): + for name, v in nd.items(): + i = items.get(name) + if not i: + i = ([], []) + items[name] = i + i[0].append(s) + i[1].append(v) + + f.write('\n_lr_goto_items = {') + for k, v in items.items(): + f.write('%r:([' % k) + for i in v[0]: + f.write('%r,' % i) + f.write('],[') + for i in v[1]: + f.write('%r,' % i) + + f.write(']),') + f.write('}\n') + + f.write(''' +_lr_goto = {} +for _k, _v in _lr_goto_items.items(): + for _x, _y in zip(_v[0], _v[1]): + if not _x in _lr_goto: _lr_goto[_x] = {} + _lr_goto[_x][_k] = _y +del _lr_goto_items +''') + else: + f.write('\n_lr_goto = { ') + for k, v in self.lr_goto.items(): + f.write('(%r,%r):%r,' % (k[0], k[1], v)) + f.write('}\n') + + # Write production table + f.write('_lr_productions = [\n') + for p in self.lr_productions: + if p.func: + f.write(' (%r,%r,%d,%r,%r,%d),\n' % (p.str, p.name, p.len, + p.func, os.path.basename(p.file), p.line)) + else: + f.write(' (%r,%r,%d,None,None,None),\n' % (str(p), p.name, p.len)) + f.write(']\n') + f.close() + + except IOError as e: + raise + + + # ----------------------------------------------------------------------------- + # pickle_table() + # + # This function pickles the LR parsing tables to a supplied file object + # ----------------------------------------------------------------------------- + + def pickle_table(self, filename, signature=''): + try: + import cPickle as pickle + except ImportError: + import pickle + with open(filename, 'wb') as outf: + pickle.dump(__tabversion__, outf, pickle_protocol) + pickle.dump(self.lr_method, outf, pickle_protocol) + pickle.dump(signature, outf, pickle_protocol) + pickle.dump(self.lr_action, outf, pickle_protocol) + pickle.dump(self.lr_goto, outf, pickle_protocol) + + outp = [] + for p in self.lr_productions: + if p.func: + outp.append((p.str, p.name, p.len, p.func, os.path.basename(p.file), p.line)) + else: + outp.append((str(p), p.name, p.len, None, None, None)) + pickle.dump(outp, outf, pickle_protocol) + +# ----------------------------------------------------------------------------- +# === INTROSPECTION === +# +# The following functions and classes are used to implement the PLY +# introspection features followed by the yacc() function itself. +# ----------------------------------------------------------------------------- + +# ----------------------------------------------------------------------------- +# get_caller_module_dict() +# +# This function returns a dictionary containing all of the symbols defined within +# a caller further down the call stack. This is used to get the environment +# associated with the yacc() call if none was provided. +# ----------------------------------------------------------------------------- + +def get_caller_module_dict(levels): + f = sys._getframe(levels) + ldict = f.f_globals.copy() + if f.f_globals != f.f_locals: + ldict.update(f.f_locals) + return ldict + +# ----------------------------------------------------------------------------- +# parse_grammar() +# +# This takes a raw grammar rule string and parses it into production data +# ----------------------------------------------------------------------------- +def parse_grammar(doc, file, line): + grammar = [] + # Split the doc string into lines + pstrings = doc.splitlines() + lastp = None + dline = line + for ps in pstrings: + dline += 1 + p = ps.split() + if not p: + continue + try: + if p[0] == '|': + # This is a continuation of a previous rule + if not lastp: + raise SyntaxError("%s:%d: Misplaced '|'" % (file, dline)) + prodname = lastp + syms = p[1:] + else: + prodname = p[0] + lastp = prodname + syms = p[2:] + assign = p[1] + if assign != ':' and assign != '::=': + raise SyntaxError("%s:%d: Syntax error. Expected ':'" % (file, dline)) + + grammar.append((file, dline, prodname, syms)) + except SyntaxError: + raise + except Exception: + raise SyntaxError('%s:%d: Syntax error in rule %r' % (file, dline, ps.strip())) + + return grammar + +# ----------------------------------------------------------------------------- +# ParserReflect() +# +# This class represents information extracted for building a parser including +# start symbol, error function, tokens, precedence list, action functions, +# etc. +# ----------------------------------------------------------------------------- +class ParserReflect(object): + def __init__(self, pdict, log=None): + self.pdict = pdict + self.start = None + self.error_func = None + self.tokens = None + self.modules = set() + self.grammar = [] + self.error = False + + if log is None: + self.log = PlyLogger(sys.stderr) + else: + self.log = log + + # Get all of the basic information + def get_all(self): + self.get_start() + self.get_error_func() + self.get_tokens() + self.get_precedence() + self.get_pfunctions() + + # Validate all of the information + def validate_all(self): + self.validate_start() + self.validate_error_func() + self.validate_tokens() + self.validate_precedence() + self.validate_pfunctions() + self.validate_modules() + return self.error + + # Compute a signature over the grammar + def signature(self): + parts = [] + try: + if self.start: + parts.append(self.start) + if self.prec: + parts.append(''.join([''.join(p) for p in self.prec])) + if self.tokens: + parts.append(' '.join(self.tokens)) + for f in self.pfuncs: + if f[3]: + parts.append(f[3]) + except (TypeError, ValueError): + pass + return ''.join(parts) + + # ----------------------------------------------------------------------------- + # validate_modules() + # + # This method checks to see if there are duplicated p_rulename() functions + # in the parser module file. Without this function, it is really easy for + # users to make mistakes by cutting and pasting code fragments (and it's a real + # bugger to try and figure out why the resulting parser doesn't work). Therefore, + # we just do a little regular expression pattern matching of def statements + # to try and detect duplicates. + # ----------------------------------------------------------------------------- + + def validate_modules(self): + # Match def p_funcname( + fre = re.compile(r'\s*def\s+(p_[a-zA-Z_0-9]*)\(') + + for module in self.modules: + try: + lines, linen = inspect.getsourcelines(module) + except IOError: + continue + + counthash = {} + for linen, line in enumerate(lines): + linen += 1 + m = fre.match(line) + if m: + name = m.group(1) + prev = counthash.get(name) + if not prev: + counthash[name] = linen + else: + filename = inspect.getsourcefile(module) + self.log.warning('%s:%d: Function %s redefined. Previously defined on line %d', + filename, linen, name, prev) + + # Get the start symbol + def get_start(self): + self.start = self.pdict.get('start') + + # Validate the start symbol + def validate_start(self): + if self.start is not None: + if not isinstance(self.start, string_types): + self.log.error("'start' must be a string") + + # Look for error handler + def get_error_func(self): + self.error_func = self.pdict.get('p_error') + + # Validate the error function + def validate_error_func(self): + if self.error_func: + if isinstance(self.error_func, types.FunctionType): + ismethod = 0 + elif isinstance(self.error_func, types.MethodType): + ismethod = 1 + else: + self.log.error("'p_error' defined, but is not a function or method") + self.error = True + return + + eline = self.error_func.__code__.co_firstlineno + efile = self.error_func.__code__.co_filename + module = inspect.getmodule(self.error_func) + self.modules.add(module) + + argcount = self.error_func.__code__.co_argcount - ismethod + if argcount != 1: + self.log.error('%s:%d: p_error() requires 1 argument', efile, eline) + self.error = True + + # Get the tokens map + def get_tokens(self): + tokens = self.pdict.get('tokens') + if not tokens: + self.log.error('No token list is defined') + self.error = True + return + + if not isinstance(tokens, (list, tuple)): + self.log.error('tokens must be a list or tuple') + self.error = True + return + + if not tokens: + self.log.error('tokens is empty') + self.error = True + return + + self.tokens = tokens + + # Validate the tokens + def validate_tokens(self): + # Validate the tokens. + if 'error' in self.tokens: + self.log.error("Illegal token name 'error'. Is a reserved word") + self.error = True + return + + terminals = set() + for n in self.tokens: + if n in terminals: + self.log.warning('Token %r multiply defined', n) + terminals.add(n) + + # Get the precedence map (if any) + def get_precedence(self): + self.prec = self.pdict.get('precedence') + + # Validate and parse the precedence map + def validate_precedence(self): + preclist = [] + if self.prec: + if not isinstance(self.prec, (list, tuple)): + self.log.error('precedence must be a list or tuple') + self.error = True + return + for level, p in enumerate(self.prec): + if not isinstance(p, (list, tuple)): + self.log.error('Bad precedence table') + self.error = True + return + + if len(p) < 2: + self.log.error('Malformed precedence entry %s. Must be (assoc, term, ..., term)', p) + self.error = True + return + assoc = p[0] + if not isinstance(assoc, string_types): + self.log.error('precedence associativity must be a string') + self.error = True + return + for term in p[1:]: + if not isinstance(term, string_types): + self.log.error('precedence items must be strings') + self.error = True + return + preclist.append((term, assoc, level+1)) + self.preclist = preclist + + # Get all p_functions from the grammar + def get_pfunctions(self): + p_functions = [] + for name, item in self.pdict.items(): + if not name.startswith('p_') or name == 'p_error': + continue + if isinstance(item, (types.FunctionType, types.MethodType)): + line = getattr(item, 'co_firstlineno', item.__code__.co_firstlineno) + module = inspect.getmodule(item) + p_functions.append((line, module, name, item.__doc__)) + + # Sort all of the actions by line number; make sure to stringify + # modules to make them sortable, since `line` may not uniquely sort all + # p functions + p_functions.sort(key=lambda p_function: ( + p_function[0], + str(p_function[1]), + p_function[2], + p_function[3])) + self.pfuncs = p_functions + + # Validate all of the p_functions + def validate_pfunctions(self): + grammar = [] + # Check for non-empty symbols + if len(self.pfuncs) == 0: + self.log.error('no rules of the form p_rulename are defined') + self.error = True + return + + for line, module, name, doc in self.pfuncs: + file = inspect.getsourcefile(module) + func = self.pdict[name] + if isinstance(func, types.MethodType): + reqargs = 2 + else: + reqargs = 1 + if func.__code__.co_argcount > reqargs: + self.log.error('%s:%d: Rule %r has too many arguments', file, line, func.__name__) + self.error = True + elif func.__code__.co_argcount < reqargs: + self.log.error('%s:%d: Rule %r requires an argument', file, line, func.__name__) + self.error = True + elif not func.__doc__: + self.log.warning('%s:%d: No documentation string specified in function %r (ignored)', + file, line, func.__name__) + else: + try: + parsed_g = parse_grammar(doc, file, line) + for g in parsed_g: + grammar.append((name, g)) + except SyntaxError as e: + self.log.error(str(e)) + self.error = True + + # Looks like a valid grammar rule + # Mark the file in which defined. + self.modules.add(module) + + # Secondary validation step that looks for p_ definitions that are not functions + # or functions that look like they might be grammar rules. + + for n, v in self.pdict.items(): + if n.startswith('p_') and isinstance(v, (types.FunctionType, types.MethodType)): + continue + if n.startswith('t_'): + continue + if n.startswith('p_') and n != 'p_error': + self.log.warning('%r not defined as a function', n) + if ((isinstance(v, types.FunctionType) and v.__code__.co_argcount == 1) or + (isinstance(v, types.MethodType) and v.__func__.__code__.co_argcount == 2)): + if v.__doc__: + try: + doc = v.__doc__.split(' ') + if doc[1] == ':': + self.log.warning('%s:%d: Possible grammar rule %r defined without p_ prefix', + v.__code__.co_filename, v.__code__.co_firstlineno, n) + except IndexError: + pass + + self.grammar = grammar + +# ----------------------------------------------------------------------------- +# yacc(module) +# +# Build a parser +# ----------------------------------------------------------------------------- + +def yacc(method='LALR', debug=yaccdebug, module=None, tabmodule=tab_module, start=None, + check_recursion=True, optimize=False, write_tables=True, debugfile=debug_file, + outputdir=None, debuglog=None, errorlog=None, picklefile=None): + + if tabmodule is None: + tabmodule = tab_module + + # Reference to the parsing method of the last built parser + global parse + + # If pickling is enabled, table files are not created + if picklefile: + write_tables = 0 + + if errorlog is None: + errorlog = PlyLogger(sys.stderr) + + # Get the module dictionary used for the parser + if module: + _items = [(k, getattr(module, k)) for k in dir(module)] + pdict = dict(_items) + # If no __file__ attribute is available, try to obtain it from the __module__ instead + if '__file__' not in pdict: + pdict['__file__'] = sys.modules[pdict['__module__']].__file__ + else: + pdict = get_caller_module_dict(2) + + if outputdir is None: + # If no output directory is set, the location of the output files + # is determined according to the following rules: + # - If tabmodule specifies a package, files go into that package directory + # - Otherwise, files go in the same directory as the specifying module + if isinstance(tabmodule, types.ModuleType): + srcfile = tabmodule.__file__ + else: + if '.' not in tabmodule: + srcfile = pdict['__file__'] + else: + parts = tabmodule.split('.') + pkgname = '.'.join(parts[:-1]) + exec('import %s' % pkgname) + srcfile = getattr(sys.modules[pkgname], '__file__', '') + outputdir = os.path.dirname(srcfile) + + # Determine if the module is package of a package or not. + # If so, fix the tabmodule setting so that tables load correctly + pkg = pdict.get('__package__') + if pkg and isinstance(tabmodule, str): + if '.' not in tabmodule: + tabmodule = pkg + '.' + tabmodule + + + + # Set start symbol if it's specified directly using an argument + if start is not None: + pdict['start'] = start + + # Collect parser information from the dictionary + pinfo = ParserReflect(pdict, log=errorlog) + pinfo.get_all() + + if pinfo.error: + raise YaccError('Unable to build parser') + + # Check signature against table files (if any) + signature = pinfo.signature() + + # Read the tables + try: + lr = LRTable() + if picklefile: + read_signature = lr.read_pickle(picklefile) + else: + read_signature = lr.read_table(tabmodule) + if optimize or (read_signature == signature): + try: + lr.bind_callables(pinfo.pdict) + parser = LRParser(lr, pinfo.error_func) + parse = parser.parse + return parser + except Exception as e: + errorlog.warning('There was a problem loading the table file: %r', e) + except VersionError as e: + errorlog.warning(str(e)) + except ImportError: + pass + + if debuglog is None: + if debug: + try: + debuglog = PlyLogger(open(os.path.join(outputdir, debugfile), 'w')) + except IOError as e: + errorlog.warning("Couldn't open %r. %s" % (debugfile, e)) + debuglog = NullLogger() + else: + debuglog = NullLogger() + + debuglog.info('Created by PLY version %s (http://www.dabeaz.com/ply)', __version__) + + errors = False + + # Validate the parser information + if pinfo.validate_all(): + raise YaccError('Unable to build parser') + + if not pinfo.error_func: + errorlog.warning('no p_error() function is defined') + + # Create a grammar object + grammar = Grammar(pinfo.tokens) + + # Set precedence level for terminals + for term, assoc, level in pinfo.preclist: + try: + grammar.set_precedence(term, assoc, level) + except GrammarError as e: + errorlog.warning('%s', e) + + # Add productions to the grammar + for funcname, gram in pinfo.grammar: + file, line, prodname, syms = gram + try: + grammar.add_production(prodname, syms, funcname, file, line) + except GrammarError as e: + errorlog.error('%s', e) + errors = True + + # Set the grammar start symbols + try: + if start is None: + grammar.set_start(pinfo.start) + else: + grammar.set_start(start) + except GrammarError as e: + errorlog.error(str(e)) + errors = True + + if errors: + raise YaccError('Unable to build parser') + + # Verify the grammar structure + undefined_symbols = grammar.undefined_symbols() + for sym, prod in undefined_symbols: + errorlog.error('%s:%d: Symbol %r used, but not defined as a token or a rule', prod.file, prod.line, sym) + errors = True + + unused_terminals = grammar.unused_terminals() + if unused_terminals: + debuglog.info('') + debuglog.info('Unused terminals:') + debuglog.info('') + for term in unused_terminals: + errorlog.warning('Token %r defined, but not used', term) + debuglog.info(' %s', term) + + # Print out all productions to the debug log + if debug: + debuglog.info('') + debuglog.info('Grammar') + debuglog.info('') + for n, p in enumerate(grammar.Productions): + debuglog.info('Rule %-5d %s', n, p) + + # Find unused non-terminals + unused_rules = grammar.unused_rules() + for prod in unused_rules: + errorlog.warning('%s:%d: Rule %r defined, but not used', prod.file, prod.line, prod.name) + + if len(unused_terminals) == 1: + errorlog.warning('There is 1 unused token') + if len(unused_terminals) > 1: + errorlog.warning('There are %d unused tokens', len(unused_terminals)) + + if len(unused_rules) == 1: + errorlog.warning('There is 1 unused rule') + if len(unused_rules) > 1: + errorlog.warning('There are %d unused rules', len(unused_rules)) + + if debug: + debuglog.info('') + debuglog.info('Terminals, with rules where they appear') + debuglog.info('') + terms = list(grammar.Terminals) + terms.sort() + for term in terms: + debuglog.info('%-20s : %s', term, ' '.join([str(s) for s in grammar.Terminals[term]])) + + debuglog.info('') + debuglog.info('Nonterminals, with rules where they appear') + debuglog.info('') + nonterms = list(grammar.Nonterminals) + nonterms.sort() + for nonterm in nonterms: + debuglog.info('%-20s : %s', nonterm, ' '.join([str(s) for s in grammar.Nonterminals[nonterm]])) + debuglog.info('') + + if check_recursion: + unreachable = grammar.find_unreachable() + for u in unreachable: + errorlog.warning('Symbol %r is unreachable', u) + + infinite = grammar.infinite_cycles() + for inf in infinite: + errorlog.error('Infinite recursion detected for symbol %r', inf) + errors = True + + unused_prec = grammar.unused_precedence() + for term, assoc in unused_prec: + errorlog.error('Precedence rule %r defined for unknown symbol %r', assoc, term) + errors = True + + if errors: + raise YaccError('Unable to build parser') + + # Run the LRGeneratedTable on the grammar + if debug: + errorlog.debug('Generating %s tables', method) + + lr = LRGeneratedTable(grammar, method, debuglog) + + if debug: + num_sr = len(lr.sr_conflicts) + + # Report shift/reduce and reduce/reduce conflicts + if num_sr == 1: + errorlog.warning('1 shift/reduce conflict') + elif num_sr > 1: + errorlog.warning('%d shift/reduce conflicts', num_sr) + + num_rr = len(lr.rr_conflicts) + if num_rr == 1: + errorlog.warning('1 reduce/reduce conflict') + elif num_rr > 1: + errorlog.warning('%d reduce/reduce conflicts', num_rr) + + # Write out conflicts to the output file + if debug and (lr.sr_conflicts or lr.rr_conflicts): + debuglog.warning('') + debuglog.warning('Conflicts:') + debuglog.warning('') + + for state, tok, resolution in lr.sr_conflicts: + debuglog.warning('shift/reduce conflict for %s in state %d resolved as %s', tok, state, resolution) + + already_reported = set() + for state, rule, rejected in lr.rr_conflicts: + if (state, id(rule), id(rejected)) in already_reported: + continue + debuglog.warning('reduce/reduce conflict in state %d resolved using rule (%s)', state, rule) + debuglog.warning('rejected rule (%s) in state %d', rejected, state) + errorlog.warning('reduce/reduce conflict in state %d resolved using rule (%s)', state, rule) + errorlog.warning('rejected rule (%s) in state %d', rejected, state) + already_reported.add((state, id(rule), id(rejected))) + + warned_never = [] + for state, rule, rejected in lr.rr_conflicts: + if not rejected.reduced and (rejected not in warned_never): + debuglog.warning('Rule (%s) is never reduced', rejected) + errorlog.warning('Rule (%s) is never reduced', rejected) + warned_never.append(rejected) + + # Write the table file if requested + if write_tables: + try: + lr.write_table(tabmodule, outputdir, signature) + except IOError as e: + errorlog.warning("Couldn't create %r. %s" % (tabmodule, e)) + + # Write a pickled version of the tables + if picklefile: + try: + lr.pickle_table(picklefile, signature) + except IOError as e: + errorlog.warning("Couldn't create %r. %s" % (picklefile, e)) + + # Build the parser + lr.bind_callables(pinfo.pdict) + parser = LRParser(lr, pinfo.error_func) + + parse = parser.parse + return parser diff --git a/third_party/python/ply/ply/ygen.py b/third_party/python/ply/ply/ygen.py new file mode 100644 index 0000000000..acf5ca1a37 --- /dev/null +++ b/third_party/python/ply/ply/ygen.py @@ -0,0 +1,74 @@ +# ply: ygen.py +# +# This is a support program that auto-generates different versions of the YACC parsing +# function with different features removed for the purposes of performance. +# +# Users should edit the method LParser.parsedebug() in yacc.py. The source code +# for that method is then used to create the other methods. See the comments in +# yacc.py for further details. + +import os.path +import shutil + +def get_source_range(lines, tag): + srclines = enumerate(lines) + start_tag = '#--! %s-start' % tag + end_tag = '#--! %s-end' % tag + + for start_index, line in srclines: + if line.strip().startswith(start_tag): + break + + for end_index, line in srclines: + if line.strip().endswith(end_tag): + break + + return (start_index + 1, end_index) + +def filter_section(lines, tag): + filtered_lines = [] + include = True + tag_text = '#--! %s' % tag + for line in lines: + if line.strip().startswith(tag_text): + include = not include + elif include: + filtered_lines.append(line) + return filtered_lines + +def main(): + dirname = os.path.dirname(__file__) + shutil.copy2(os.path.join(dirname, 'yacc.py'), os.path.join(dirname, 'yacc.py.bak')) + with open(os.path.join(dirname, 'yacc.py'), 'r') as f: + lines = f.readlines() + + parse_start, parse_end = get_source_range(lines, 'parsedebug') + parseopt_start, parseopt_end = get_source_range(lines, 'parseopt') + parseopt_notrack_start, parseopt_notrack_end = get_source_range(lines, 'parseopt-notrack') + + # Get the original source + orig_lines = lines[parse_start:parse_end] + + # Filter the DEBUG sections out + parseopt_lines = filter_section(orig_lines, 'DEBUG') + + # Filter the TRACKING sections out + parseopt_notrack_lines = filter_section(parseopt_lines, 'TRACKING') + + # Replace the parser source sections with updated versions + lines[parseopt_notrack_start:parseopt_notrack_end] = parseopt_notrack_lines + lines[parseopt_start:parseopt_end] = parseopt_lines + + lines = [line.rstrip()+'\n' for line in lines] + with open(os.path.join(dirname, 'yacc.py'), 'w') as f: + f.writelines(lines) + + print('Updated yacc.py') + +if __name__ == '__main__': + main() + + + + + diff --git a/third_party/python/ply/setup.cfg b/third_party/python/ply/setup.cfg new file mode 100644 index 0000000000..4ec8a167da --- /dev/null +++ b/third_party/python/ply/setup.cfg @@ -0,0 +1,11 @@ +[bdist_wheel] +universal = 1 + +[metadata] +description-file = README.md + +[egg_info] +tag_build = +tag_date = 0 +tag_svn_revision = 0 + diff --git a/third_party/python/ply/setup.py b/third_party/python/ply/setup.py new file mode 100644 index 0000000000..ee8ccd0ccf --- /dev/null +++ b/third_party/python/ply/setup.py @@ -0,0 +1,31 @@ +try: + from setuptools import setup +except ImportError: + from distutils.core import setup + +setup(name = "ply", + description="Python Lex & Yacc", + long_description = """ +PLY is yet another implementation of lex and yacc for Python. Some notable +features include the fact that its implemented entirely in Python and it +uses LALR(1) parsing which is efficient and well suited for larger grammars. + +PLY provides most of the standard lex/yacc features including support for empty +productions, precedence rules, error recovery, and support for ambiguous grammars. + +PLY is extremely easy to use and provides very extensive error checking. +It is compatible with both Python 2 and Python 3. +""", + license="""BSD""", + version = "3.10", + author = "David Beazley", + author_email = "dave@dabeaz.com", + maintainer = "David Beazley", + maintainer_email = "dave@dabeaz.com", + url = "http://www.dabeaz.com/ply/", + packages = ['ply'], + classifiers = [ + 'Programming Language :: Python :: 3', + 'Programming Language :: Python :: 2', + ] + ) |