Peter M. Groen / oletools

Commit 2ea8bb47d026ee5b3e0de38153b8339d54b1d2f3

Authored by decalage2
1 parent 985f5574

removed the thirdparty/pyparsing folder, replaced by the regular pyparsing as a dependency

Inline Side-by-side
Showing 7 changed files with 9 additions and 3863 deletions
oletools/olevba.py
  View file @2ea8bb4
... ... @@ -202,8 +202,9 @@ from __future__ import print_function
202 202 # 2017-11-20 PL: - fixed issue #219, do not close the file too early
203 203 # 2017-11-24 PL: - added keywords to detect self-modifying macros and
204 204 # attempts to disable macro security (issue #221)
  205 +# 2018-03-19 PL: - removed pyparsing from the thirdparty subfolder
205 206  
206   -__version__ = '0.52'
  207 +__version__ = '0.52.3'
207 208  
208 209 #------------------------------------------------------------------------------
209 210 # TODO:
... ... @@ -286,7 +287,7 @@ if not _parent_dir in sys.path:
286 287 from oletools.thirdparty import olefile
287 288 from oletools.thirdparty.prettytable import prettytable
288 289 from oletools.thirdparty.xglob import xglob, PathNotFoundException
289   -from oletools.thirdparty.pyparsing.pyparsing import \
  290 +from pyparsing import \
290 291 CaselessKeyword, CaselessLiteral, Combine, Forward, Literal, \
291 292 Optional, QuotedString,Regex, Suppress, Word, WordStart, \
292 293 alphanums, alphas, hexnums,nums, opAssoc, srange, \
... ...
oletools/olevba3.py
  View file @2ea8bb4
... ... @@ -197,8 +197,9 @@ from __future__ import print_function
197 197 # 2017-06-08 PL: - fixed issue #122 Chr() with negative numbers
198 198 # 2017-06-15 PL: - deobfuscation line by line to handle large files
199 199 # 2017-07-11 v0.52 PL: - raise exception instead of sys.exit (issue #180)
  200 +# 2018-03-19 PL: - removed pyparsing from the thirdparty subfolder
200 201  
201   -__version__ = '0.52'
  202 +__version__ = '0.52.3'
202 203  
203 204 #------------------------------------------------------------------------------
204 205 # TODO:
... ... @@ -281,7 +282,7 @@ if not _parent_dir in sys.path:
281 282 from oletools.thirdparty import olefile
282 283 from oletools.thirdparty.prettytable import prettytable
283 284 from oletools.thirdparty.xglob import xglob, PathNotFoundException
284   -from oletools.thirdparty.pyparsing.pyparsing import \
  285 +from pyparsing import \
285 286 CaselessKeyword, CaselessLiteral, Combine, Forward, Literal, \
286 287 Optional, QuotedString,Regex, Suppress, Word, WordStart, \
287 288 alphanums, alphas, hexnums,nums, opAssoc, srange, \
... ...
oletools/thirdparty/pyparsing/LICENSE deleted
View file @985f557
1   -Permission is hereby granted, free of charge, to any person obtaining
2   -a copy of this software and associated documentation files (the
3   -"Software"), to deal in the Software without restriction, including
4   -without limitation the rights to use, copy, modify, merge, publish,
5   -distribute, sublicense, and/or sell copies of the Software, and to
6   -permit persons to whom the Software is furnished to do so, subject to
7   -the following conditions:
8   -
9   -The above copyright notice and this permission notice shall be
10   -included in all copies or substantial portions of the Software.
11   -
12   -THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
13   -EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
14   -MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
15   -IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
16   -CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
17   -TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
18   -SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
oletools/thirdparty/pyparsing/README deleted
View file @985f557
1   -====================================
2   -PyParsing -- A Python Parsing Module
3   -====================================
4   -
5   -Introduction
6   -============
7   -
8   -The pyparsing module is an alternative approach to creating and executing
9   -simple grammars, vs. the traditional lex/yacc approach, or the use of
10   -regular expressions. The pyparsing module provides a library of classes
11   -that client code uses to construct the grammar directly in Python code.
12   -
13   -Here is a program to parse "Hello, World!" (or any greeting of the form
14   -"<salutation>, <addressee>!"):
15   -
16   - from pyparsing import Word, alphas
17   - greet = Word( alphas ) + "," + Word( alphas ) + "!"
18   - hello = "Hello, World!"
19   - print hello, "->", greet.parseString( hello )
20   -
21   -The program outputs the following:
22   -
23   - Hello, World! -> ['Hello', ',', 'World', '!']
24   -
25   -The Python representation of the grammar is quite readable, owing to the
26   -self-explanatory class names, and the use of '+', '|' and '^' operator
27   -definitions.
28   -
29   -The parsed results returned from parseString() can be accessed as a
30   -nested list, a dictionary, or an object with named attributes.
31   -
32   -The pyparsing module handles some of the problems that are typically
33   -vexing when writing text parsers:
34   -- extra or missing whitespace (the above program will also handle
35   - "Hello,World!", "Hello , World !", etc.)
36   -- quoted strings
37   -- embedded comments
38   -
39   -The .zip file includes examples of a simple SQL parser, simple CORBA IDL
40   -parser, a config file parser, a chemical formula parser, and a four-
41   -function algebraic notation parser. It also includes a simple how-to
42   -document, and a UML class diagram of the library's classes.
43   -
44   -
45   -
46   -Installation
47   -============
48   -
49   -Do the usual:
50   -
51   - python setup.py install
52   -
53   -(pyparsing requires Python 2.3.2 or later.)
54   -
55   -
56   -Documentation
57   -=============
58   -
59   -See:
60   -
61   - HowToUsePyparsing.html
62   -
63   -
64   -License
65   -=======
66   -
67   - MIT License. See header of pyparsing.py
68   -
69   -History
70   -=======
71   -
72   - See CHANGES file.
oletools/thirdparty/pyparsing/__init__.py deleted
View file @985f557
oletools/thirdparty/pyparsing/pyparsing.py deleted
View file @985f557
1   -# module pyparsing.py
2   -#
3   -# Copyright (c) 2003-2013 Paul T. McGuire
4   -#
5   -# Permission is hereby granted, free of charge, to any person obtaining
6   -# a copy of this software and associated documentation files (the
7   -# "Software"), to deal in the Software without restriction, including
8   -# without limitation the rights to use, copy, modify, merge, publish,
9   -# distribute, sublicense, and/or sell copies of the Software, and to
10   -# permit persons to whom the Software is furnished to do so, subject to
11   -# the following conditions:
12   -#
13   -# The above copyright notice and this permission notice shall be
14   -# included in all copies or substantial portions of the Software.
15   -#
16   -# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
17   -# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
18   -# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
19   -# IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
20   -# CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
21   -# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
22   -# SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
23   -#
24   -
25   -__doc__ = \
26   -"""
27   -pyparsing module - Classes and methods to define and execute parsing grammars
28   -
29   -The pyparsing module is an alternative approach to creating and executing simple grammars,
30   -vs. the traditional lex/yacc approach, or the use of regular expressions. With pyparsing, you
31   -don't need to learn a new syntax for defining grammars or matching expressions - the parsing module
32   -provides a library of classes that you use to construct the grammar directly in Python.
33   -
34   -Here is a program to parse "Hello, World!" (or any greeting of the form C{"<salutation>, <addressee>!"})::
35   -
36   - from pyparsing import Word, alphas
37   -
38   - # define grammar of a greeting
39   - greet = Word( alphas ) + "," + Word( alphas ) + "!"
40   -
41   - hello = "Hello, World!"
42   - print (hello, "->", greet.parseString( hello ))
43   -
44   -The program outputs the following::
45   -
46   - Hello, World! -> ['Hello', ',', 'World', '!']
47   -
48   -The Python representation of the grammar is quite readable, owing to the self-explanatory
49   -class names, and the use of '+', '|' and '^' operators.
50   -
51   -The parsed results returned from C{parseString()} can be accessed as a nested list, a dictionary, or an
52   -object with named attributes.
53   -
54   -The pyparsing module handles some of the problems that are typically vexing when writing text parsers:
55   - - extra or missing whitespace (the above program will also handle "Hello,World!", "Hello , World !", etc.)
56   - - quoted strings
57   - - embedded comments
58   -"""
59   -
60   -__version__ = "2.0.3"
61   -__versionTime__ = "16 Aug 2014 00:12"
62   -__author__ = "Paul McGuire <ptmcg@users.sourceforge.net>"
63   -
64   -import string
65   -from weakref import ref as wkref
66   -import copy
67   -import sys
68   -import warnings
69   -import re
70   -import sre_constants
71   -import collections
72   -import pprint
73   -#~ sys.stderr.write( "testing pyparsing module, version %s, %s\n" % (__version__,__versionTime__ ) )
74   -
75   -__all__ = [
76   -'And', 'CaselessKeyword', 'CaselessLiteral', 'CharsNotIn', 'Combine', 'Dict', 'Each', 'Empty',
77   -'FollowedBy', 'Forward', 'GoToColumn', 'Group', 'Keyword', 'LineEnd', 'LineStart', 'Literal',
78   -'MatchFirst', 'NoMatch', 'NotAny', 'OneOrMore', 'OnlyOnce', 'Optional', 'Or',
79   -'ParseBaseException', 'ParseElementEnhance', 'ParseException', 'ParseExpression', 'ParseFatalException',
80   -'ParseResults', 'ParseSyntaxException', 'ParserElement', 'QuotedString', 'RecursiveGrammarException',
81   -'Regex', 'SkipTo', 'StringEnd', 'StringStart', 'Suppress', 'Token', 'TokenConverter', 'Upcase',
82   -'White', 'Word', 'WordEnd', 'WordStart', 'ZeroOrMore',
83   -'alphanums', 'alphas', 'alphas8bit', 'anyCloseTag', 'anyOpenTag', 'cStyleComment', 'col',
84   -'commaSeparatedList', 'commonHTMLEntity', 'countedArray', 'cppStyleComment', 'dblQuotedString',
85   -'dblSlashComment', 'delimitedList', 'dictOf', 'downcaseTokens', 'empty', 'hexnums',
86   -'htmlComment', 'javaStyleComment', 'keepOriginalText', 'line', 'lineEnd', 'lineStart', 'lineno',
87   -'makeHTMLTags', 'makeXMLTags', 'matchOnlyAtCol', 'matchPreviousExpr', 'matchPreviousLiteral',
88   -'nestedExpr', 'nullDebugAction', 'nums', 'oneOf', 'opAssoc', 'operatorPrecedence', 'printables',
89   -'punc8bit', 'pythonStyleComment', 'quotedString', 'removeQuotes', 'replaceHTMLEntity',
90   -'replaceWith', 'restOfLine', 'sglQuotedString', 'srange', 'stringEnd',
91   -'stringStart', 'traceParseAction', 'unicodeString', 'upcaseTokens', 'withAttribute',
92   -'indentedBlock', 'originalTextFor', 'ungroup', 'infixNotation','locatedExpr',
93   -]
94   -
95   -PY_3 = sys.version.startswith('3')
96   -if PY_3:
97   - _MAX_INT = sys.maxsize
98   - basestring = str
99   - unichr = chr
100   - _ustr = str
101   -
102   - # build list of single arg builtins, that can be used as parse actions
103   - singleArgBuiltins = [sum, len, sorted, reversed, list, tuple, set, any, all, min, max]
104   -
105   -else:
106   - _MAX_INT = sys.maxint
107   - range = xrange
108   -
109   - def _ustr(obj):
110   - """Drop-in replacement for str(obj) that tries to be Unicode friendly. It first tries
111   - str(obj). If that fails with a UnicodeEncodeError, then it tries unicode(obj). It
112   - then < returns the unicode object | encodes it with the default encoding | ... >.
113   - """
114   - if isinstance(obj,unicode):
115   - return obj
116   -
117   - try:
118   - # If this works, then _ustr(obj) has the same behaviour as str(obj), so
119   - # it won't break any existing code.
120   - return str(obj)
121   -
122   - except UnicodeEncodeError:
123   - # The Python docs (http://docs.python.org/ref/customization.html#l2h-182)
124   - # state that "The return value must be a string object". However, does a
125   - # unicode object (being a subclass of basestring) count as a "string
126   - # object"?
127   - # If so, then return a unicode object:
128   - return unicode(obj)
129   - # Else encode it... but how? There are many choices... :)
130   - # Replace unprintables with escape codes?
131   - #return unicode(obj).encode(sys.getdefaultencoding(), 'backslashreplace_errors')
132   - # Replace unprintables with question marks?
133   - #return unicode(obj).encode(sys.getdefaultencoding(), 'replace')
134   - # ...
135   -
136   - # build list of single arg builtins, tolerant of Python version, that can be used as parse actions
137   - singleArgBuiltins = []
138   - import __builtin__
139   - for fname in "sum len sorted reversed list tuple set any all min max".split():
140   - try:
141   - singleArgBuiltins.append(getattr(__builtin__,fname))
142   - except AttributeError:
143   - continue
144   -
145   -_generatorType = type((y for y in range(1)))
146   -
147   -def _xml_escape(data):
148   - """Escape &, <, >, ", ', etc. in a string of data."""
149   -
150   - # ampersand must be replaced first
151   - from_symbols = '&><"\''
152   - to_symbols = ('&'+s+';' for s in "amp gt lt quot apos".split())
153   - for from_,to_ in zip(from_symbols, to_symbols):
154   - data = data.replace(from_, to_)
155   - return data
156   -
157   -class _Constants(object):
158   - pass
159   -
160   -alphas = string.ascii_lowercase + string.ascii_uppercase
161   -nums = "0123456789"
162   -hexnums = nums + "ABCDEFabcdef"
163   -alphanums = alphas + nums
164   -_bslash = chr(92)
165   -printables = "".join(c for c in string.printable if c not in string.whitespace)
166   -
167   -class ParseBaseException(Exception):
168   - """base exception class for all parsing runtime exceptions"""
169   - # Performance tuning: we construct a *lot* of these, so keep this
170   - # constructor as small and fast as possible
171   - def __init__( self, pstr, loc=0, msg=None, elem=None ):
172   - self.loc = loc
173   - if msg is None:
174   - self.msg = pstr
175   - self.pstr = ""
176   - else:
177   - self.msg = msg
178   - self.pstr = pstr
179   - self.parserElement = elem
180   -
181   - def __getattr__( self, aname ):
182   - """supported attributes by name are:
183   - - lineno - returns the line number of the exception text
184   - - col - returns the column number of the exception text
185   - - line - returns the line containing the exception text
186   - """
187   - if( aname == "lineno" ):
188   - return lineno( self.loc, self.pstr )
189   - elif( aname in ("col", "column") ):
190   - return col( self.loc, self.pstr )
191   - elif( aname == "line" ):
192   - return line( self.loc, self.pstr )
193   - else:
194   - raise AttributeError(aname)
195   -
196   - def __str__( self ):
197   - return "%s (at char %d), (line:%d, col:%d)" % \
198   - ( self.msg, self.loc, self.lineno, self.column )
199   - def __repr__( self ):
200   - return _ustr(self)
201   - def markInputline( self, markerString = ">!<" ):
202   - """Extracts the exception line from the input string, and marks
203   - the location of the exception with a special symbol.
204   - """
205   - line_str = self.line
206   - line_column = self.column - 1
207   - if markerString:
208   - line_str = "".join((line_str[:line_column],
209   - markerString, line_str[line_column:]))
210   - return line_str.strip()
211   - def __dir__(self):
212   - return "loc msg pstr parserElement lineno col line " \
213   - "markInputline __str__ __repr__".split()
214   -
215   -class ParseException(ParseBaseException):
216   - """exception thrown when parse expressions don't match class;
217   - supported attributes by name are:
218   - - lineno - returns the line number of the exception text
219   - - col - returns the column number of the exception text
220   - - line - returns the line containing the exception text
221   - """
222   - pass
223   -
224   -class ParseFatalException(ParseBaseException):
225   - """user-throwable exception thrown when inconsistent parse content
226   - is found; stops all parsing immediately"""
227   - pass
228   -
229   -class ParseSyntaxException(ParseFatalException):
230   - """just like C{L{ParseFatalException}}, but thrown internally when an
231   - C{L{ErrorStop<And._ErrorStop>}} ('-' operator) indicates that parsing is to stop immediately because
232   - an unbacktrackable syntax error has been found"""
233   - def __init__(self, pe):
234   - super(ParseSyntaxException, self).__init__(
235   - pe.pstr, pe.loc, pe.msg, pe.parserElement)
236   -
237   -#~ class ReparseException(ParseBaseException):
238   - #~ """Experimental class - parse actions can raise this exception to cause
239   - #~ pyparsing to reparse the input string:
240   - #~ - with a modified input string, and/or
241   - #~ - with a modified start location
242   - #~ Set the values of the ReparseException in the constructor, and raise the
243   - #~ exception in a parse action to cause pyparsing to use the new string/location.
244   - #~ Setting the values as None causes no change to be made.
245   - #~ """
246   - #~ def __init_( self, newstring, restartLoc ):
247   - #~ self.newParseText = newstring
248   - #~ self.reparseLoc = restartLoc
249   -
250   -class RecursiveGrammarException(Exception):
251   - """exception thrown by C{validate()} if the grammar could be improperly recursive"""
252   - def __init__( self, parseElementList ):
253   - self.parseElementTrace = parseElementList
254   -
255   - def __str__( self ):
256   - return "RecursiveGrammarException: %s" % self.parseElementTrace
257   -
258   -class _ParseResultsWithOffset(object):
259   - def __init__(self,p1,p2):
260   - self.tup = (p1,p2)
261   - def __getitem__(self,i):
262   - return self.tup[i]
263   - def __repr__(self):
264   - return repr(self.tup)
265   - def setOffset(self,i):
266   - self.tup = (self.tup[0],i)
267   -
268   -class ParseResults(object):
269   - """Structured parse results, to provide multiple means of access to the parsed data:
270   - - as a list (C{len(results)})
271   - - by list index (C{results[0], results[1]}, etc.)
272   - - by attribute (C{results.<resultsName>})
273   - """
274   - def __new__(cls, toklist, name=None, asList=True, modal=True ):
275   - if isinstance(toklist, cls):
276   - return toklist
277   - retobj = object.__new__(cls)
278   - retobj.__doinit = True
279   - return retobj
280   -
281   - # Performance tuning: we construct a *lot* of these, so keep this
282   - # constructor as small and fast as possible
283   - def __init__( self, toklist, name=None, asList=True, modal=True, isinstance=isinstance ):
284   - if self.__doinit:
285   - self.__doinit = False
286   - self.__name = None
287   - self.__parent = None
288   - self.__accumNames = {}
289   - if isinstance(toklist, list):
290   - self.__toklist = toklist[:]
291   - elif isinstance(toklist, _generatorType):
292   - self.__toklist = list(toklist)
293   - else:
294   - self.__toklist = [toklist]
295   - self.__tokdict = dict()
296   -
297   - if name is not None and name:
298   - if not modal:
299   - self.__accumNames[name] = 0
300   - if isinstance(name,int):
301   - name = _ustr(name) # will always return a str, but use _ustr for consistency
302   - self.__name = name
303   - if not (isinstance(toklist, (type(None), basestring, list)) and toklist in (None,'',[])):
304   - if isinstance(toklist,basestring):
305   - toklist = [ toklist ]
306   - if asList:
307   - if isinstance(toklist,ParseResults):
308   - self[name] = _ParseResultsWithOffset(toklist.copy(),0)
309   - else:
310   - self[name] = _ParseResultsWithOffset(ParseResults(toklist[0]),0)
311   - self[name].__name = name
312   - else:
313   - try:
314   - self[name] = toklist[0]
315   - except (KeyError,TypeError,IndexError):
316   - self[name] = toklist
317   -
318   - def __getitem__( self, i ):
319   - if isinstance( i, (int,slice) ):
320   - return self.__toklist[i]
321   - else:
322   - if i not in self.__accumNames:
323   - return self.__tokdict[i][-1][0]
324   - else:
325   - return ParseResults([ v[0] for v in self.__tokdict[i] ])
326   -
327   - def __setitem__( self, k, v, isinstance=isinstance ):
328   - if isinstance(v,_ParseResultsWithOffset):
329   - self.__tokdict[k] = self.__tokdict.get(k,list()) + [v]
330   - sub = v[0]
331   - elif isinstance(k,int):
332   - self.__toklist[k] = v
333   - sub = v
334   - else:
335   - self.__tokdict[k] = self.__tokdict.get(k,list()) + [_ParseResultsWithOffset(v,0)]
336   - sub = v
337   - if isinstance(sub,ParseResults):
338   - sub.__parent = wkref(self)
339   -
340   - def __delitem__( self, i ):
341   - if isinstance(i,(int,slice)):
342   - mylen = len( self.__toklist )
343   - del self.__toklist[i]
344   -
345   - # convert int to slice
346   - if isinstance(i, int):
347   - if i < 0:
348   - i += mylen
349   - i = slice(i, i+1)
350   - # get removed indices
351   - removed = list(range(*i.indices(mylen)))
352   - removed.reverse()
353   - # fixup indices in token dictionary
354   - for name in self.__tokdict:
355   - occurrences = self.__tokdict[name]
356   - for j in removed:
357   - for k, (value, position) in enumerate(occurrences):
358   - occurrences[k] = _ParseResultsWithOffset(value, position - (position > j))
359   - else:
360   - del self.__tokdict[i]
361   -
362   - def __contains__( self, k ):
363   - return k in self.__tokdict
364   -
365   - def __len__( self ): return len( self.__toklist )
366   - def __bool__(self): return len( self.__toklist ) > 0
367   - __nonzero__ = __bool__
368   - def __iter__( self ): return iter( self.__toklist )
369   - def __reversed__( self ): return iter( self.__toklist[::-1] )
370   - def iterkeys( self ):
371   - """Returns all named result keys."""
372   - if hasattr(self.__tokdict, "iterkeys"):
373   - return self.__tokdict.iterkeys()
374   - else:
375   - return iter(self.__tokdict)
376   -
377   - def itervalues( self ):
378   - """Returns all named result values."""
379   - return (self[k] for k in self.iterkeys())
380   -
381   - def iteritems( self ):
382   - return ((k, self[k]) for k in self.iterkeys())
383   -
384   - if PY_3:
385   - keys = iterkeys
386   - values = itervalues
387   - items = iteritems
388   - else:
389   - def keys( self ):
390   - """Returns all named result keys."""
391   - return list(self.iterkeys())
392   -
393   - def values( self ):
394   - """Returns all named result values."""
395   - return list(self.itervalues())
396   -
397   - def items( self ):
398   - """Returns all named result keys and values as a list of tuples."""
399   - return list(self.iteritems())
400   -
401   - def haskeys( self ):
402   - """Since keys() returns an iterator, this method is helpful in bypassing
403   - code that looks for the existence of any defined results names."""
404   - return bool(self.__tokdict)
405   -
406   - def pop( self, *args, **kwargs):
407   - """Removes and returns item at specified index (default=last).
408   - Supports both list and dict semantics for pop(). If passed no
409   - argument or an integer argument, it will use list semantics
410   - and pop tokens from the list of parsed tokens. If passed a
411   - non-integer argument (most likely a string), it will use dict
412   - semantics and pop the corresponding value from any defined
413   - results names. A second default return value argument is
414   - supported, just as in dict.pop()."""
415   - if not args:
416   - args = [-1]
417   - for k,v in kwargs.items():
418   - if k == 'default':
419   - args = (args[0], v)
420   - else:
421   - raise TypeError("pop() got an unexpected keyword argument '%s'" % k)
422   - if (isinstance(args[0], int) or
423   - len(args) == 1 or
424   - args[0] in self):
425   - index = args[0]
426   - ret = self[index]
427   - del self[index]
428   - return ret
429   - else:
430   - defaultvalue = args[1]
431   - return defaultvalue
432   -
433   - def get(self, key, defaultValue=None):
434   - """Returns named result matching the given key, or if there is no
435   - such name, then returns the given C{defaultValue} or C{None} if no
436   - C{defaultValue} is specified."""
437   - if key in self:
438   - return self[key]
439   - else:
440   - return defaultValue
441   -
442   - def insert( self, index, insStr ):
443   - """Inserts new element at location index in the list of parsed tokens."""
444   - self.__toklist.insert(index, insStr)
445   - # fixup indices in token dictionary
446   - for name in self.__tokdict:
447   - occurrences = self.__tokdict[name]
448   - for k, (value, position) in enumerate(occurrences):
449   - occurrences[k] = _ParseResultsWithOffset(value, position + (position > index))
450   -
451   - def append( self, item ):
452   - """Add single element to end of ParseResults list of elements."""
453   - self.__toklist.append(item)
454   -
455   - def extend( self, itemseq ):
456   - """Add sequence of elements to end of ParseResults list of elements."""
457   - if isinstance(itemseq, ParseResults):
458   - self += itemseq
459   - else:
460   - self.__toklist.extend(itemseq)
461   -
462   - def clear( self ):
463   - """Clear all elements and results names."""
464   - del self.__toklist[:]
465   - self.__tokdict.clear()
466   -
467   - def __getattr__( self, name ):
468   - try:
469   - return self[name]
470   - except KeyError:
471   - return ""
472   -
473   - if name in self.__tokdict:
474   - if name not in self.__accumNames:
475   - return self.__tokdict[name][-1][0]
476   - else:
477   - return ParseResults([ v[0] for v in self.__tokdict[name] ])
478   - else:
479   - return ""
480   -
481   - def __add__( self, other ):
482   - ret = self.copy()
483   - ret += other
484   - return ret
485   -
486   - def __iadd__( self, other ):
487   - if other.__tokdict:
488   - offset = len(self.__toklist)
489   - addoffset = ( lambda a: (a<0 and offset) or (a+offset) )
490   - otheritems = other.__tokdict.items()
491   - otherdictitems = [(k, _ParseResultsWithOffset(v[0],addoffset(v[1])) )
492   - for (k,vlist) in otheritems for v in vlist]
493   - for k,v in otherdictitems:
494   - self[k] = v
495   - if isinstance(v[0],ParseResults):
496   - v[0].__parent = wkref(self)
497   -
498   - self.__toklist += other.__toklist
499   - self.__accumNames.update( other.__accumNames )
500   - return self
501   -
502   - def __radd__(self, other):
503   - if isinstance(other,int) and other == 0:
504   - return self.copy()
505   -
506   - def __repr__( self ):
507   - return "(%s, %s)" % ( repr( self.__toklist ), repr( self.__tokdict ) )
508   -
509   - def __str__( self ):
510   - out = []
511   - for i in self.__toklist:
512   - if isinstance(i, ParseResults):
513   - out.append(_ustr(i))
514   - else:
515   - out.append(repr(i))
516   - return '[' + ', '.join(out) + ']'
517   -
518   - def _asStringList( self, sep='' ):
519   - out = []
520   - for item in self.__toklist:
521   - if out and sep:
522   - out.append(sep)
523   - if isinstance( item, ParseResults ):
524   - out += item._asStringList()
525   - else:
526   - out.append( _ustr(item) )
527   - return out
528   -
529   - def asList( self ):
530   - """Returns the parse results as a nested list of matching tokens, all converted to strings."""
531   - out = []
532   - for res in self.__toklist:
533   - if isinstance(res,ParseResults):
534   - out.append( res.asList() )
535   - else:
536   - out.append( res )
537   - return out
538   -
539   - def asDict( self ):
540   - """Returns the named parse results as dictionary."""
541   - if PY_3:
542   - return dict( self.items() )
543   - else:
544   - return dict( self.iteritems() )
545   -
546   - def copy( self ):
547   - """Returns a new copy of a C{ParseResults} object."""
548   - ret = ParseResults( self.__toklist )
549   - ret.__tokdict = self.__tokdict.copy()
550   - ret.__parent = self.__parent
551   - ret.__accumNames.update( self.__accumNames )
552   - ret.__name = self.__name
553   - return ret
554   -
555   - def asXML( self, doctag=None, namedItemsOnly=False, indent="", formatted=True ):
556   - """Returns the parse results as XML. Tags are created for tokens and lists that have defined results names."""
557   - nl = "\n"
558   - out = []
559   - namedItems = dict((v[1],k) for (k,vlist) in self.__tokdict.items()
560   - for v in vlist)
561   - nextLevelIndent = indent + " "
562   -
563   - # collapse out indents if formatting is not desired
564   - if not formatted:
565   - indent = ""
566   - nextLevelIndent = ""
567   - nl = ""
568   -
569   - selfTag = None
570   - if doctag is not None:
571   - selfTag = doctag
572   - else:
573   - if self.__name:
574   - selfTag = self.__name
575   -
576   - if not selfTag:
577   - if namedItemsOnly:
578   - return ""
579   - else:
580   - selfTag = "ITEM"
581   -
582   - out += [ nl, indent, "<", selfTag, ">" ]
583   -
584   - worklist = self.__toklist
585   - for i,res in enumerate(worklist):
586   - if isinstance(res,ParseResults):
587   - if i in namedItems:
588   - out += [ res.asXML(namedItems[i],
589   - namedItemsOnly and doctag is None,
590   - nextLevelIndent,
591   - formatted)]
592   - else:
593   - out += [ res.asXML(None,
594   - namedItemsOnly and doctag is None,
595   - nextLevelIndent,
596   - formatted)]
597   - else:
598   - # individual token, see if there is a name for it
599   - resTag = None
600   - if i in namedItems:
601   - resTag = namedItems[i]
602   - if not resTag:
603   - if namedItemsOnly:
604   - continue
605   - else:
606   - resTag = "ITEM"
607   - xmlBodyText = _xml_escape(_ustr(res))
608   - out += [ nl, nextLevelIndent, "<", resTag, ">",
609   - xmlBodyText,
610   - "</", resTag, ">" ]
611   -
612   - out += [ nl, indent, "</", selfTag, ">" ]
613   - return "".join(out)
614   -
615   - def __lookup(self,sub):
616   - for k,vlist in self.__tokdict.items():
617   - for v,loc in vlist:
618   - if sub is v:
619   - return k
620   - return None
621   -
622   - def getName(self):
623   - """Returns the results name for this token expression."""
624   - if self.__name:
625   - return self.__name
626   - elif self.__parent:
627   - par = self.__parent()
628   - if par:
629   - return par.__lookup(self)
630   - else:
631   - return None
632   - elif (len(self) == 1 and
633   - len(self.__tokdict) == 1 and
634   - self.__tokdict.values()[0][0][1] in (0,-1)):
635   - return self.__tokdict.keys()[0]
636   - else:
637   - return None
638   -
639   - def dump(self,indent='',depth=0):
640   - """Diagnostic method for listing out the contents of a C{ParseResults}.
641   - Accepts an optional C{indent} argument so that this string can be embedded
642   - in a nested display of other data."""
643   - out = []
644   - NL = '\n'
645   - out.append( indent+_ustr(self.asList()) )
646   - items = sorted(self.items())
647   - for k,v in items:
648   - if out:
649   - out.append(NL)
650   - out.append( "%s%s- %s: " % (indent,(' '*depth), k) )
651   - if isinstance(v,ParseResults):
652   - if v:
653   - if v.haskeys():
654   - out.append( v.dump(indent,depth+1) )
655   - elif any(isinstance(vv,ParseResults) for vv in v):
656   - for i,vv in enumerate(v):
657   - if isinstance(vv,ParseResults):
658   - out.append("\n%s%s[%d]:\n%s%s%s" % (indent,(' '*(depth+1)),i,indent,(' '*(depth+2)),vv.dump(indent,depth+2) ))
659   - else:
660   - out.append("\n%s%s[%d]:\n%s%s%s" % (indent,(' '*(depth+1)),i,indent,(' '*(depth+2)),_ustr(vv)))
661   - else:
662   - out.append(_ustr(v))
663   - else:
664   - out.append(_ustr(v))
665   - else:
666   - out.append(_ustr(v))
667   - return "".join(out)
668   -
669   - def pprint(self, *args, **kwargs):
670   - """Pretty-printer for parsed results as a list, using the C{pprint} module.
671   - Accepts additional positional or keyword args as defined for the
672   - C{pprint.pprint} method. (U{http://docs.python.org/3/library/pprint.html#pprint.pprint})"""
673   - pprint.pprint(self.asList(), *args, **kwargs)
674   -
675   - # add support for pickle protocol
676   - def __getstate__(self):
677   - return ( self.__toklist,
678   - ( self.__tokdict.copy(),
679   - self.__parent is not None and self.__parent() or None,
680   - self.__accumNames,
681   - self.__name ) )
682   -
683   - def __setstate__(self,state):
684   - self.__toklist = state[0]
685   - (self.__tokdict,
686   - par,
687   - inAccumNames,
688   - self.__name) = state[1]
689   - self.__accumNames = {}
690   - self.__accumNames.update(inAccumNames)
691   - if par is not None:
692   - self.__parent = wkref(par)
693   - else:
694   - self.__parent = None
695   -
696   - def __dir__(self):
697   - return dir(super(ParseResults,self)) + list(self.keys())
698   -
699   -collections.MutableMapping.register(ParseResults)
700   -
701   -def col (loc,strg):
702   - """Returns current column within a string, counting newlines as line separators.
703   - The first column is number 1.
704   -
705   - Note: the default parsing behavior is to expand tabs in the input string
706   - before starting the parsing process. See L{I{ParserElement.parseString}<ParserElement.parseString>} for more information
707   - on parsing strings containing C{<TAB>}s, and suggested methods to maintain a
708   - consistent view of the parsed string, the parse location, and line and column
709   - positions within the parsed string.
710   - """
711   - return (loc<len(strg) and strg[loc] == '\n') and 1 or loc - strg.rfind("\n", 0, loc)
712   -
713   -def lineno(loc,strg):
714   - """Returns current line number within a string, counting newlines as line separators.
715   - The first line is number 1.
716   -
717   - Note: the default parsing behavior is to expand tabs in the input string
718   - before starting the parsing process. See L{I{ParserElement.parseString}<ParserElement.parseString>} for more information
719   - on parsing strings containing C{<TAB>}s, and suggested methods to maintain a
720   - consistent view of the parsed string, the parse location, and line and column
721   - positions within the parsed string.
722   - """
723   - return strg.count("\n",0,loc) + 1
724   -
725   -def line( loc, strg ):
726   - """Returns the line of text containing loc within a string, counting newlines as line separators.
727   - """
728   - lastCR = strg.rfind("\n", 0, loc)
729   - nextCR = strg.find("\n", loc)
730   - if nextCR >= 0:
731   - return strg[lastCR+1:nextCR]
732   - else:
733   - return strg[lastCR+1:]
734   -
735   -def _defaultStartDebugAction( instring, loc, expr ):
736   - print (("Match " + _ustr(expr) + " at loc " + _ustr(loc) + "(%d,%d)" % ( lineno(loc,instring), col(loc,instring) )))
737   -
738   -def _defaultSuccessDebugAction( instring, startloc, endloc, expr, toks ):
739   - print ("Matched " + _ustr(expr) + " -> " + str(toks.asList()))
740   -
741   -def _defaultExceptionDebugAction( instring, loc, expr, exc ):
742   - print ("Exception raised:" + _ustr(exc))
743   -
744   -def nullDebugAction(*args):
745   - """'Do-nothing' debug action, to suppress debugging output during parsing."""
746   - pass
747   -
748   -# Only works on Python 3.x - nonlocal is toxic to Python 2 installs
749   -#~ 'decorator to trim function calls to match the arity of the target'
750   -#~ def _trim_arity(func, maxargs=3):
751   - #~ if func in singleArgBuiltins:
752   - #~ return lambda s,l,t: func(t)
753   - #~ limit = 0
754   - #~ foundArity = False
755   - #~ def wrapper(*args):
756   - #~ nonlocal limit,foundArity
757   - #~ while 1:
758   - #~ try:
759   - #~ ret = func(*args[limit:])
760   - #~ foundArity = True
761   - #~ return ret
762   - #~ except TypeError:
763   - #~ if limit == maxargs or foundArity:
764   - #~ raise
765   - #~ limit += 1
766   - #~ continue
767   - #~ return wrapper
768   -
769   -# this version is Python 2.x-3.x cross-compatible
770   -'decorator to trim function calls to match the arity of the target'
771   -def _trim_arity(func, maxargs=2):
772   - if func in singleArgBuiltins:
773   - return lambda s,l,t: func(t)
774   - limit = [0]
775   - foundArity = [False]
776   - def wrapper(*args):
777   - while 1:
778   - try:
779   - ret = func(*args[limit[0]:])
780   - foundArity[0] = True
781   - return ret
782   - except TypeError:
783   - if limit[0] <= maxargs and not foundArity[0]:
784   - limit[0] += 1
785   - continue
786   - raise
787   - return wrapper
788   -
789   -class ParserElement(object):
790   - """Abstract base level parser element class."""
791   - DEFAULT_WHITE_CHARS = " \n\t\r"
792   - verbose_stacktrace = False
793   -
794   - def setDefaultWhitespaceChars( chars ):
795   - """Overrides the default whitespace chars
796   - """
797   - ParserElement.DEFAULT_WHITE_CHARS = chars
798   - setDefaultWhitespaceChars = staticmethod(setDefaultWhitespaceChars)
799   -
800   - def inlineLiteralsUsing(cls):
801   - """
802   - Set class to be used for inclusion of string literals into a parser.
803   - """
804   - ParserElement.literalStringClass = cls
805   - inlineLiteralsUsing = staticmethod(inlineLiteralsUsing)
806   -
807   - def __init__( self, savelist=False ):
808   - self.parseAction = list()
809   - self.failAction = None
810   - #~ self.name = "<unknown>" # don't define self.name, let subclasses try/except upcall
811   - self.strRepr = None
812   - self.resultsName = None
813   - self.saveAsList = savelist
814   - self.skipWhitespace = True
815   - self.whiteChars = ParserElement.DEFAULT_WHITE_CHARS
816   - self.copyDefaultWhiteChars = True
817   - self.mayReturnEmpty = False # used when checking for left-recursion
818   - self.keepTabs = False
819   - self.ignoreExprs = list()
820   - self.debug = False
821   - self.streamlined = False
822   - self.mayIndexError = True # used to optimize exception handling for subclasses that don't advance parse index
823   - self.errmsg = ""
824   - self.modalResults = True # used to mark results names as modal (report only last) or cumulative (list all)
825   - self.debugActions = ( None, None, None ) #custom debug actions
826   - self.re = None
827   - self.callPreparse = True # used to avoid redundant calls to preParse
828   - self.callDuringTry = False
829   -
830   - def copy( self ):
831   - """Make a copy of this C{ParserElement}. Useful for defining different parse actions
832   - for the same parsing pattern, using copies of the original parse element."""
833   - cpy = copy.copy( self )
834   - cpy.parseAction = self.parseAction[:]
835   - cpy.ignoreExprs = self.ignoreExprs[:]
836   - if self.copyDefaultWhiteChars:
837   - cpy.whiteChars = ParserElement.DEFAULT_WHITE_CHARS
838   - return cpy
839   -
840   - def setName( self, name ):
841   - """Define name for this expression, for use in debugging."""
842   - self.name = name
843   - self.errmsg = "Expected " + self.name
844   - if hasattr(self,"exception"):
845   - self.exception.msg = self.errmsg
846   - return self
847   -
848   - def setResultsName( self, name, listAllMatches=False ):
849   - """Define name for referencing matching tokens as a nested attribute
850   - of the returned parse results.
851   - NOTE: this returns a *copy* of the original C{ParserElement} object;
852   - this is so that the client can define a basic element, such as an
853   - integer, and reference it in multiple places with different names.
854   -
855   - You can also set results names using the abbreviated syntax,
856   - C{expr("name")} in place of C{expr.setResultsName("name")} -
857   - see L{I{__call__}<__call__>}.
858   - """
859   - newself = self.copy()
860   - if name.endswith("*"):
861   - name = name[:-1]
862   - listAllMatches=True
863   - newself.resultsName = name
864   - newself.modalResults = not listAllMatches
865   - return newself
866   -
867   - def setBreak(self,breakFlag = True):
868   - """Method to invoke the Python pdb debugger when this element is
869   - about to be parsed. Set C{breakFlag} to True to enable, False to
870   - disable.
871   - """
872   - if breakFlag:
873   - _parseMethod = self._parse
874   - def breaker(instring, loc, doActions=True, callPreParse=True):
875   - import pdb
876   - pdb.set_trace()
877   - return _parseMethod( instring, loc, doActions, callPreParse )
878   - breaker._originalParseMethod = _parseMethod
879   - self._parse = breaker
880   - else:
881   - if hasattr(self._parse,"_originalParseMethod"):
882   - self._parse = self._parse._originalParseMethod
883   - return self
884   -
885   - def setParseAction( self, *fns, **kwargs ):
886   - """Define action to perform when successfully matching parse element definition.
887   - Parse action fn is a callable method with 0-3 arguments, called as C{fn(s,loc,toks)},
888   - C{fn(loc,toks)}, C{fn(toks)}, or just C{fn()}, where:
889   - - s = the original string being parsed (see note below)
890   - - loc = the location of the matching substring
891   - - toks = a list of the matched tokens, packaged as a C{L{ParseResults}} object
892   - If the functions in fns modify the tokens, they can return them as the return
893   - value from fn, and the modified list of tokens will replace the original.
894   - Otherwise, fn does not need to return any value.
895   -
896   - Note: the default parsing behavior is to expand tabs in the input string
897   - before starting the parsing process. See L{I{parseString}<parseString>} for more information
898   - on parsing strings containing C{<TAB>}s, and suggested methods to maintain a
899   - consistent view of the parsed string, the parse location, and line and column
900   - positions within the parsed string.
901   - """
902   - self.parseAction = list(map(_trim_arity, list(fns)))
903   - self.callDuringTry = ("callDuringTry" in kwargs and kwargs["callDuringTry"])
904   - return self
905   -
906   - def addParseAction( self, *fns, **kwargs ):
907   - """Add parse action to expression's list of parse actions. See L{I{setParseAction}<setParseAction>}."""
908   - self.parseAction += list(map(_trim_arity, list(fns)))
909   - self.callDuringTry = self.callDuringTry or ("callDuringTry" in kwargs and kwargs["callDuringTry"])
910   - return self
911   -
912   - def setFailAction( self, fn ):
913   - """Define action to perform if parsing fails at this expression.
914   - Fail acton fn is a callable function that takes the arguments
915   - C{fn(s,loc,expr,err)} where:
916   - - s = string being parsed
917   - - loc = location where expression match was attempted and failed
918   - - expr = the parse expression that failed
919   - - err = the exception thrown
920   - The function returns no value. It may throw C{L{ParseFatalException}}
921   - if it is desired to stop parsing immediately."""
922   - self.failAction = fn
923   - return self
924   -
925   - def _skipIgnorables( self, instring, loc ):
926   - exprsFound = True
927   - while exprsFound:
928   - exprsFound = False
929   - for e in self.ignoreExprs:
930   - try:
931   - while 1:
932   - loc,dummy = e._parse( instring, loc )
933   - exprsFound = True
934   - except ParseException:
935   - pass
936   - return loc
937   -
938   - def preParse( self, instring, loc ):
939   - if self.ignoreExprs:
940   - loc = self._skipIgnorables( instring, loc )
941   -
942   - if self.skipWhitespace:
943   - wt = self.whiteChars
944   - instrlen = len(instring)
945   - while loc < instrlen and instring[loc] in wt:
946   - loc += 1
947   -
948   - return loc
949   -
950   - def parseImpl( self, instring, loc, doActions=True ):
951   - return loc, []
952   -
953   - def postParse( self, instring, loc, tokenlist ):
954   - return tokenlist
955   -
956   - #~ @profile
957   - def _parseNoCache( self, instring, loc, doActions=True, callPreParse=True ):
958   - debugging = ( self.debug ) #and doActions )
959   -
960   - if debugging or self.failAction:
961   - #~ print ("Match",self,"at loc",loc,"(%d,%d)" % ( lineno(loc,instring), col(loc,instring) ))
962   - if (self.debugActions[0] ):
963   - self.debugActions[0]( instring, loc, self )
964   - if callPreParse and self.callPreparse:
965   - preloc = self.preParse( instring, loc )
966   - else:
967   - preloc = loc
968   - tokensStart = preloc
969   - try:
970   - try:
971   - loc,tokens = self.parseImpl( instring, preloc, doActions )
972   - except IndexError:
973   - raise ParseException( instring, len(instring), self.errmsg, self )
974   - except ParseBaseException as err:
975   - #~ print ("Exception raised:", err)
976   - if self.debugActions[2]:
977   - self.debugActions[2]( instring, tokensStart, self, err )
978   - if self.failAction:
979   - self.failAction( instring, tokensStart, self, err )
980   - raise
981   - else:
982   - if callPreParse and self.callPreparse:
983   - preloc = self.preParse( instring, loc )
984   - else:
985   - preloc = loc
986   - tokensStart = preloc
987   - if self.mayIndexError or loc >= len(instring):
988   - try:
989   - loc,tokens = self.parseImpl( instring, preloc, doActions )
990   - except IndexError:
991   - raise ParseException( instring, len(instring), self.errmsg, self )
992   - else:
993   - loc,tokens = self.parseImpl( instring, preloc, doActions )
994   -
995   - tokens = self.postParse( instring, loc, tokens )
996   -
997   - retTokens = ParseResults( tokens, self.resultsName, asList=self.saveAsList, modal=self.modalResults )
998   - if self.parseAction and (doActions or self.callDuringTry):
999   - if debugging:
1000   - try:
1001   - for fn in self.parseAction:
1002   - tokens = fn( instring, tokensStart, retTokens )
1003   - if tokens is not None:
1004   - retTokens = ParseResults( tokens,
1005   - self.resultsName,
1006   - asList=self.saveAsList and isinstance(tokens,(ParseResults,list)),
1007   - modal=self.modalResults )
1008   - except ParseBaseException as err:
1009   - #~ print "Exception raised in user parse action:", err
1010   - if (self.debugActions[2] ):
1011   - self.debugActions[2]( instring, tokensStart, self, err )
1012   - raise
1013   - else:
1014   - for fn in self.parseAction:
1015   - tokens = fn( instring, tokensStart, retTokens )
1016   - if tokens is not None:
1017   - retTokens = ParseResults( tokens,
1018   - self.resultsName,
1019   - asList=self.saveAsList and isinstance(tokens,(ParseResults,list)),
1020   - modal=self.modalResults )
1021   -
1022   - if debugging:
1023   - #~ print ("Matched",self,"->",retTokens.asList())
1024   - if (self.debugActions[1] ):
1025   - self.debugActions[1]( instring, tokensStart, loc, self, retTokens )
1026   -
1027   - return loc, retTokens
1028   -
1029   - def tryParse( self, instring, loc ):
1030   - try:
1031   - return self._parse( instring, loc, doActions=False )[0]
1032   - except ParseFatalException:
1033   - raise ParseException( instring, loc, self.errmsg, self)
1034   -
1035   - # this method gets repeatedly called during backtracking with the same arguments -
1036   - # we can cache these arguments and save ourselves the trouble of re-parsing the contained expression
1037   - def _parseCache( self, instring, loc, doActions=True, callPreParse=True ):
1038   - lookup = (self,instring,loc,callPreParse,doActions)
1039   - if lookup in ParserElement._exprArgCache:
1040   - value = ParserElement._exprArgCache[ lookup ]
1041   - if isinstance(value, Exception):
1042   - raise value
1043   - return (value[0],value[1].copy())
1044   - else:
1045   - try:
1046   - value = self._parseNoCache( instring, loc, doActions, callPreParse )
1047   - ParserElement._exprArgCache[ lookup ] = (value[0],value[1].copy())
1048   - return value
1049   - except ParseBaseException as pe:
1050   - pe.__traceback__ = None
1051   - ParserElement._exprArgCache[ lookup ] = pe
1052   - raise
1053   -
1054   - _parse = _parseNoCache
1055   -
1056   - # argument cache for optimizing repeated calls when backtracking through recursive expressions
1057   - _exprArgCache = {}
1058   - def resetCache():
1059   - ParserElement._exprArgCache.clear()
1060   - resetCache = staticmethod(resetCache)
1061   -
1062   - _packratEnabled = False
1063   - def enablePackrat():
1064   - """Enables "packrat" parsing, which adds memoizing to the parsing logic.
1065   - Repeated parse attempts at the same string location (which happens
1066   - often in many complex grammars) can immediately return a cached value,
1067   - instead of re-executing parsing/validating code. Memoizing is done of
1068   - both valid results and parsing exceptions.
1069   -
1070   - This speedup may break existing programs that use parse actions that
1071   - have side-effects. For this reason, packrat parsing is disabled when
1072   - you first import pyparsing. To activate the packrat feature, your
1073   - program must call the class method C{ParserElement.enablePackrat()}. If
1074   - your program uses C{psyco} to "compile as you go", you must call
1075   - C{enablePackrat} before calling C{psyco.full()}. If you do not do this,
1076   - Python will crash. For best results, call C{enablePackrat()} immediately
1077   - after importing pyparsing.
1078   - """
1079   - if not ParserElement._packratEnabled:
1080   - ParserElement._packratEnabled = True
1081   - ParserElement._parse = ParserElement._parseCache
1082   - enablePackrat = staticmethod(enablePackrat)
1083   -
1084   - def parseString( self, instring, parseAll=False ):
1085   - """Execute the parse expression with the given string.
1086   - This is the main interface to the client code, once the complete
1087   - expression has been built.
1088   -
1089   - If you want the grammar to require that the entire input string be
1090   - successfully parsed, then set C{parseAll} to True (equivalent to ending
1091   - the grammar with C{L{StringEnd()}}).
1092   -
1093   - Note: C{parseString} implicitly calls C{expandtabs()} on the input string,
1094   - in order to report proper column numbers in parse actions.
1095   - If the input string contains tabs and
1096   - the grammar uses parse actions that use the C{loc} argument to index into the
1097   - string being parsed, you can ensure you have a consistent view of the input
1098   - string by:
1099   - - calling C{parseWithTabs} on your grammar before calling C{parseString}
1100   - (see L{I{parseWithTabs}<parseWithTabs>})
1101   - - define your parse action using the full C{(s,loc,toks)} signature, and
1102   - reference the input string using the parse action's C{s} argument
1103   - - explictly expand the tabs in your input string before calling
1104   - C{parseString}
1105   - """
1106   - ParserElement.resetCache()
1107   - if not self.streamlined:
1108   - self.streamline()
1109   - #~ self.saveAsList = True
1110   - for e in self.ignoreExprs:
1111   - e.streamline()
1112   - if not self.keepTabs:
1113   - instring = instring.expandtabs()
1114   - try:
1115   - loc, tokens = self._parse( instring, 0 )
1116   - if parseAll:
1117   - loc = self.preParse( instring, loc )
1118   - se = Empty() + StringEnd()
1119   - se._parse( instring, loc )
1120   - except ParseBaseException as exc:
1121   - if ParserElement.verbose_stacktrace:
1122   - raise
1123   - else:
1124   - # catch and re-raise exception from here, clears out pyparsing internal stack trace
1125   - raise exc
1126   - else:
1127   - return tokens
1128   -
1129   - def scanString( self, instring, maxMatches=_MAX_INT, overlap=False ):
1130   - """Scan the input string for expression matches. Each match will return the
1131   - matching tokens, start location, and end location. May be called with optional
1132   - C{maxMatches} argument, to clip scanning after 'n' matches are found. If
1133   - C{overlap} is specified, then overlapping matches will be reported.
1134   -
1135   - Note that the start and end locations are reported relative to the string
1136   - being parsed. See L{I{parseString}<parseString>} for more information on parsing
1137   - strings with embedded tabs."""
1138   - if not self.streamlined:
1139   - self.streamline()
1140   - for e in self.ignoreExprs:
1141   - e.streamline()
1142   -
1143   - if not self.keepTabs:
1144   - instring = _ustr(instring).expandtabs()
1145   - instrlen = len(instring)
1146   - loc = 0
1147   - preparseFn = self.preParse
1148   - parseFn = self._parse
1149   - ParserElement.resetCache()
1150   - matches = 0
1151   - try:
1152   - while loc <= instrlen and matches < maxMatches:
1153   - try:
1154   - preloc = preparseFn( instring, loc )
1155   - nextLoc,tokens = parseFn( instring, preloc, callPreParse=False )
1156   - except ParseException:
1157   - loc = preloc+1
1158   - else:
1159   - if nextLoc > loc:
1160   - matches += 1
1161   - yield tokens, preloc, nextLoc
1162   - if overlap:
1163   - nextloc = preparseFn( instring, loc )
1164   - if nextloc > loc:
1165   - loc = nextLoc
1166   - else:
1167   - loc += 1
1168   - else:
1169   - loc = nextLoc
1170   - else:
1171   - loc = preloc+1
1172   - except ParseBaseException as exc:
1173   - if ParserElement.verbose_stacktrace:
1174   - raise
1175   - else:
1176   - # catch and re-raise exception from here, clears out pyparsing internal stack trace
1177   - raise exc
1178   -
1179   - def transformString( self, instring ):
1180   - """Extension to C{L{scanString}}, to modify matching text with modified tokens that may
1181   - be returned from a parse action. To use C{transformString}, define a grammar and
1182   - attach a parse action to it that modifies the returned token list.
1183   - Invoking C{transformString()} on a target string will then scan for matches,
1184   - and replace the matched text patterns according to the logic in the parse
1185   - action. C{transformString()} returns the resulting transformed string."""
1186   - out = []
1187   - lastE = 0
1188   - # force preservation of <TAB>s, to minimize unwanted transformation of string, and to
1189   - # keep string locs straight between transformString and scanString
1190   - self.keepTabs = True
1191   - try:
1192   - for t,s,e in self.scanString( instring ):
1193   - out.append( instring[lastE:s] )
1194   - if t:
1195   - if isinstance(t,ParseResults):
1196   - out += t.asList()
1197   - elif isinstance(t,list):
1198   - out += t
1199   - else:
1200   - out.append(t)
1201   - lastE = e
1202   - out.append(instring[lastE:])
1203   - out = [o for o in out if o]
1204   - return "".join(map(_ustr,_flatten(out)))
1205   - except ParseBaseException as exc:
1206   - if ParserElement.verbose_stacktrace:
1207   - raise
1208   - else:
1209   - # catch and re-raise exception from here, clears out pyparsing internal stack trace
1210   - raise exc
1211   -
1212   - def searchString( self, instring, maxMatches=_MAX_INT ):
1213   - """Another extension to C{L{scanString}}, simplifying the access to the tokens found
1214   - to match the given parse expression. May be called with optional
1215   - C{maxMatches} argument, to clip searching after 'n' matches are found.
1216   - """
1217   - try:
1218   - return ParseResults([ t for t,s,e in self.scanString( instring, maxMatches ) ])
1219   - except ParseBaseException as exc:
1220   - if ParserElement.verbose_stacktrace:
1221   - raise
1222   - else:
1223   - # catch and re-raise exception from here, clears out pyparsing internal stack trace
1224   - raise exc
1225   -
1226   - def __add__(self, other ):
1227   - """Implementation of + operator - returns C{L{And}}"""
1228   - if isinstance( other, basestring ):
1229   - other = ParserElement.literalStringClass( other )
1230   - if not isinstance( other, ParserElement ):
1231   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1232   - SyntaxWarning, stacklevel=2)
1233   - return None
1234   - return And( [ self, other ] )
1235   -
1236   - def __radd__(self, other ):
1237   - """Implementation of + operator when left operand is not a C{L{ParserElement}}"""
1238   - if isinstance( other, basestring ):
1239   - other = ParserElement.literalStringClass( other )
1240   - if not isinstance( other, ParserElement ):
1241   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1242   - SyntaxWarning, stacklevel=2)
1243   - return None
1244   - return other + self
1245   -
1246   - def __sub__(self, other):
1247   - """Implementation of - operator, returns C{L{And}} with error stop"""
1248   - if isinstance( other, basestring ):
1249   - other = ParserElement.literalStringClass( other )
1250   - if not isinstance( other, ParserElement ):
1251   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1252   - SyntaxWarning, stacklevel=2)
1253   - return None
1254   - return And( [ self, And._ErrorStop(), other ] )
1255   -
1256   - def __rsub__(self, other ):
1257   - """Implementation of - operator when left operand is not a C{L{ParserElement}}"""
1258   - if isinstance( other, basestring ):
1259   - other = ParserElement.literalStringClass( other )
1260   - if not isinstance( other, ParserElement ):
1261   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1262   - SyntaxWarning, stacklevel=2)
1263   - return None
1264   - return other - self
1265   -
1266   - def __mul__(self,other):
1267   - """Implementation of * operator, allows use of C{expr * 3} in place of
1268   - C{expr + expr + expr}. Expressions may also me multiplied by a 2-integer
1269   - tuple, similar to C{{min,max}} multipliers in regular expressions. Tuples
1270   - may also include C{None} as in:
1271   - - C{expr*(n,None)} or C{expr*(n,)} is equivalent
1272   - to C{expr*n + L{ZeroOrMore}(expr)}
1273   - (read as "at least n instances of C{expr}")
1274   - - C{expr*(None,n)} is equivalent to C{expr*(0,n)}
1275   - (read as "0 to n instances of C{expr}")
1276   - - C{expr*(None,None)} is equivalent to C{L{ZeroOrMore}(expr)}
1277   - - C{expr*(1,None)} is equivalent to C{L{OneOrMore}(expr)}
1278   -
1279   - Note that C{expr*(None,n)} does not raise an exception if
1280   - more than n exprs exist in the input stream; that is,
1281   - C{expr*(None,n)} does not enforce a maximum number of expr
1282   - occurrences. If this behavior is desired, then write
1283   - C{expr*(None,n) + ~expr}
1284   -
1285   - """
1286   - if isinstance(other,int):
1287   - minElements, optElements = other,0
1288   - elif isinstance(other,tuple):
1289   - other = (other + (None, None))[:2]
1290   - if other[0] is None:
1291   - other = (0, other[1])
1292   - if isinstance(other[0],int) and other[1] is None:
1293   - if other[0] == 0:
1294   - return ZeroOrMore(self)
1295   - if other[0] == 1:
1296   - return OneOrMore(self)
1297   - else:
1298   - return self*other[0] + ZeroOrMore(self)
1299   - elif isinstance(other[0],int) and isinstance(other[1],int):
1300   - minElements, optElements = other
1301   - optElements -= minElements
1302   - else:
1303   - raise TypeError("cannot multiply 'ParserElement' and ('%s','%s') objects", type(other[0]),type(other[1]))
1304   - else:
1305   - raise TypeError("cannot multiply 'ParserElement' and '%s' objects", type(other))
1306   -
1307   - if minElements < 0:
1308   - raise ValueError("cannot multiply ParserElement by negative value")
1309   - if optElements < 0:
1310   - raise ValueError("second tuple value must be greater or equal to first tuple value")
1311   - if minElements == optElements == 0:
1312   - raise ValueError("cannot multiply ParserElement by 0 or (0,0)")
1313   -
1314   - if (optElements):
1315   - def makeOptionalList(n):
1316   - if n>1:
1317   - return Optional(self + makeOptionalList(n-1))
1318   - else:
1319   - return Optional(self)
1320   - if minElements:
1321   - if minElements == 1:
1322   - ret = self + makeOptionalList(optElements)
1323   - else:
1324   - ret = And([self]*minElements) + makeOptionalList(optElements)
1325   - else:
1326   - ret = makeOptionalList(optElements)
1327   - else:
1328   - if minElements == 1:
1329   - ret = self
1330   - else:
1331   - ret = And([self]*minElements)
1332   - return ret
1333   -
1334   - def __rmul__(self, other):
1335   - return self.__mul__(other)
1336   -
1337   - def __or__(self, other ):
1338   - """Implementation of | operator - returns C{L{MatchFirst}}"""
1339   - if isinstance( other, basestring ):
1340   - other = ParserElement.literalStringClass( other )
1341   - if not isinstance( other, ParserElement ):
1342   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1343   - SyntaxWarning, stacklevel=2)
1344   - return None
1345   - return MatchFirst( [ self, other ] )
1346   -
1347   - def __ror__(self, other ):
1348   - """Implementation of | operator when left operand is not a C{L{ParserElement}}"""
1349   - if isinstance( other, basestring ):
1350   - other = ParserElement.literalStringClass( other )
1351   - if not isinstance( other, ParserElement ):
1352   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1353   - SyntaxWarning, stacklevel=2)
1354   - return None
1355   - return other | self
1356   -
1357   - def __xor__(self, other ):
1358   - """Implementation of ^ operator - returns C{L{Or}}"""
1359   - if isinstance( other, basestring ):
1360   - other = ParserElement.literalStringClass( other )
1361   - if not isinstance( other, ParserElement ):
1362   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1363   - SyntaxWarning, stacklevel=2)
1364   - return None
1365   - return Or( [ self, other ] )
1366   -
1367   - def __rxor__(self, other ):
1368   - """Implementation of ^ operator when left operand is not a C{L{ParserElement}}"""
1369   - if isinstance( other, basestring ):
1370   - other = ParserElement.literalStringClass( other )
1371   - if not isinstance( other, ParserElement ):
1372   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1373   - SyntaxWarning, stacklevel=2)
1374   - return None
1375   - return other ^ self
1376   -
1377   - def __and__(self, other ):
1378   - """Implementation of & operator - returns C{L{Each}}"""
1379   - if isinstance( other, basestring ):
1380   - other = ParserElement.literalStringClass( other )
1381   - if not isinstance( other, ParserElement ):
1382   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1383   - SyntaxWarning, stacklevel=2)
1384   - return None
1385   - return Each( [ self, other ] )
1386   -
1387   - def __rand__(self, other ):
1388   - """Implementation of & operator when left operand is not a C{L{ParserElement}}"""
1389   - if isinstance( other, basestring ):
1390   - other = ParserElement.literalStringClass( other )
1391   - if not isinstance( other, ParserElement ):
1392   - warnings.warn("Cannot combine element of type %s with ParserElement" % type(other),
1393   - SyntaxWarning, stacklevel=2)
1394   - return None
1395   - return other & self
1396   -
1397   - def __invert__( self ):
1398   - """Implementation of ~ operator - returns C{L{NotAny}}"""
1399   - return NotAny( self )
1400   -
1401   - def __call__(self, name=None):
1402   - """Shortcut for C{L{setResultsName}}, with C{listAllMatches=default}::
1403   - userdata = Word(alphas).setResultsName("name") + Word(nums+"-").setResultsName("socsecno")
1404   - could be written as::
1405   - userdata = Word(alphas)("name") + Word(nums+"-")("socsecno")
1406   -
1407   - If C{name} is given with a trailing C{'*'} character, then C{listAllMatches} will be
1408   - passed as C{True}.
1409   -
1410   - If C{name} is omitted, same as calling C{L{copy}}.
1411   - """
1412   - if name is not None:
1413   - return self.setResultsName(name)
1414   - else:
1415   - return self.copy()
1416   -
1417   - def suppress( self ):
1418   - """Suppresses the output of this C{ParserElement}; useful to keep punctuation from
1419   - cluttering up returned output.
1420   - """
1421   - return Suppress( self )
1422   -
1423   - def leaveWhitespace( self ):
1424   - """Disables the skipping of whitespace before matching the characters in the
1425   - C{ParserElement}'s defined pattern. This is normally only used internally by
1426   - the pyparsing module, but may be needed in some whitespace-sensitive grammars.
1427   - """
1428   - self.skipWhitespace = False
1429   - return self
1430   -
1431   - def setWhitespaceChars( self, chars ):
1432   - """Overrides the default whitespace chars
1433   - """
1434   - self.skipWhitespace = True
1435   - self.whiteChars = chars
1436   - self.copyDefaultWhiteChars = False
1437   - return self
1438   -
1439   - def parseWithTabs( self ):
1440   - """Overrides default behavior to expand C{<TAB>}s to spaces before parsing the input string.
1441   - Must be called before C{parseString} when the input grammar contains elements that
1442   - match C{<TAB>} characters."""
1443   - self.keepTabs = True
1444   - return self
1445   -
1446   - def ignore( self, other ):
1447   - """Define expression to be ignored (e.g., comments) while doing pattern
1448   - matching; may be called repeatedly, to define multiple comment or other
1449   - ignorable patterns.
1450   - """
1451   - if isinstance( other, Suppress ):
1452   - if other not in self.ignoreExprs:
1453   - self.ignoreExprs.append( other.copy() )
1454   - else:
1455   - self.ignoreExprs.append( Suppress( other.copy() ) )
1456   - return self
1457   -
1458   - def setDebugActions( self, startAction, successAction, exceptionAction ):
1459   - """Enable display of debugging messages while doing pattern matching."""
1460   - self.debugActions = (startAction or _defaultStartDebugAction,
1461   - successAction or _defaultSuccessDebugAction,
1462   - exceptionAction or _defaultExceptionDebugAction)
1463   - self.debug = True
1464   - return self
1465   -
1466   - def setDebug( self, flag=True ):
1467   - """Enable display of debugging messages while doing pattern matching.
1468   - Set C{flag} to True to enable, False to disable."""
1469   - if flag:
1470   - self.setDebugActions( _defaultStartDebugAction, _defaultSuccessDebugAction, _defaultExceptionDebugAction )
1471   - else:
1472   - self.debug = False
1473   - return self
1474   -
1475   - def __str__( self ):
1476   - return self.name
1477   -
1478   - def __repr__( self ):
1479   - return _ustr(self)
1480   -
1481   - def streamline( self ):
1482   - self.streamlined = True
1483   - self.strRepr = None
1484   - return self
1485   -
1486   - def checkRecursion( self, parseElementList ):
1487   - pass
1488   -
1489   - def validate( self, validateTrace=[] ):
1490   - """Check defined expressions for valid structure, check for infinite recursive definitions."""
1491   - self.checkRecursion( [] )
1492   -
1493   - def parseFile( self, file_or_filename, parseAll=False ):
1494   - """Execute the parse expression on the given file or filename.
1495   - If a filename is specified (instead of a file object),
1496   - the entire file is opened, read, and closed before parsing.
1497   - """
1498   - try:
1499   - file_contents = file_or_filename.read()
1500   - except AttributeError:
1501   - f = open(file_or_filename, "r")
1502   - file_contents = f.read()
1503   - f.close()
1504   - try:
1505   - return self.parseString(file_contents, parseAll)
1506   - except ParseBaseException as exc:
1507   - if ParserElement.verbose_stacktrace:
1508   - raise
1509   - else:
1510   - # catch and re-raise exception from here, clears out pyparsing internal stack trace
1511   - raise exc
1512   -
1513   - def __eq__(self,other):
1514   - if isinstance(other, ParserElement):
1515   - return self is other or self.__dict__ == other.__dict__
1516   - elif isinstance(other, basestring):
1517   - try:
1518   - self.parseString(_ustr(other), parseAll=True)
1519   - return True
1520   - except ParseBaseException:
1521   - return False
1522   - else:
1523   - return super(ParserElement,self)==other
1524   -
1525   - def __ne__(self,other):
1526   - return not (self == other)
1527   -
1528   - def __hash__(self):
1529   - return hash(id(self))
1530   -
1531   - def __req__(self,other):
1532   - return self == other
1533   -
1534   - def __rne__(self,other):
1535   - return not (self == other)
1536   -
1537   -
1538   -class Token(ParserElement):
1539   - """Abstract C{ParserElement} subclass, for defining atomic matching patterns."""
1540   - def __init__( self ):
1541   - super(Token,self).__init__( savelist=False )
1542   -
1543   - def setName(self, name):
1544   - s = super(Token,self).setName(name)
1545   - self.errmsg = "Expected " + self.name
1546   - return s
1547   -
1548   -
1549   -class Empty(Token):
1550   - """An empty token, will always match."""
1551   - def __init__( self ):
1552   - super(Empty,self).__init__()
1553   - self.name = "Empty"
1554   - self.mayReturnEmpty = True
1555   - self.mayIndexError = False
1556   -
1557   -
1558   -class NoMatch(Token):
1559   - """A token that will never match."""
1560   - def __init__( self ):
1561   - super(NoMatch,self).__init__()
1562   - self.name = "NoMatch"
1563   - self.mayReturnEmpty = True
1564   - self.mayIndexError = False
1565   - self.errmsg = "Unmatchable token"
1566   -
1567   - def parseImpl( self, instring, loc, doActions=True ):
1568   - raise ParseException(instring, loc, self.errmsg, self)
1569   -
1570   -
1571   -class Literal(Token):
1572   - """Token to exactly match a specified string."""
1573   - def __init__( self, matchString ):
1574   - super(Literal,self).__init__()
1575   - self.match = matchString
1576   - self.matchLen = len(matchString)
1577   - try:
1578   - self.firstMatchChar = matchString[0]
1579   - except IndexError:
1580   - warnings.warn("null string passed to Literal; use Empty() instead",
1581   - SyntaxWarning, stacklevel=2)
1582   - self.__class__ = Empty
1583   - self.name = '"%s"' % _ustr(self.match)
1584   - self.errmsg = "Expected " + self.name
1585   - self.mayReturnEmpty = False
1586   - self.mayIndexError = False
1587   -
1588   - # Performance tuning: this routine gets called a *lot*
1589   - # if this is a single character match string and the first character matches,
1590   - # short-circuit as quickly as possible, and avoid calling startswith
1591   - #~ @profile
1592   - def parseImpl( self, instring, loc, doActions=True ):
1593   - if (instring[loc] == self.firstMatchChar and
1594   - (self.matchLen==1 or instring.startswith(self.match,loc)) ):
1595   - return loc+self.matchLen, self.match
1596   - raise ParseException(instring, loc, self.errmsg, self)
1597   -_L = Literal
1598   -ParserElement.literalStringClass = Literal
1599   -
1600   -class Keyword(Token):
1601   - """Token to exactly match a specified string as a keyword, that is, it must be
1602   - immediately followed by a non-keyword character. Compare with C{L{Literal}}::
1603   - Literal("if") will match the leading C{'if'} in C{'ifAndOnlyIf'}.
1604   - Keyword("if") will not; it will only match the leading C{'if'} in C{'if x=1'}, or C{'if(y==2)'}
1605   - Accepts two optional constructor arguments in addition to the keyword string:
1606   - C{identChars} is a string of characters that would be valid identifier characters,
1607   - defaulting to all alphanumerics + "_" and "$"; C{caseless} allows case-insensitive
1608   - matching, default is C{False}.
1609   - """
1610   - DEFAULT_KEYWORD_CHARS = alphanums+"_$"
1611   -
1612   - def __init__( self, matchString, identChars=DEFAULT_KEYWORD_CHARS, caseless=False ):
1613   - super(Keyword,self).__init__()
1614   - self.match = matchString
1615   - self.matchLen = len(matchString)
1616   - try:
1617   - self.firstMatchChar = matchString[0]
1618   - except IndexError:
1619   - warnings.warn("null string passed to Keyword; use Empty() instead",
1620   - SyntaxWarning, stacklevel=2)
1621   - self.name = '"%s"' % self.match
1622   - self.errmsg = "Expected " + self.name
1623   - self.mayReturnEmpty = False
1624   - self.mayIndexError = False
1625   - self.caseless = caseless
1626   - if caseless:
1627   - self.caselessmatch = matchString.upper()
1628   - identChars = identChars.upper()
1629   - self.identChars = set(identChars)
1630   -
1631   - def parseImpl( self, instring, loc, doActions=True ):
1632   - if self.caseless:
1633   - if ( (instring[ loc:loc+self.matchLen ].upper() == self.caselessmatch) and
1634   - (loc >= len(instring)-self.matchLen or instring[loc+self.matchLen].upper() not in self.identChars) and
1635   - (loc == 0 or instring[loc-1].upper() not in self.identChars) ):
1636   - return loc+self.matchLen, self.match
1637   - else:
1638   - if (instring[loc] == self.firstMatchChar and
1639   - (self.matchLen==1 or instring.startswith(self.match,loc)) and
1640   - (loc >= len(instring)-self.matchLen or instring[loc+self.matchLen] not in self.identChars) and
1641   - (loc == 0 or instring[loc-1] not in self.identChars) ):
1642   - return loc+self.matchLen, self.match
1643   - raise ParseException(instring, loc, self.errmsg, self)
1644   -
1645   - def copy(self):
1646   - c = super(Keyword,self).copy()
1647   - c.identChars = Keyword.DEFAULT_KEYWORD_CHARS
1648   - return c
1649   -
1650   - def setDefaultKeywordChars( chars ):
1651   - """Overrides the default Keyword chars
1652   - """
1653   - Keyword.DEFAULT_KEYWORD_CHARS = chars
1654   - setDefaultKeywordChars = staticmethod(setDefaultKeywordChars)
1655   -
1656   -class CaselessLiteral(Literal):
1657   - """Token to match a specified string, ignoring case of letters.
1658   - Note: the matched results will always be in the case of the given
1659   - match string, NOT the case of the input text.
1660   - """
1661   - def __init__( self, matchString ):
1662   - super(CaselessLiteral,self).__init__( matchString.upper() )
1663   - # Preserve the defining literal.
1664   - self.returnString = matchString
1665   - self.name = "'%s'" % self.returnString
1666   - self.errmsg = "Expected " + self.name
1667   -
1668   - def parseImpl( self, instring, loc, doActions=True ):
1669   - if instring[ loc:loc+self.matchLen ].upper() == self.match:
1670   - return loc+self.matchLen, self.returnString
1671   - raise ParseException(instring, loc, self.errmsg, self)
1672   -
1673   -class CaselessKeyword(Keyword):
1674   - def __init__( self, matchString, identChars=Keyword.DEFAULT_KEYWORD_CHARS ):
1675   - super(CaselessKeyword,self).__init__( matchString, identChars, caseless=True )
1676   -
1677   - def parseImpl( self, instring, loc, doActions=True ):
1678   - if ( (instring[ loc:loc+self.matchLen ].upper() == self.caselessmatch) and
1679   - (loc >= len(instring)-self.matchLen or instring[loc+self.matchLen].upper() not in self.identChars) ):
1680   - return loc+self.matchLen, self.match
1681   - raise ParseException(instring, loc, self.errmsg, self)
1682   -
1683   -class Word(Token):
1684   - """Token for matching words composed of allowed character sets.
1685   - Defined with string containing all allowed initial characters,
1686   - an optional string containing allowed body characters (if omitted,
1687   - defaults to the initial character set), and an optional minimum,
1688   - maximum, and/or exact length. The default value for C{min} is 1 (a
1689   - minimum value < 1 is not valid); the default values for C{max} and C{exact}
1690   - are 0, meaning no maximum or exact length restriction. An optional
1691   - C{exclude} parameter can list characters that might be found in
1692   - the input C{bodyChars} string; useful to define a word of all printables
1693   - except for one or two characters, for instance.
1694   - """
1695   - def __init__( self, initChars, bodyChars=None, min=1, max=0, exact=0, asKeyword=False, excludeChars=None ):
1696   - super(Word,self).__init__()
1697   - if excludeChars:
1698   - initChars = ''.join(c for c in initChars if c not in excludeChars)
1699   - if bodyChars:
1700   - bodyChars = ''.join(c for c in bodyChars if c not in excludeChars)
1701   - self.initCharsOrig = initChars
1702   - self.initChars = set(initChars)
1703   - if bodyChars :
1704   - self.bodyCharsOrig = bodyChars
1705   - self.bodyChars = set(bodyChars)
1706   - else:
1707   - self.bodyCharsOrig = initChars
1708   - self.bodyChars = set(initChars)
1709   -
1710   - self.maxSpecified = max > 0
1711   -
1712   - if min < 1:
1713   - raise ValueError("cannot specify a minimum length < 1; use Optional(Word()) if zero-length word is permitted")
1714   -
1715   - self.minLen = min
1716   -
1717   - if max > 0:
1718   - self.maxLen = max
1719   - else:
1720   - self.maxLen = _MAX_INT
1721   -
1722   - if exact > 0:
1723   - self.maxLen = exact
1724   - self.minLen = exact
1725   -
1726   - self.name = _ustr(self)
1727   - self.errmsg = "Expected " + self.name
1728   - self.mayIndexError = False
1729   - self.asKeyword = asKeyword
1730   -
1731   - if ' ' not in self.initCharsOrig+self.bodyCharsOrig and (min==1 and max==0 and exact==0):
1732   - if self.bodyCharsOrig == self.initCharsOrig:
1733   - self.reString = "[%s]+" % _escapeRegexRangeChars(self.initCharsOrig)
1734   - elif len(self.bodyCharsOrig) == 1:
1735   - self.reString = "%s[%s]*" % \
1736   - (re.escape(self.initCharsOrig),
1737   - _escapeRegexRangeChars(self.bodyCharsOrig),)
1738   - else:
1739   - self.reString = "[%s][%s]*" % \
1740   - (_escapeRegexRangeChars(self.initCharsOrig),
1741   - _escapeRegexRangeChars(self.bodyCharsOrig),)
1742   - if self.asKeyword:
1743   - self.reString = r"\b"+self.reString+r"\b"
1744   - try:
1745   - self.re = re.compile( self.reString )
1746   - except:
1747   - self.re = None
1748   -
1749   - def parseImpl( self, instring, loc, doActions=True ):
1750   - if self.re:
1751   - result = self.re.match(instring,loc)
1752   - if not result:
1753   - raise ParseException(instring, loc, self.errmsg, self)
1754   -
1755   - loc = result.end()
1756   - return loc, result.group()
1757   -
1758   - if not(instring[ loc ] in self.initChars):
1759   - raise ParseException(instring, loc, self.errmsg, self)
1760   -
1761   - start = loc
1762   - loc += 1
1763   - instrlen = len(instring)
1764   - bodychars = self.bodyChars
1765   - maxloc = start + self.maxLen
1766   - maxloc = min( maxloc, instrlen )
1767   - while loc < maxloc and instring[loc] in bodychars:
1768   - loc += 1
1769   -
1770   - throwException = False
1771   - if loc - start < self.minLen:
1772   - throwException = True
1773   - if self.maxSpecified and loc < instrlen and instring[loc] in bodychars:
1774   - throwException = True
1775   - if self.asKeyword:
1776   - if (start>0 and instring[start-1] in bodychars) or (loc<instrlen and instring[loc] in bodychars):
1777   - throwException = True
1778   -
1779   - if throwException:
1780   - raise ParseException(instring, loc, self.errmsg, self)
1781   -
1782   - return loc, instring[start:loc]
1783   -
1784   - def __str__( self ):
1785   - try:
1786   - return super(Word,self).__str__()
1787   - except:
1788   - pass
1789   -
1790   -
1791   - if self.strRepr is None:
1792   -
1793   - def charsAsStr(s):
1794   - if len(s)>4:
1795   - return s[:4]+"..."
1796   - else:
1797   - return s
1798   -
1799   - if ( self.initCharsOrig != self.bodyCharsOrig ):
1800   - self.strRepr = "W:(%s,%s)" % ( charsAsStr(self.initCharsOrig), charsAsStr(self.bodyCharsOrig) )
1801   - else:
1802   - self.strRepr = "W:(%s)" % charsAsStr(self.initCharsOrig)
1803   -
1804   - return self.strRepr
1805   -
1806   -
1807   -class Regex(Token):
1808   - """Token for matching strings that match a given regular expression.
1809   - Defined with string specifying the regular expression in a form recognized by the inbuilt Python re module.
1810   - """
1811   - compiledREtype = type(re.compile("[A-Z]"))
1812   - def __init__( self, pattern, flags=0):
1813   - """The parameters C{pattern} and C{flags} are passed to the C{re.compile()} function as-is. See the Python C{re} module for an explanation of the acceptable patterns and flags."""
1814   - super(Regex,self).__init__()
1815   -
1816   - if isinstance(pattern, basestring):
1817   - if len(pattern) == 0:
1818   - warnings.warn("null string passed to Regex; use Empty() instead",
1819   - SyntaxWarning, stacklevel=2)
1820   -
1821   - self.pattern = pattern
1822   - self.flags = flags
1823   -
1824   - try:
1825   - self.re = re.compile(self.pattern, self.flags)
1826   - self.reString = self.pattern
1827   - except sre_constants.error:
1828   - warnings.warn("invalid pattern (%s) passed to Regex" % pattern,
1829   - SyntaxWarning, stacklevel=2)
1830   - raise
1831   -
1832   - elif isinstance(pattern, Regex.compiledREtype):
1833   - self.re = pattern
1834   - self.pattern = \
1835   - self.reString = str(pattern)
1836   - self.flags = flags
1837   -
1838   - else:
1839   - raise ValueError("Regex may only be constructed with a string or a compiled RE object")
1840   -
1841   - self.name = _ustr(self)
1842   - self.errmsg = "Expected " + self.name
1843   - self.mayIndexError = False
1844   - self.mayReturnEmpty = True
1845   -
1846   - def parseImpl( self, instring, loc, doActions=True ):
1847   - result = self.re.match(instring,loc)
1848   - if not result:
1849   - raise ParseException(instring, loc, self.errmsg, self)
1850   -
1851   - loc = result.end()
1852   - d = result.groupdict()
1853   - ret = ParseResults(result.group())
1854   - if d:
1855   - for k in d:
1856   - ret[k] = d[k]
1857   - return loc,ret
1858   -
1859   - def __str__( self ):
1860   - try:
1861   - return super(Regex,self).__str__()
1862   - except:
1863   - pass
1864   -
1865   - if self.strRepr is None:
1866   - self.strRepr = "Re:(%s)" % repr(self.pattern)
1867   -
1868   - return self.strRepr
1869   -
1870   -
1871   -class QuotedString(Token):
1872   - """Token for matching strings that are delimited by quoting characters.
1873   - """
1874   - def __init__( self, quoteChar, escChar=None, escQuote=None, multiline=False, unquoteResults=True, endQuoteChar=None):
1875   - """
1876   - Defined with the following parameters:
1877   - - quoteChar - string of one or more characters defining the quote delimiting string
1878   - - escChar - character to escape quotes, typically backslash (default=None)
1879   - - escQuote - special quote sequence to escape an embedded quote string (such as SQL's "" to escape an embedded ") (default=None)
1880   - - multiline - boolean indicating whether quotes can span multiple lines (default=C{False})
1881   - - unquoteResults - boolean indicating whether the matched text should be unquoted (default=C{True})
1882   - - endQuoteChar - string of one or more characters defining the end of the quote delimited string (default=C{None} => same as quoteChar)
1883   - """
1884   - super(QuotedString,self).__init__()
1885   -
1886   - # remove white space from quote chars - wont work anyway
1887   - quoteChar = quoteChar.strip()
1888   - if len(quoteChar) == 0:
1889   - warnings.warn("quoteChar cannot be the empty string",SyntaxWarning,stacklevel=2)
1890   - raise SyntaxError()
1891   -
1892   - if endQuoteChar is None:
1893   - endQuoteChar = quoteChar
1894   - else:
1895   - endQuoteChar = endQuoteChar.strip()
1896   - if len(endQuoteChar) == 0:
1897   - warnings.warn("endQuoteChar cannot be the empty string",SyntaxWarning,stacklevel=2)
1898   - raise SyntaxError()
1899   -
1900   - self.quoteChar = quoteChar
1901   - self.quoteCharLen = len(quoteChar)
1902   - self.firstQuoteChar = quoteChar[0]
1903   - self.endQuoteChar = endQuoteChar
1904   - self.endQuoteCharLen = len(endQuoteChar)
1905   - self.escChar = escChar
1906   - self.escQuote = escQuote
1907   - self.unquoteResults = unquoteResults
1908   -
1909   - if multiline:
1910   - self.flags = re.MULTILINE | re.DOTALL
1911   - self.pattern = r'%s(?:[^%s%s]' % \
1912   - ( re.escape(self.quoteChar),
1913   - _escapeRegexRangeChars(self.endQuoteChar[0]),
1914   - (escChar is not None and _escapeRegexRangeChars(escChar) or '') )
1915   - else:
1916   - self.flags = 0
1917   - self.pattern = r'%s(?:[^%s\n\r%s]' % \
1918   - ( re.escape(self.quoteChar),
1919   - _escapeRegexRangeChars(self.endQuoteChar[0]),
1920   - (escChar is not None and _escapeRegexRangeChars(escChar) or '') )
1921   - if len(self.endQuoteChar) > 1:
1922   - self.pattern += (
1923   - '|(?:' + ')|(?:'.join("%s[^%s]" % (re.escape(self.endQuoteChar[:i]),
1924   - _escapeRegexRangeChars(self.endQuoteChar[i]))
1925   - for i in range(len(self.endQuoteChar)-1,0,-1)) + ')'
1926   - )
1927   - if escQuote:
1928   - self.pattern += (r'|(?:%s)' % re.escape(escQuote))
1929   - if escChar:
1930   - self.pattern += (r'|(?:%s.)' % re.escape(escChar))
1931   - self.escCharReplacePattern = re.escape(self.escChar)+"(.)"
1932   - self.pattern += (r')*%s' % re.escape(self.endQuoteChar))
1933   -
1934   - try:
1935   - self.re = re.compile(self.pattern, self.flags)
1936   - self.reString = self.pattern
1937   - except sre_constants.error:
1938   - warnings.warn("invalid pattern (%s) passed to Regex" % self.pattern,
1939   - SyntaxWarning, stacklevel=2)
1940   - raise
1941   -
1942   - self.name = _ustr(self)
1943   - self.errmsg = "Expected " + self.name
1944   - self.mayIndexError = False
1945   - self.mayReturnEmpty = True
1946   -
1947   - def parseImpl( self, instring, loc, doActions=True ):
1948   - result = instring[loc] == self.firstQuoteChar and self.re.match(instring,loc) or None
1949   - if not result:
1950   - raise ParseException(instring, loc, self.errmsg, self)
1951   -
1952   - loc = result.end()
1953   - ret = result.group()
1954   -
1955   - if self.unquoteResults:
1956   -
1957   - # strip off quotes
1958   - ret = ret[self.quoteCharLen:-self.endQuoteCharLen]
1959   -
1960   - if isinstance(ret,basestring):
1961   - # replace escaped characters
1962   - if self.escChar:
1963   - ret = re.sub(self.escCharReplacePattern,"\g<1>",ret)
1964   -
1965   - # replace escaped quotes
1966   - if self.escQuote:
1967   - ret = ret.replace(self.escQuote, self.endQuoteChar)
1968   -
1969   - return loc, ret
1970   -
1971   - def __str__( self ):
1972   - try:
1973   - return super(QuotedString,self).__str__()
1974   - except:
1975   - pass
1976   -
1977   - if self.strRepr is None:
1978   - self.strRepr = "quoted string, starting with %s ending with %s" % (self.quoteChar, self.endQuoteChar)
1979   -
1980   - return self.strRepr
1981   -
1982   -
1983   -class CharsNotIn(Token):
1984   - """Token for matching words composed of characters *not* in a given set.
1985   - Defined with string containing all disallowed characters, and an optional
1986   - minimum, maximum, and/or exact length. The default value for C{min} is 1 (a
1987   - minimum value < 1 is not valid); the default values for C{max} and C{exact}
1988   - are 0, meaning no maximum or exact length restriction.
1989   - """
1990   - def __init__( self, notChars, min=1, max=0, exact=0 ):
1991   - super(CharsNotIn,self).__init__()
1992   - self.skipWhitespace = False
1993   - self.notChars = notChars
1994   -
1995   - if min < 1:
1996   - raise ValueError("cannot specify a minimum length < 1; use Optional(CharsNotIn()) if zero-length char group is permitted")
1997   -
1998   - self.minLen = min
1999   -
2000   - if max > 0:
2001   - self.maxLen = max
2002   - else:
2003   - self.maxLen = _MAX_INT
2004   -
2005   - if exact > 0:
2006   - self.maxLen = exact
2007   - self.minLen = exact
2008   -
2009   - self.name = _ustr(self)
2010   - self.errmsg = "Expected " + self.name
2011   - self.mayReturnEmpty = ( self.minLen == 0 )
2012   - self.mayIndexError = False
2013   -
2014   - def parseImpl( self, instring, loc, doActions=True ):
2015   - if instring[loc] in self.notChars:
2016   - raise ParseException(instring, loc, self.errmsg, self)
2017   -
2018   - start = loc
2019   - loc += 1
2020   - notchars = self.notChars
2021   - maxlen = min( start+self.maxLen, len(instring) )
2022   - while loc < maxlen and \
2023   - (instring[loc] not in notchars):
2024   - loc += 1
2025   -
2026   - if loc - start < self.minLen:
2027   - raise ParseException(instring, loc, self.errmsg, self)
2028   -
2029   - return loc, instring[start:loc]
2030   -
2031   - def __str__( self ):
2032   - try:
2033   - return super(CharsNotIn, self).__str__()
2034   - except:
2035   - pass
2036   -
2037   - if self.strRepr is None:
2038   - if len(self.notChars) > 4:
2039   - self.strRepr = "!W:(%s...)" % self.notChars[:4]
2040   - else:
2041   - self.strRepr = "!W:(%s)" % self.notChars
2042   -
2043   - return self.strRepr
2044   -
2045   -class White(Token):
2046   - """Special matching class for matching whitespace. Normally, whitespace is ignored
2047   - by pyparsing grammars. This class is included when some whitespace structures
2048   - are significant. Define with a string containing the whitespace characters to be
2049   - matched; default is C{" \\t\\r\\n"}. Also takes optional C{min}, C{max}, and C{exact} arguments,
2050   - as defined for the C{L{Word}} class."""
2051   - whiteStrs = {
2052   - " " : "<SPC>",
2053   - "\t": "<TAB>",
2054   - "\n": "<LF>",
2055   - "\r": "<CR>",
2056   - "\f": "<FF>",
2057   - }
2058   - def __init__(self, ws=" \t\r\n", min=1, max=0, exact=0):
2059   - super(White,self).__init__()
2060   - self.matchWhite = ws
2061   - self.setWhitespaceChars( "".join(c for c in self.whiteChars if c not in self.matchWhite) )
2062   - #~ self.leaveWhitespace()
2063   - self.name = ("".join(White.whiteStrs[c] for c in self.matchWhite))
2064   - self.mayReturnEmpty = True
2065   - self.errmsg = "Expected " + self.name
2066   -
2067   - self.minLen = min
2068   -
2069   - if max > 0:
2070   - self.maxLen = max
2071   - else:
2072   - self.maxLen = _MAX_INT
2073   -
2074   - if exact > 0:
2075   - self.maxLen = exact
2076   - self.minLen = exact
2077   -
2078   - def parseImpl( self, instring, loc, doActions=True ):
2079   - if not(instring[ loc ] in self.matchWhite):
2080   - raise ParseException(instring, loc, self.errmsg, self)
2081   - start = loc
2082   - loc += 1
2083   - maxloc = start + self.maxLen
2084   - maxloc = min( maxloc, len(instring) )
2085   - while loc < maxloc and instring[loc] in self.matchWhite:
2086   - loc += 1
2087   -
2088   - if loc - start < self.minLen:
2089   - raise ParseException(instring, loc, self.errmsg, self)
2090   -
2091   - return loc, instring[start:loc]
2092   -
2093   -
2094   -class _PositionToken(Token):
2095   - def __init__( self ):
2096   - super(_PositionToken,self).__init__()
2097   - self.name=self.__class__.__name__
2098   - self.mayReturnEmpty = True
2099   - self.mayIndexError = False
2100   -
2101   -class GoToColumn(_PositionToken):
2102   - """Token to advance to a specific column of input text; useful for tabular report scraping."""
2103   - def __init__( self, colno ):
2104   - super(GoToColumn,self).__init__()
2105   - self.col = colno
2106   -
2107   - def preParse( self, instring, loc ):
2108   - if col(loc,instring) != self.col:
2109   - instrlen = len(instring)
2110   - if self.ignoreExprs:
2111   - loc = self._skipIgnorables( instring, loc )
2112   - while loc < instrlen and instring[loc].isspace() and col( loc, instring ) != self.col :
2113   - loc += 1
2114   - return loc
2115   -
2116   - def parseImpl( self, instring, loc, doActions=True ):
2117   - thiscol = col( loc, instring )
2118   - if thiscol > self.col:
2119   - raise ParseException( instring, loc, "Text not in expected column", self )
2120   - newloc = loc + self.col - thiscol
2121   - ret = instring[ loc: newloc ]
2122   - return newloc, ret
2123   -
2124   -class LineStart(_PositionToken):
2125   - """Matches if current position is at the beginning of a line within the parse string"""
2126   - def __init__( self ):
2127   - super(LineStart,self).__init__()
2128   - self.setWhitespaceChars( ParserElement.DEFAULT_WHITE_CHARS.replace("\n","") )
2129   - self.errmsg = "Expected start of line"
2130   -
2131   - def preParse( self, instring, loc ):
2132   - preloc = super(LineStart,self).preParse(instring,loc)
2133   - if instring[preloc] == "\n":
2134   - loc += 1
2135   - return loc
2136   -
2137   - def parseImpl( self, instring, loc, doActions=True ):
2138   - if not( loc==0 or
2139   - (loc == self.preParse( instring, 0 )) or
2140   - (instring[loc-1] == "\n") ): #col(loc, instring) != 1:
2141   - raise ParseException(instring, loc, self.errmsg, self)
2142   - return loc, []
2143   -
2144   -class LineEnd(_PositionToken):
2145   - """Matches if current position is at the end of a line within the parse string"""
2146   - def __init__( self ):
2147   - super(LineEnd,self).__init__()
2148   - self.setWhitespaceChars( ParserElement.DEFAULT_WHITE_CHARS.replace("\n","") )
2149   - self.errmsg = "Expected end of line"
2150   -
2151   - def parseImpl( self, instring, loc, doActions=True ):
2152   - if loc<len(instring):
2153   - if instring[loc] == "\n":
2154   - return loc+1, "\n"
2155   - else:
2156   - raise ParseException(instring, loc, self.errmsg, self)
2157   - elif loc == len(instring):
2158   - return loc+1, []
2159   - else:
2160   - raise ParseException(instring, loc, self.errmsg, self)
2161   -
2162   -class StringStart(_PositionToken):
2163   - """Matches if current position is at the beginning of the parse string"""
2164   - def __init__( self ):
2165   - super(StringStart,self).__init__()
2166   - self.errmsg = "Expected start of text"
2167   -
2168   - def parseImpl( self, instring, loc, doActions=True ):
2169   - if loc != 0:
2170   - # see if entire string up to here is just whitespace and ignoreables
2171   - if loc != self.preParse( instring, 0 ):
2172   - raise ParseException(instring, loc, self.errmsg, self)
2173   - return loc, []
2174   -
2175   -class StringEnd(_PositionToken):
2176   - """Matches if current position is at the end of the parse string"""
2177   - def __init__( self ):
2178   - super(StringEnd,self).__init__()
2179   - self.errmsg = "Expected end of text"
2180   -
2181   - def parseImpl( self, instring, loc, doActions=True ):
2182   - if loc < len(instring):
2183   - raise ParseException(instring, loc, self.errmsg, self)
2184   - elif loc == len(instring):
2185   - return loc+1, []
2186   - elif loc > len(instring):
2187   - return loc, []
2188   - else:
2189   - raise ParseException(instring, loc, self.errmsg, self)
2190   -
2191   -class WordStart(_PositionToken):
2192   - """Matches if the current position is at the beginning of a Word, and
2193   - is not preceded by any character in a given set of C{wordChars}
2194   - (default=C{printables}). To emulate the C{\b} behavior of regular expressions,
2195   - use C{WordStart(alphanums)}. C{WordStart} will also match at the beginning of
2196   - the string being parsed, or at the beginning of a line.
2197   - """
2198   - def __init__(self, wordChars = printables):
2199   - super(WordStart,self).__init__()
2200   - self.wordChars = set(wordChars)
2201   - self.errmsg = "Not at the start of a word"
2202   -
2203   - def parseImpl(self, instring, loc, doActions=True ):
2204   - if loc != 0:
2205   - if (instring[loc-1] in self.wordChars or
2206   - instring[loc] not in self.wordChars):
2207   - raise ParseException(instring, loc, self.errmsg, self)
2208   - return loc, []
2209   -
2210   -class WordEnd(_PositionToken):
2211   - """Matches if the current position is at the end of a Word, and
2212   - is not followed by any character in a given set of C{wordChars}
2213   - (default=C{printables}). To emulate the C{\b} behavior of regular expressions,
2214   - use C{WordEnd(alphanums)}. C{WordEnd} will also match at the end of
2215   - the string being parsed, or at the end of a line.
2216   - """
2217   - def __init__(self, wordChars = printables):
2218   - super(WordEnd,self).__init__()
2219   - self.wordChars = set(wordChars)
2220   - self.skipWhitespace = False
2221   - self.errmsg = "Not at the end of a word"
2222   -
2223   - def parseImpl(self, instring, loc, doActions=True ):
2224   - instrlen = len(instring)
2225   - if instrlen>0 and loc<instrlen:
2226   - if (instring[loc] in self.wordChars or
2227   - instring[loc-1] not in self.wordChars):
2228   - raise ParseException(instring, loc, self.errmsg, self)
2229   - return loc, []
2230   -
2231   -
2232   -class ParseExpression(ParserElement):
2233   - """Abstract subclass of ParserElement, for combining and post-processing parsed tokens."""
2234   - def __init__( self, exprs, savelist = False ):
2235   - super(ParseExpression,self).__init__(savelist)
2236   - if isinstance( exprs, _generatorType ):
2237   - exprs = list(exprs)
2238   -
2239   - if isinstance( exprs, basestring ):
2240   - self.exprs = [ Literal( exprs ) ]
2241   - elif isinstance( exprs, collections.Sequence ):
2242   - # if sequence of strings provided, wrap with Literal
2243   - if all(isinstance(expr, basestring) for expr in exprs):
2244   - exprs = map(Literal, exprs)
2245   - self.exprs = list(exprs)
2246   - else:
2247   - try:
2248   - self.exprs = list( exprs )
2249   - except TypeError:
2250   - self.exprs = [ exprs ]
2251   - self.callPreparse = False
2252   -
2253   - def __getitem__( self, i ):
2254   - return self.exprs[i]
2255   -
2256   - def append( self, other ):
2257   - self.exprs.append( other )
2258   - self.strRepr = None
2259   - return self
2260   -
2261   - def leaveWhitespace( self ):
2262   - """Extends C{leaveWhitespace} defined in base class, and also invokes C{leaveWhitespace} on
2263   - all contained expressions."""
2264   - self.skipWhitespace = False
2265   - self.exprs = [ e.copy() for e in self.exprs ]
2266   - for e in self.exprs:
2267   - e.leaveWhitespace()
2268   - return self
2269   -
2270   - def ignore( self, other ):
2271   - if isinstance( other, Suppress ):
2272   - if other not in self.ignoreExprs:
2273   - super( ParseExpression, self).ignore( other )
2274   - for e in self.exprs:
2275   - e.ignore( self.ignoreExprs[-1] )
2276   - else:
2277   - super( ParseExpression, self).ignore( other )
2278   - for e in self.exprs:
2279   - e.ignore( self.ignoreExprs[-1] )
2280   - return self
2281   -
2282   - def __str__( self ):
2283   - try:
2284   - return super(ParseExpression,self).__str__()
2285   - except:
2286   - pass
2287   -
2288   - if self.strRepr is None:
2289   - self.strRepr = "%s:(%s)" % ( self.__class__.__name__, _ustr(self.exprs) )
2290   - return self.strRepr
2291   -
2292   - def streamline( self ):
2293   - super(ParseExpression,self).streamline()
2294   -
2295   - for e in self.exprs:
2296   - e.streamline()
2297   -
2298   - # collapse nested And's of the form And( And( And( a,b), c), d) to And( a,b,c,d )
2299   - # but only if there are no parse actions or resultsNames on the nested And's
2300   - # (likewise for Or's and MatchFirst's)
2301   - if ( len(self.exprs) == 2 ):
2302   - other = self.exprs[0]
2303   - if ( isinstance( other, self.__class__ ) and
2304   - not(other.parseAction) and
2305   - other.resultsName is None and
2306   - not other.debug ):
2307   - self.exprs = other.exprs[:] + [ self.exprs[1] ]
2308   - self.strRepr = None
2309   - self.mayReturnEmpty |= other.mayReturnEmpty
2310   - self.mayIndexError |= other.mayIndexError
2311   -
2312   - other = self.exprs[-1]
2313   - if ( isinstance( other, self.__class__ ) and
2314   - not(other.parseAction) and
2315   - other.resultsName is None and
2316   - not other.debug ):
2317   - self.exprs = self.exprs[:-1] + other.exprs[:]
2318   - self.strRepr = None
2319   - self.mayReturnEmpty |= other.mayReturnEmpty
2320   - self.mayIndexError |= other.mayIndexError
2321   -
2322   - return self
2323   -
2324   - def setResultsName( self, name, listAllMatches=False ):
2325   - ret = super(ParseExpression,self).setResultsName(name,listAllMatches)
2326   - return ret
2327   -
2328   - def validate( self, validateTrace=[] ):
2329   - tmp = validateTrace[:]+[self]
2330   - for e in self.exprs:
2331   - e.validate(tmp)
2332   - self.checkRecursion( [] )
2333   -
2334   - def copy(self):
2335   - ret = super(ParseExpression,self).copy()
2336   - ret.exprs = [e.copy() for e in self.exprs]
2337   - return ret
2338   -
2339   -class And(ParseExpression):
2340   - """Requires all given C{ParseExpression}s to be found in the given order.
2341   - Expressions may be separated by whitespace.
2342   - May be constructed using the C{'+'} operator.
2343   - """
2344   -
2345   - class _ErrorStop(Empty):
2346   - def __init__(self, *args, **kwargs):
2347   - super(And._ErrorStop,self).__init__(*args, **kwargs)
2348   - self.name = '-'
2349   - self.leaveWhitespace()
2350   -
2351   - def __init__( self, exprs, savelist = True ):
2352   - super(And,self).__init__(exprs, savelist)
2353   - self.mayReturnEmpty = all(e.mayReturnEmpty for e in self.exprs)
2354   - self.setWhitespaceChars( self.exprs[0].whiteChars )
2355   - self.skipWhitespace = self.exprs[0].skipWhitespace
2356   - self.callPreparse = True
2357   -
2358   - def parseImpl( self, instring, loc, doActions=True ):
2359   - # pass False as last arg to _parse for first element, since we already
2360   - # pre-parsed the string as part of our And pre-parsing
2361   - loc, resultlist = self.exprs[0]._parse( instring, loc, doActions, callPreParse=False )
2362   - errorStop = False
2363   - for e in self.exprs[1:]:
2364   - if isinstance(e, And._ErrorStop):
2365   - errorStop = True
2366   - continue
2367   - if errorStop:
2368   - try:
2369   - loc, exprtokens = e._parse( instring, loc, doActions )
2370   - except ParseSyntaxException:
2371   - raise
2372   - except ParseBaseException as pe:
2373   - pe.__traceback__ = None
2374   - raise ParseSyntaxException(pe)
2375   - except IndexError:
2376   - raise ParseSyntaxException( ParseException(instring, len(instring), self.errmsg, self) )
2377   - else:
2378   - loc, exprtokens = e._parse( instring, loc, doActions )
2379   - if exprtokens or exprtokens.haskeys():
2380   - resultlist += exprtokens
2381   - return loc, resultlist
2382   -
2383   - def __iadd__(self, other ):
2384   - if isinstance( other, basestring ):
2385   - other = Literal( other )
2386   - return self.append( other ) #And( [ self, other ] )
2387   -
2388   - def checkRecursion( self, parseElementList ):
2389   - subRecCheckList = parseElementList[:] + [ self ]
2390   - for e in self.exprs:
2391   - e.checkRecursion( subRecCheckList )
2392   - if not e.mayReturnEmpty:
2393   - break
2394   -
2395   - def __str__( self ):
2396   - if hasattr(self,"name"):
2397   - return self.name
2398   -
2399   - if self.strRepr is None:
2400   - self.strRepr = "{" + " ".join(_ustr(e) for e in self.exprs) + "}"
2401   -
2402   - return self.strRepr
2403   -
2404   -
2405   -class Or(ParseExpression):
2406   - """Requires that at least one C{ParseExpression} is found.
2407   - If two expressions match, the expression that matches the longest string will be used.
2408   - May be constructed using the C{'^'} operator.
2409   - """
2410   - def __init__( self, exprs, savelist = False ):
2411   - super(Or,self).__init__(exprs, savelist)
2412   - if self.exprs:
2413   - self.mayReturnEmpty = any(e.mayReturnEmpty for e in self.exprs)
2414   - else:
2415   - self.mayReturnEmpty = True
2416   -
2417   - def parseImpl( self, instring, loc, doActions=True ):
2418   - maxExcLoc = -1
2419   - maxMatchLoc = -1
2420   - maxException = None
2421   - for e in self.exprs:
2422   - try:
2423   - loc2 = e.tryParse( instring, loc )
2424   - except ParseException as err:
2425   - err.__traceback__ = None
2426   - if err.loc > maxExcLoc:
2427   - maxException = err
2428   - maxExcLoc = err.loc
2429   - except IndexError:
2430   - if len(instring) > maxExcLoc:
2431   - maxException = ParseException(instring,len(instring),e.errmsg,self)
2432   - maxExcLoc = len(instring)
2433   - else:
2434   - if loc2 > maxMatchLoc:
2435   - maxMatchLoc = loc2
2436   - maxMatchExp = e
2437   -
2438   - if maxMatchLoc < 0:
2439   - if maxException is not None:
2440   - raise maxException
2441   - else:
2442   - raise ParseException(instring, loc, "no defined alternatives to match", self)
2443   -
2444   - return maxMatchExp._parse( instring, loc, doActions )
2445   -
2446   - def __ixor__(self, other ):
2447   - if isinstance( other, basestring ):
2448   - other = ParserElement.literalStringClass( other )
2449   - return self.append( other ) #Or( [ self, other ] )
2450   -
2451   - def __str__( self ):
2452   - if hasattr(self,"name"):
2453   - return self.name
2454   -
2455   - if self.strRepr is None:
2456   - self.strRepr = "{" + " ^ ".join(_ustr(e) for e in self.exprs) + "}"
2457   -
2458   - return self.strRepr
2459   -
2460   - def checkRecursion( self, parseElementList ):
2461   - subRecCheckList = parseElementList[:] + [ self ]
2462   - for e in self.exprs:
2463   - e.checkRecursion( subRecCheckList )
2464   -
2465   -
2466   -class MatchFirst(ParseExpression):
2467   - """Requires that at least one C{ParseExpression} is found.
2468   - If two expressions match, the first one listed is the one that will match.
2469   - May be constructed using the C{'|'} operator.
2470   - """
2471   - def __init__( self, exprs, savelist = False ):
2472   - super(MatchFirst,self).__init__(exprs, savelist)
2473   - if self.exprs:
2474   - self.mayReturnEmpty = any(e.mayReturnEmpty for e in self.exprs)
2475   - else:
2476   - self.mayReturnEmpty = True
2477   -
2478   - def parseImpl( self, instring, loc, doActions=True ):
2479   - maxExcLoc = -1
2480   - maxException = None
2481   - for e in self.exprs:
2482   - try:
2483   - ret = e._parse( instring, loc, doActions )
2484   - return ret
2485   - except ParseException as err:
2486   - if err.loc > maxExcLoc:
2487   - maxException = err
2488   - maxExcLoc = err.loc
2489   - except IndexError:
2490   - if len(instring) > maxExcLoc:
2491   - maxException = ParseException(instring,len(instring),e.errmsg,self)
2492   - maxExcLoc = len(instring)
2493   -
2494   - # only got here if no expression matched, raise exception for match that made it the furthest
2495   - else:
2496   - if maxException is not None:
2497   - raise maxException
2498   - else:
2499   - raise ParseException(instring, loc, "no defined alternatives to match", self)
2500   -
2501   - def __ior__(self, other ):
2502   - if isinstance( other, basestring ):
2503   - other = ParserElement.literalStringClass( other )
2504   - return self.append( other ) #MatchFirst( [ self, other ] )
2505   -
2506   - def __str__( self ):
2507   - if hasattr(self,"name"):
2508   - return self.name
2509   -
2510   - if self.strRepr is None:
2511   - self.strRepr = "{" + " | ".join(_ustr(e) for e in self.exprs) + "}"
2512   -
2513   - return self.strRepr
2514   -
2515   - def checkRecursion( self, parseElementList ):
2516   - subRecCheckList = parseElementList[:] + [ self ]
2517   - for e in self.exprs:
2518   - e.checkRecursion( subRecCheckList )
2519   -
2520   -
2521   -class Each(ParseExpression):
2522   - """Requires all given C{ParseExpression}s to be found, but in any order.
2523   - Expressions may be separated by whitespace.
2524   - May be constructed using the C{'&'} operator.
2525   - """
2526   - def __init__( self, exprs, savelist = True ):
2527   - super(Each,self).__init__(exprs, savelist)
2528   - self.mayReturnEmpty = all(e.mayReturnEmpty for e in self.exprs)
2529   - self.skipWhitespace = True
2530   - self.initExprGroups = True
2531   -
2532   - def parseImpl( self, instring, loc, doActions=True ):
2533   - if self.initExprGroups:
2534   - opt1 = [ e.expr for e in self.exprs if isinstance(e,Optional) ]
2535   - opt2 = [ e for e in self.exprs if e.mayReturnEmpty and e not in opt1 ]
2536   - self.optionals = opt1 + opt2
2537   - self.multioptionals = [ e.expr for e in self.exprs if isinstance(e,ZeroOrMore) ]
2538   - self.multirequired = [ e.expr for e in self.exprs if isinstance(e,OneOrMore) ]
2539   - self.required = [ e for e in self.exprs if not isinstance(e,(Optional,ZeroOrMore,OneOrMore)) ]
2540   - self.required += self.multirequired
2541   - self.initExprGroups = False
2542   - tmpLoc = loc
2543   - tmpReqd = self.required[:]
2544   - tmpOpt = self.optionals[:]
2545   - matchOrder = []
2546   -
2547   - keepMatching = True
2548   - while keepMatching:
2549   - tmpExprs = tmpReqd + tmpOpt + self.multioptionals + self.multirequired
2550   - failed = []
2551   - for e in tmpExprs:
2552   - try:
2553   - tmpLoc = e.tryParse( instring, tmpLoc )
2554   - except ParseException:
2555   - failed.append(e)
2556   - else:
2557   - matchOrder.append(e)
2558   - if e in tmpReqd:
2559   - tmpReqd.remove(e)
2560   - elif e in tmpOpt:
2561   - tmpOpt.remove(e)
2562   - if len(failed) == len(tmpExprs):
2563   - keepMatching = False
2564   -
2565   - if tmpReqd:
2566   - missing = ", ".join(_ustr(e) for e in tmpReqd)
2567   - raise ParseException(instring,loc,"Missing one or more required elements (%s)" % missing )
2568   -
2569   - # add any unmatched Optionals, in case they have default values defined
2570   - matchOrder += [e for e in self.exprs if isinstance(e,Optional) and e.expr in tmpOpt]
2571   -
2572   - resultlist = []
2573   - for e in matchOrder:
2574   - loc,results = e._parse(instring,loc,doActions)
2575   - resultlist.append(results)
2576   -
2577   - finalResults = ParseResults([])
2578   - for r in resultlist:
2579   - dups = {}
2580   - for k in r.keys():
2581   - if k in finalResults:
2582   - tmp = ParseResults(finalResults[k])
2583   - tmp += ParseResults(r[k])
2584   - dups[k] = tmp
2585   - finalResults += ParseResults(r)
2586   - for k,v in dups.items():
2587   - finalResults[k] = v
2588   - return loc, finalResults
2589   -
2590   - def __str__( self ):
2591   - if hasattr(self,"name"):
2592   - return self.name
2593   -
2594   - if self.strRepr is None:
2595   - self.strRepr = "{" + " & ".join(_ustr(e) for e in self.exprs) + "}"
2596   -
2597   - return self.strRepr
2598   -
2599   - def checkRecursion( self, parseElementList ):
2600   - subRecCheckList = parseElementList[:] + [ self ]
2601   - for e in self.exprs:
2602   - e.checkRecursion( subRecCheckList )
2603   -
2604   -
2605   -class ParseElementEnhance(ParserElement):
2606   - """Abstract subclass of C{ParserElement}, for combining and post-processing parsed tokens."""
2607   - def __init__( self, expr, savelist=False ):
2608   - super(ParseElementEnhance,self).__init__(savelist)
2609   - if isinstance( expr, basestring ):
2610   - expr = Literal(expr)
2611   - self.expr = expr
2612   - self.strRepr = None
2613   - if expr is not None:
2614   - self.mayIndexError = expr.mayIndexError
2615   - self.mayReturnEmpty = expr.mayReturnEmpty
2616   - self.setWhitespaceChars( expr.whiteChars )
2617   - self.skipWhitespace = expr.skipWhitespace
2618   - self.saveAsList = expr.saveAsList
2619   - self.callPreparse = expr.callPreparse
2620   - self.ignoreExprs.extend(expr.ignoreExprs)
2621   -
2622   - def parseImpl( self, instring, loc, doActions=True ):
2623   - if self.expr is not None:
2624   - return self.expr._parse( instring, loc, doActions, callPreParse=False )
2625   - else:
2626   - raise ParseException("",loc,self.errmsg,self)
2627   -
2628   - def leaveWhitespace( self ):
2629   - self.skipWhitespace = False
2630   - self.expr = self.expr.copy()
2631   - if self.expr is not None:
2632   - self.expr.leaveWhitespace()
2633   - return self
2634   -
2635   - def ignore( self, other ):
2636   - if isinstance( other, Suppress ):
2637   - if other not in self.ignoreExprs:
2638   - super( ParseElementEnhance, self).ignore( other )
2639   - if self.expr is not None:
2640   - self.expr.ignore( self.ignoreExprs[-1] )
2641   - else:
2642   - super( ParseElementEnhance, self).ignore( other )
2643   - if self.expr is not None:
2644   - self.expr.ignore( self.ignoreExprs[-1] )
2645   - return self
2646   -
2647   - def streamline( self ):
2648   - super(ParseElementEnhance,self).streamline()
2649   - if self.expr is not None:
2650   - self.expr.streamline()
2651   - return self
2652   -
2653   - def checkRecursion( self, parseElementList ):
2654   - if self in parseElementList:
2655   - raise RecursiveGrammarException( parseElementList+[self] )
2656   - subRecCheckList = parseElementList[:] + [ self ]
2657   - if self.expr is not None:
2658   - self.expr.checkRecursion( subRecCheckList )
2659   -
2660   - def validate( self, validateTrace=[] ):
2661   - tmp = validateTrace[:]+[self]
2662   - if self.expr is not None:
2663   - self.expr.validate(tmp)
2664   - self.checkRecursion( [] )
2665   -
2666   - def __str__( self ):
2667   - try:
2668   - return super(ParseElementEnhance,self).__str__()
2669   - except:
2670   - pass
2671   -
2672   - if self.strRepr is None and self.expr is not None:
2673   - self.strRepr = "%s:(%s)" % ( self.__class__.__name__, _ustr(self.expr) )
2674   - return self.strRepr
2675   -
2676   -
2677   -class FollowedBy(ParseElementEnhance):
2678   - """Lookahead matching of the given parse expression. C{FollowedBy}
2679   - does *not* advance the parsing position within the input string, it only
2680   - verifies that the specified parse expression matches at the current
2681   - position. C{FollowedBy} always returns a null token list."""
2682   - def __init__( self, expr ):
2683   - super(FollowedBy,self).__init__(expr)
2684   - self.mayReturnEmpty = True
2685   -
2686   - def parseImpl( self, instring, loc, doActions=True ):
2687   - self.expr.tryParse( instring, loc )
2688   - return loc, []
2689   -
2690   -
2691   -class NotAny(ParseElementEnhance):
2692   - """Lookahead to disallow matching with the given parse expression. C{NotAny}
2693   - does *not* advance the parsing position within the input string, it only
2694   - verifies that the specified parse expression does *not* match at the current
2695   - position. Also, C{NotAny} does *not* skip over leading whitespace. C{NotAny}
2696   - always returns a null token list. May be constructed using the '~' operator."""
2697   - def __init__( self, expr ):
2698   - super(NotAny,self).__init__(expr)
2699   - #~ self.leaveWhitespace()
2700   - self.skipWhitespace = False # do NOT use self.leaveWhitespace(), don't want to propagate to exprs
2701   - self.mayReturnEmpty = True
2702   - self.errmsg = "Found unwanted token, "+_ustr(self.expr)
2703   -
2704   - def parseImpl( self, instring, loc, doActions=True ):
2705   - try:
2706   - self.expr.tryParse( instring, loc )
2707   - except (ParseException,IndexError):
2708   - pass
2709   - else:
2710   - raise ParseException(instring, loc, self.errmsg, self)
2711   - return loc, []
2712   -
2713   - def __str__( self ):
2714   - if hasattr(self,"name"):
2715   - return self.name
2716   -
2717   - if self.strRepr is None:
2718   - self.strRepr = "~{" + _ustr(self.expr) + "}"
2719   -
2720   - return self.strRepr
2721   -
2722   -
2723   -class ZeroOrMore(ParseElementEnhance):
2724   - """Optional repetition of zero or more of the given expression."""
2725   - def __init__( self, expr ):
2726   - super(ZeroOrMore,self).__init__(expr)
2727   - self.mayReturnEmpty = True
2728   -
2729   - def parseImpl( self, instring, loc, doActions=True ):
2730   - tokens = []
2731   - try:
2732   - loc, tokens = self.expr._parse( instring, loc, doActions, callPreParse=False )
2733   - hasIgnoreExprs = ( len(self.ignoreExprs) > 0 )
2734   - while 1:
2735   - if hasIgnoreExprs:
2736   - preloc = self._skipIgnorables( instring, loc )
2737   - else:
2738   - preloc = loc
2739   - loc, tmptokens = self.expr._parse( instring, preloc, doActions )
2740   - if tmptokens or tmptokens.haskeys():
2741   - tokens += tmptokens
2742   - except (ParseException,IndexError):
2743   - pass
2744   -
2745   - return loc, tokens
2746   -
2747   - def __str__( self ):
2748   - if hasattr(self,"name"):
2749   - return self.name
2750   -
2751   - if self.strRepr is None:
2752   - self.strRepr = "[" + _ustr(self.expr) + "]..."
2753   -
2754   - return self.strRepr
2755   -
2756   - def setResultsName( self, name, listAllMatches=False ):
2757   - ret = super(ZeroOrMore,self).setResultsName(name,listAllMatches)
2758   - ret.saveAsList = True
2759   - return ret
2760   -
2761   -
2762   -class OneOrMore(ParseElementEnhance):
2763   - """Repetition of one or more of the given expression."""
2764   - def parseImpl( self, instring, loc, doActions=True ):
2765   - # must be at least one
2766   - loc, tokens = self.expr._parse( instring, loc, doActions, callPreParse=False )
2767   - try:
2768   - hasIgnoreExprs = ( len(self.ignoreExprs) > 0 )
2769   - while 1:
2770   - if hasIgnoreExprs:
2771   - preloc = self._skipIgnorables( instring, loc )
2772   - else:
2773   - preloc = loc
2774   - loc, tmptokens = self.expr._parse( instring, preloc, doActions )
2775   - if tmptokens or tmptokens.haskeys():
2776   - tokens += tmptokens
2777   - except (ParseException,IndexError):
2778   - pass
2779   -
2780   - return loc, tokens
2781   -
2782   - def __str__( self ):
2783   - if hasattr(self,"name"):
2784   - return self.name
2785   -
2786   - if self.strRepr is None:
2787   - self.strRepr = "{" + _ustr(self.expr) + "}..."
2788   -
2789   - return self.strRepr
2790   -
2791   - def setResultsName( self, name, listAllMatches=False ):
2792   - ret = super(OneOrMore,self).setResultsName(name,listAllMatches)
2793   - ret.saveAsList = True
2794   - return ret
2795   -
2796   -class _NullToken(object):
2797   - def __bool__(self):
2798   - return False
2799   - __nonzero__ = __bool__
2800   - def __str__(self):
2801   - return ""
2802   -
2803   -_optionalNotMatched = _NullToken()
2804   -class Optional(ParseElementEnhance):
2805   - """Optional matching of the given expression.
2806   - A default return string can also be specified, if the optional expression
2807   - is not found.
2808   - """
2809   - def __init__( self, expr, default=_optionalNotMatched ):
2810   - super(Optional,self).__init__( expr, savelist=False )
2811   - self.defaultValue = default
2812   - self.mayReturnEmpty = True
2813   -
2814   - def parseImpl( self, instring, loc, doActions=True ):
2815   - try:
2816   - loc, tokens = self.expr._parse( instring, loc, doActions, callPreParse=False )
2817   - except (ParseException,IndexError):
2818   - if self.defaultValue is not _optionalNotMatched:
2819   - if self.expr.resultsName:
2820   - tokens = ParseResults([ self.defaultValue ])
2821   - tokens[self.expr.resultsName] = self.defaultValue
2822   - else:
2823   - tokens = [ self.defaultValue ]
2824   - else:
2825   - tokens = []
2826   - return loc, tokens
2827   -
2828   - def __str__( self ):
2829   - if hasattr(self,"name"):
2830   - return self.name
2831   -
2832   - if self.strRepr is None:
2833   - self.strRepr = "[" + _ustr(self.expr) + "]"
2834   -
2835   - return self.strRepr
2836   -
2837   -
2838   -class SkipTo(ParseElementEnhance):
2839   - """Token for skipping over all undefined text until the matched expression is found.
2840   - If C{include} is set to true, the matched expression is also parsed (the skipped text
2841   - and matched expression are returned as a 2-element list). The C{ignore}
2842   - argument is used to define grammars (typically quoted strings and comments) that
2843   - might contain false matches.
2844   - """
2845   - def __init__( self, other, include=False, ignore=None, failOn=None ):
2846   - super( SkipTo, self ).__init__( other )
2847   - self.ignoreExpr = ignore
2848   - self.mayReturnEmpty = True
2849   - self.mayIndexError = False
2850   - self.includeMatch = include
2851   - self.asList = False
2852   - if failOn is not None and isinstance(failOn, basestring):
2853   - self.failOn = Literal(failOn)
2854   - else:
2855   - self.failOn = failOn
2856   - self.errmsg = "No match found for "+_ustr(self.expr)
2857   -
2858   - def parseImpl( self, instring, loc, doActions=True ):
2859   - startLoc = loc
2860   - instrlen = len(instring)
2861   - expr = self.expr
2862   - failParse = False
2863   - while loc <= instrlen:
2864   - try:
2865   - if self.failOn:
2866   - try:
2867   - self.failOn.tryParse(instring, loc)
2868   - except ParseBaseException:
2869   - pass
2870   - else:
2871   - failParse = True
2872   - raise ParseException(instring, loc, "Found expression " + str(self.failOn))
2873   - failParse = False
2874   - if self.ignoreExpr is not None:
2875   - while 1:
2876   - try:
2877   - loc = self.ignoreExpr.tryParse(instring,loc)
2878   - # print("found ignoreExpr, advance to", loc)
2879   - except ParseBaseException:
2880   - break
2881   - expr._parse( instring, loc, doActions=False, callPreParse=False )
2882   - skipText = instring[startLoc:loc]
2883   - if self.includeMatch:
2884   - loc,mat = expr._parse(instring,loc,doActions,callPreParse=False)
2885   - if mat:
2886   - skipRes = ParseResults( skipText )
2887   - skipRes += mat
2888   - return loc, [ skipRes ]
2889   - else:
2890   - return loc, [ skipText ]
2891   - else:
2892   - return loc, [ skipText ]
2893   - except (ParseException,IndexError):
2894   - if failParse:
2895   - raise
2896   - else:
2897   - loc += 1
2898   - raise ParseException(instring, loc, self.errmsg, self)
2899   -
2900   -class Forward(ParseElementEnhance):
2901   - """Forward declaration of an expression to be defined later -
2902   - used for recursive grammars, such as algebraic infix notation.
2903   - When the expression is known, it is assigned to the C{Forward} variable using the '<<' operator.
2904   -
2905   - Note: take care when assigning to C{Forward} not to overlook precedence of operators.
2906   - Specifically, '|' has a lower precedence than '<<', so that::
2907   - fwdExpr << a | b | c
2908   - will actually be evaluated as::
2909   - (fwdExpr << a) | b | c
2910   - thereby leaving b and c out as parseable alternatives. It is recommended that you
2911   - explicitly group the values inserted into the C{Forward}::
2912   - fwdExpr << (a | b | c)
2913   - Converting to use the '<<=' operator instead will avoid this problem.
2914   - """
2915   - def __init__( self, other=None ):
2916   - super(Forward,self).__init__( other, savelist=False )
2917   -
2918   - def __lshift__( self, other ):
2919   - if isinstance( other, basestring ):
2920   - other = ParserElement.literalStringClass(other)
2921   - self.expr = other
2922   - self.mayReturnEmpty = other.mayReturnEmpty
2923   - self.strRepr = None
2924   - self.mayIndexError = self.expr.mayIndexError
2925   - self.mayReturnEmpty = self.expr.mayReturnEmpty
2926   - self.setWhitespaceChars( self.expr.whiteChars )
2927   - self.skipWhitespace = self.expr.skipWhitespace
2928   - self.saveAsList = self.expr.saveAsList
2929   - self.ignoreExprs.extend(self.expr.ignoreExprs)
2930   - return self
2931   -
2932   - def __ilshift__(self, other):
2933   - return self << other
2934   -
2935   - def leaveWhitespace( self ):
2936   - self.skipWhitespace = False
2937   - return self
2938   -
2939   - def streamline( self ):
2940   - if not self.streamlined:
2941   - self.streamlined = True
2942   - if self.expr is not None:
2943   - self.expr.streamline()
2944   - return self
2945   -
2946   - def validate( self, validateTrace=[] ):
2947   - if self not in validateTrace:
2948   - tmp = validateTrace[:]+[self]
2949   - if self.expr is not None:
2950   - self.expr.validate(tmp)
2951   - self.checkRecursion([])
2952   -
2953   - def __str__( self ):
2954   - if hasattr(self,"name"):
2955   - return self.name
2956   -
2957   - self._revertClass = self.__class__
2958   - self.__class__ = _ForwardNoRecurse
2959   - try:
2960   - if self.expr is not None:
2961   - retString = _ustr(self.expr)
2962   - else:
2963   - retString = "None"
2964   - finally:
2965   - self.__class__ = self._revertClass
2966   - return self.__class__.__name__ + ": " + retString
2967   -
2968   - def copy(self):
2969   - if self.expr is not None:
2970   - return super(Forward,self).copy()
2971   - else:
2972   - ret = Forward()
2973   - ret <<= self
2974   - return ret
2975   -
2976   -class _ForwardNoRecurse(Forward):
2977   - def __str__( self ):
2978   - return "..."
2979   -
2980   -class TokenConverter(ParseElementEnhance):
2981   - """Abstract subclass of C{ParseExpression}, for converting parsed results."""
2982   - def __init__( self, expr, savelist=False ):
2983   - super(TokenConverter,self).__init__( expr )#, savelist )
2984   - self.saveAsList = False
2985   -
2986   -class Upcase(TokenConverter):
2987   - """Converter to upper case all matching tokens."""
2988   - def __init__(self, *args):
2989   - super(Upcase,self).__init__(*args)
2990   - warnings.warn("Upcase class is deprecated, use upcaseTokens parse action instead",
2991   - DeprecationWarning,stacklevel=2)
2992   -
2993   - def postParse( self, instring, loc, tokenlist ):
2994   - return list(map( str.upper, tokenlist ))
2995   -
2996   -
2997   -class Combine(TokenConverter):
2998   - """Converter to concatenate all matching tokens to a single string.
2999   - By default, the matching patterns must also be contiguous in the input string;
3000   - this can be disabled by specifying C{'adjacent=False'} in the constructor.
3001   - """
3002   - def __init__( self, expr, joinString="", adjacent=True ):
3003   - super(Combine,self).__init__( expr )
3004   - # suppress whitespace-stripping in contained parse expressions, but re-enable it on the Combine itself
3005   - if adjacent:
3006   - self.leaveWhitespace()
3007   - self.adjacent = adjacent
3008   - self.skipWhitespace = True
3009   - self.joinString = joinString
3010   - self.callPreparse = True
3011   -
3012   - def ignore( self, other ):
3013   - if self.adjacent:
3014   - ParserElement.ignore(self, other)
3015   - else:
3016   - super( Combine, self).ignore( other )
3017   - return self
3018   -
3019   - def postParse( self, instring, loc, tokenlist ):
3020   - retToks = tokenlist.copy()
3021   - del retToks[:]
3022   - retToks += ParseResults([ "".join(tokenlist._asStringList(self.joinString)) ], modal=self.modalResults)
3023   -
3024   - if self.resultsName and retToks.haskeys():
3025   - return [ retToks ]
3026   - else:
3027   - return retToks
3028   -
3029   -class Group(TokenConverter):
3030   - """Converter to return the matched tokens as a list - useful for returning tokens of C{L{ZeroOrMore}} and C{L{OneOrMore}} expressions."""
3031   - def __init__( self, expr ):
3032   - super(Group,self).__init__( expr )
3033   - self.saveAsList = True
3034   -
3035   - def postParse( self, instring, loc, tokenlist ):
3036   - return [ tokenlist ]
3037   -
3038   -class Dict(TokenConverter):
3039   - """Converter to return a repetitive expression as a list, but also as a dictionary.
3040   - Each element can also be referenced using the first token in the expression as its key.
3041   - Useful for tabular report scraping when the first column can be used as a item key.
3042   - """
3043   - def __init__( self, expr ):
3044   - super(Dict,self).__init__( expr )
3045   - self.saveAsList = True
3046   -
3047   - def postParse( self, instring, loc, tokenlist ):
3048   - for i,tok in enumerate(tokenlist):
3049   - if len(tok) == 0:
3050   - continue
3051   - ikey = tok[0]
3052   - if isinstance(ikey,int):
3053   - ikey = _ustr(tok[0]).strip()
3054   - if len(tok)==1:
3055   - tokenlist[ikey] = _ParseResultsWithOffset("",i)
3056   - elif len(tok)==2 and not isinstance(tok[1],ParseResults):
3057   - tokenlist[ikey] = _ParseResultsWithOffset(tok[1],i)
3058   - else:
3059   - dictvalue = tok.copy() #ParseResults(i)
3060   - del dictvalue[0]
3061   - if len(dictvalue)!= 1 or (isinstance(dictvalue,ParseResults) and dictvalue.haskeys()):
3062   - tokenlist[ikey] = _ParseResultsWithOffset(dictvalue,i)
3063   - else:
3064   - tokenlist[ikey] = _ParseResultsWithOffset(dictvalue[0],i)
3065   -
3066   - if self.resultsName:
3067   - return [ tokenlist ]
3068   - else:
3069   - return tokenlist
3070   -
3071   -
3072   -class Suppress(TokenConverter):
3073   - """Converter for ignoring the results of a parsed expression."""
3074   - def postParse( self, instring, loc, tokenlist ):
3075   - return []
3076   -
3077   - def suppress( self ):
3078   - return self
3079   -
3080   -
3081   -class OnlyOnce(object):
3082   - """Wrapper for parse actions, to ensure they are only called once."""
3083   - def __init__(self, methodCall):
3084   - self.callable = _trim_arity(methodCall)
3085   - self.called = False
3086   - def __call__(self,s,l,t):
3087   - if not self.called:
3088   - results = self.callable(s,l,t)
3089   - self.called = True
3090   - return results
3091   - raise ParseException(s,l,"")
3092   - def reset(self):
3093   - self.called = False
3094   -
3095   -def traceParseAction(f):
3096   - """Decorator for debugging parse actions."""
3097   - f = _trim_arity(f)
3098   - def z(*paArgs):
3099   - thisFunc = f.func_name
3100   - s,l,t = paArgs[-3:]
3101   - if len(paArgs)>3:
3102   - thisFunc = paArgs[0].__class__.__name__ + '.' + thisFunc
3103   - sys.stderr.write( ">>entering %s(line: '%s', %d, %s)\n" % (thisFunc,line(l,s),l,t) )
3104   - try:
3105   - ret = f(*paArgs)
3106   - except Exception as exc:
3107   - sys.stderr.write( "<<leaving %s (exception: %s)\n" % (thisFunc,exc) )
3108   - raise
3109   - sys.stderr.write( "<<leaving %s (ret: %s)\n" % (thisFunc,ret) )
3110   - return ret
3111   - try:
3112   - z.__name__ = f.__name__
3113   - except AttributeError:
3114   - pass
3115   - return z
3116   -
3117   -#
3118   -# global helpers
3119   -#
3120   -def delimitedList( expr, delim=",", combine=False ):
3121   - """Helper to define a delimited list of expressions - the delimiter defaults to ','.
3122   - By default, the list elements and delimiters can have intervening whitespace, and
3123   - comments, but this can be overridden by passing C{combine=True} in the constructor.
3124   - If C{combine} is set to C{True}, the matching tokens are returned as a single token
3125   - string, with the delimiters included; otherwise, the matching tokens are returned
3126   - as a list of tokens, with the delimiters suppressed.
3127   - """
3128   - dlName = _ustr(expr)+" ["+_ustr(delim)+" "+_ustr(expr)+"]..."
3129   - if combine:
3130   - return Combine( expr + ZeroOrMore( delim + expr ) ).setName(dlName)
3131   - else:
3132   - return ( expr + ZeroOrMore( Suppress( delim ) + expr ) ).setName(dlName)
3133   -
3134   -def countedArray( expr, intExpr=None ):
3135   - """Helper to define a counted list of expressions.
3136   - This helper defines a pattern of the form::
3137   - integer expr expr expr...
3138   - where the leading integer tells how many expr expressions follow.
3139   - The matched tokens returns the array of expr tokens as a list - the leading count token is suppressed.
3140   - """
3141   - arrayExpr = Forward()
3142   - def countFieldParseAction(s,l,t):
3143   - n = t[0]
3144   - arrayExpr << (n and Group(And([expr]*n)) or Group(empty))
3145   - return []
3146   - if intExpr is None:
3147   - intExpr = Word(nums).setParseAction(lambda t:int(t[0]))
3148   - else:
3149   - intExpr = intExpr.copy()
3150   - intExpr.setName("arrayLen")
3151   - intExpr.addParseAction(countFieldParseAction, callDuringTry=True)
3152   - return ( intExpr + arrayExpr )
3153   -
3154   -def _flatten(L):
3155   - ret = []
3156   - for i in L:
3157   - if isinstance(i,list):
3158   - ret.extend(_flatten(i))
3159   - else:
3160   - ret.append(i)
3161   - return ret
3162   -
3163   -def matchPreviousLiteral(expr):
3164   - """Helper to define an expression that is indirectly defined from
3165   - the tokens matched in a previous expression, that is, it looks
3166   - for a 'repeat' of a previous expression. For example::
3167   - first = Word(nums)
3168   - second = matchPreviousLiteral(first)
3169   - matchExpr = first + ":" + second
3170   - will match C{"1:1"}, but not C{"1:2"}. Because this matches a
3171   - previous literal, will also match the leading C{"1:1"} in C{"1:10"}.
3172   - If this is not desired, use C{matchPreviousExpr}.
3173   - Do *not* use with packrat parsing enabled.
3174   - """
3175   - rep = Forward()
3176   - def copyTokenToRepeater(s,l,t):
3177   - if t:
3178   - if len(t) == 1:
3179   - rep << t[0]
3180   - else:
3181   - # flatten t tokens
3182   - tflat = _flatten(t.asList())
3183   - rep << And( [ Literal(tt) for tt in tflat ] )
3184   - else:
3185   - rep << Empty()
3186   - expr.addParseAction(copyTokenToRepeater, callDuringTry=True)
3187   - return rep
3188   -
3189   -def matchPreviousExpr(expr):
3190   - """Helper to define an expression that is indirectly defined from
3191   - the tokens matched in a previous expression, that is, it looks
3192   - for a 'repeat' of a previous expression. For example::
3193   - first = Word(nums)
3194   - second = matchPreviousExpr(first)
3195   - matchExpr = first + ":" + second
3196   - will match C{"1:1"}, but not C{"1:2"}. Because this matches by
3197   - expressions, will *not* match the leading C{"1:1"} in C{"1:10"};
3198   - the expressions are evaluated first, and then compared, so
3199   - C{"1"} is compared with C{"10"}.
3200   - Do *not* use with packrat parsing enabled.
3201   - """
3202   - rep = Forward()
3203   - e2 = expr.copy()
3204   - rep <<= e2
3205   - def copyTokenToRepeater(s,l,t):
3206   - matchTokens = _flatten(t.asList())
3207   - def mustMatchTheseTokens(s,l,t):
3208   - theseTokens = _flatten(t.asList())
3209   - if theseTokens != matchTokens:
3210   - raise ParseException("",0,"")
3211   - rep.setParseAction( mustMatchTheseTokens, callDuringTry=True )
3212   - expr.addParseAction(copyTokenToRepeater, callDuringTry=True)
3213   - return rep
3214   -
3215   -def _escapeRegexRangeChars(s):
3216   - #~ escape these chars: ^-]
3217   - for c in r"\^-]":
3218   - s = s.replace(c,_bslash+c)
3219   - s = s.replace("\n",r"\n")
3220   - s = s.replace("\t",r"\t")
3221   - return _ustr(s)
3222   -
3223   -def oneOf( strs, caseless=False, useRegex=True ):
3224   - """Helper to quickly define a set of alternative Literals, and makes sure to do
3225   - longest-first testing when there is a conflict, regardless of the input order,
3226   - but returns a C{L{MatchFirst}} for best performance.
3227   -
3228   - Parameters:
3229   - - strs - a string of space-delimited literals, or a list of string literals
3230   - - caseless - (default=False) - treat all literals as caseless
3231   - - useRegex - (default=True) - as an optimization, will generate a Regex
3232   - object; otherwise, will generate a C{MatchFirst} object (if C{caseless=True}, or
3233   - if creating a C{Regex} raises an exception)
3234   - """
3235   - if caseless:
3236   - isequal = ( lambda a,b: a.upper() == b.upper() )
3237   - masks = ( lambda a,b: b.upper().startswith(a.upper()) )
3238   - parseElementClass = CaselessLiteral
3239   - else:
3240   - isequal = ( lambda a,b: a == b )
3241   - masks = ( lambda a,b: b.startswith(a) )
3242   - parseElementClass = Literal
3243   -
3244   - symbols = []
3245   - if isinstance(strs,basestring):
3246   - symbols = strs.split()
3247   - elif isinstance(strs, collections.Sequence):
3248   - symbols = list(strs[:])
3249   - elif isinstance(strs, _generatorType):
3250   - symbols = list(strs)
3251   - else:
3252   - warnings.warn("Invalid argument to oneOf, expected string or list",
3253   - SyntaxWarning, stacklevel=2)
3254   - if not symbols:
3255   - return NoMatch()
3256   -
3257   - i = 0
3258   - while i < len(symbols)-1:
3259   - cur = symbols[i]
3260   - for j,other in enumerate(symbols[i+1:]):
3261   - if ( isequal(other, cur) ):
3262   - del symbols[i+j+1]
3263   - break
3264   - elif ( masks(cur, other) ):
3265   - del symbols[i+j+1]
3266   - symbols.insert(i,other)
3267   - cur = other
3268   - break
3269   - else:
3270   - i += 1
3271   -
3272   - if not caseless and useRegex:
3273   - #~ print (strs,"->", "|".join( [ _escapeRegexChars(sym) for sym in symbols] ))
3274   - try:
3275   - if len(symbols)==len("".join(symbols)):
3276   - return Regex( "[%s]" % "".join(_escapeRegexRangeChars(sym) for sym in symbols) )
3277   - else:
3278   - return Regex( "|".join(re.escape(sym) for sym in symbols) )
3279   - except:
3280   - warnings.warn("Exception creating Regex for oneOf, building MatchFirst",
3281   - SyntaxWarning, stacklevel=2)
3282   -
3283   -
3284   - # last resort, just use MatchFirst
3285   - return MatchFirst( [ parseElementClass(sym) for sym in symbols ] )
3286   -
3287   -def dictOf( key, value ):
3288   - """Helper to easily and clearly define a dictionary by specifying the respective patterns
3289   - for the key and value. Takes care of defining the C{L{Dict}}, C{L{ZeroOrMore}}, and C{L{Group}} tokens
3290   - in the proper order. The key pattern can include delimiting markers or punctuation,
3291   - as long as they are suppressed, thereby leaving the significant key text. The value
3292   - pattern can include named results, so that the C{Dict} results can include named token
3293   - fields.
3294   - """
3295   - return Dict( ZeroOrMore( Group ( key + value ) ) )
3296   -
3297   -def originalTextFor(expr, asString=True):
3298   - """Helper to return the original, untokenized text for a given expression. Useful to
3299   - restore the parsed fields of an HTML start tag into the raw tag text itself, or to
3300   - revert separate tokens with intervening whitespace back to the original matching
3301   - input text. Simpler to use than the parse action C{L{keepOriginalText}}, and does not
3302   - require the inspect module to chase up the call stack. By default, returns a
3303   - string containing the original parsed text.
3304   -
3305   - If the optional C{asString} argument is passed as C{False}, then the return value is a
3306   - C{L{ParseResults}} containing any results names that were originally matched, and a
3307   - single token containing the original matched text from the input string. So if
3308   - the expression passed to C{L{originalTextFor}} contains expressions with defined
3309   - results names, you must set C{asString} to C{False} if you want to preserve those
3310   - results name values."""
3311   - locMarker = Empty().setParseAction(lambda s,loc,t: loc)
3312   - endlocMarker = locMarker.copy()
3313   - endlocMarker.callPreparse = False
3314   - matchExpr = locMarker("_original_start") + expr + endlocMarker("_original_end")
3315   - if asString:
3316   - extractText = lambda s,l,t: s[t._original_start:t._original_end]
3317   - else:
3318   - def extractText(s,l,t):
3319   - del t[:]
3320   - t.insert(0, s[t._original_start:t._original_end])
3321   - del t["_original_start"]
3322   - del t["_original_end"]
3323   - matchExpr.setParseAction(extractText)
3324   - return matchExpr
3325   -
3326   -def ungroup(expr):
3327   - """Helper to undo pyparsing's default grouping of And expressions, even
3328   - if all but one are non-empty."""
3329   - return TokenConverter(expr).setParseAction(lambda t:t[0])
3330   -
3331   -def locatedExpr(expr):
3332   - """Helper to decorate a returned token with its starting and ending locations in the input string.
3333   - This helper adds the following results names:
3334   - - locn_start = location where matched expression begins
3335   - - locn_end = location where matched expression ends
3336   - - value = the actual parsed results
3337   -
3338   - Be careful if the input text contains C{<TAB>} characters, you may want to call
3339   - C{L{ParserElement.parseWithTabs}}
3340   - """
3341   - locator = Empty().setParseAction(lambda s,l,t: l)
3342   - return Group(locator("locn_start") + expr("value") + locator.copy().leaveWhitespace()("locn_end"))
3343   -
3344   -
3345   -# convenience constants for positional expressions
3346   -empty = Empty().setName("empty")
3347   -lineStart = LineStart().setName("lineStart")
3348   -lineEnd = LineEnd().setName("lineEnd")
3349   -stringStart = StringStart().setName("stringStart")
3350   -stringEnd = StringEnd().setName("stringEnd")
3351   -
3352   -_escapedPunc = Word( _bslash, r"\[]-*.$+^?()~ ", exact=2 ).setParseAction(lambda s,l,t:t[0][1])
3353   -_escapedHexChar = Regex(r"\\0?[xX][0-9a-fA-F]+").setParseAction(lambda s,l,t:unichr(int(t[0].lstrip(r'\0x'),16)))
3354   -_escapedOctChar = Regex(r"\\0[0-7]+").setParseAction(lambda s,l,t:unichr(int(t[0][1:],8)))
3355   -_singleChar = _escapedPunc | _escapedHexChar | _escapedOctChar | Word(printables, excludeChars=r'\]', exact=1)
3356   -_charRange = Group(_singleChar + Suppress("-") + _singleChar)
3357   -_reBracketExpr = Literal("[") + Optional("^").setResultsName("negate") + Group( OneOrMore( _charRange | _singleChar ) ).setResultsName("body") + "]"
3358   -
3359   -def srange(s):
3360   - r"""Helper to easily define string ranges for use in Word construction. Borrows
3361   - syntax from regexp '[]' string range definitions::
3362   - srange("[0-9]") -> "0123456789"
3363   - srange("[a-z]") -> "abcdefghijklmnopqrstuvwxyz"
3364   - srange("[a-z$_]") -> "abcdefghijklmnopqrstuvwxyz$_"
3365   - The input string must be enclosed in []'s, and the returned string is the expanded
3366   - character set joined into a single string.
3367   - The values enclosed in the []'s may be::
3368   - a single character
3369   - an escaped character with a leading backslash (such as \- or \])
3370   - an escaped hex character with a leading '\x' (\x21, which is a '!' character)
3371   - (\0x## is also supported for backwards compatibility)
3372   - an escaped octal character with a leading '\0' (\041, which is a '!' character)
3373   - a range of any of the above, separated by a dash ('a-z', etc.)
3374   - any combination of the above ('aeiouy', 'a-zA-Z0-9_$', etc.)
3375   - """
3376   - _expanded = lambda p: p if not isinstance(p,ParseResults) else ''.join(unichr(c) for c in range(ord(p[0]),ord(p[1])+1))
3377   - try:
3378   - return "".join(_expanded(part) for part in _reBracketExpr.parseString(s).body)
3379   - except:
3380   - return ""
3381   -
3382   -def matchOnlyAtCol(n):
3383   - """Helper method for defining parse actions that require matching at a specific
3384   - column in the input text.
3385   - """
3386   - def verifyCol(strg,locn,toks):
3387   - if col(locn,strg) != n:
3388   - raise ParseException(strg,locn,"matched token not at column %d" % n)
3389   - return verifyCol
3390   -
3391   -def replaceWith(replStr):
3392   - """Helper method for common parse actions that simply return a literal value. Especially
3393   - useful when used with C{L{transformString<ParserElement.transformString>}()}.
3394   - """
3395   - def _replFunc(*args):
3396   - return [replStr]
3397   - return _replFunc
3398   -
3399   -def removeQuotes(s,l,t):
3400   - """Helper parse action for removing quotation marks from parsed quoted strings.
3401   - To use, add this parse action to quoted string using::
3402   - quotedString.setParseAction( removeQuotes )
3403   - """
3404   - return t[0][1:-1]
3405   -
3406   -def upcaseTokens(s,l,t):
3407   - """Helper parse action to convert tokens to upper case."""
3408   - return [ tt.upper() for tt in map(_ustr,t) ]
3409   -
3410   -def downcaseTokens(s,l,t):
3411   - """Helper parse action to convert tokens to lower case."""
3412   - return [ tt.lower() for tt in map(_ustr,t) ]
3413   -
3414   -def keepOriginalText(s,startLoc,t):
3415   - """DEPRECATED - use new helper method C{L{originalTextFor}}.
3416   - Helper parse action to preserve original parsed text,
3417   - overriding any nested parse actions."""
3418   - try:
3419   - endloc = getTokensEndLoc()
3420   - except ParseException:
3421   - raise ParseFatalException("incorrect usage of keepOriginalText - may only be called as a parse action")
3422   - del t[:]
3423   - t += ParseResults(s[startLoc:endloc])
3424   - return t
3425   -
3426   -def getTokensEndLoc():
3427   - """Method to be called from within a parse action to determine the end
3428   - location of the parsed tokens."""
3429   - import inspect
3430   - fstack = inspect.stack()
3431   - try:
3432   - # search up the stack (through intervening argument normalizers) for correct calling routine
3433   - for f in fstack[2:]:
3434   - if f[3] == "_parseNoCache":
3435   - endloc = f[0].f_locals["loc"]
3436   - return endloc
3437   - else:
3438   - raise ParseFatalException("incorrect usage of getTokensEndLoc - may only be called from within a parse action")
3439   - finally:
3440   - del fstack
3441   -
3442   -def _makeTags(tagStr, xml):
3443   - """Internal helper to construct opening and closing tag expressions, given a tag name"""
3444   - if isinstance(tagStr,basestring):
3445   - resname = tagStr
3446   - tagStr = Keyword(tagStr, caseless=not xml)
3447   - else:
3448   - resname = tagStr.name
3449   -
3450   - tagAttrName = Word(alphas,alphanums+"_-:")
3451   - if (xml):
3452   - tagAttrValue = dblQuotedString.copy().setParseAction( removeQuotes )
3453   - openTag = Suppress("<") + tagStr("tag") + \
3454   - Dict(ZeroOrMore(Group( tagAttrName + Suppress("=") + tagAttrValue ))) + \
3455   - Optional("/",default=[False]).setResultsName("empty").setParseAction(lambda s,l,t:t[0]=='/') + Suppress(">")
3456   - else:
3457   - printablesLessRAbrack = "".join(c for c in printables if c not in ">")
3458   - tagAttrValue = quotedString.copy().setParseAction( removeQuotes ) | Word(printablesLessRAbrack)
3459   - openTag = Suppress("<") + tagStr("tag") + \
3460   - Dict(ZeroOrMore(Group( tagAttrName.setParseAction(downcaseTokens) + \
3461   - Optional( Suppress("=") + tagAttrValue ) ))) + \
3462   - Optional("/",default=[False]).setResultsName("empty").setParseAction(lambda s,l,t:t[0]=='/') + Suppress(">")
3463   - closeTag = Combine(_L("</") + tagStr + ">")
3464   -
3465   - openTag = openTag.setResultsName("start"+"".join(resname.replace(":"," ").title().split())).setName("<%s>" % tagStr)
3466   - closeTag = closeTag.setResultsName("end"+"".join(resname.replace(":"," ").title().split())).setName("</%s>" % tagStr)
3467   - openTag.tag = resname
3468   - closeTag.tag = resname
3469   - return openTag, closeTag
3470   -
3471   -def makeHTMLTags(tagStr):
3472   - """Helper to construct opening and closing tag expressions for HTML, given a tag name"""
3473   - return _makeTags( tagStr, False )
3474   -
3475   -def makeXMLTags(tagStr):
3476   - """Helper to construct opening and closing tag expressions for XML, given a tag name"""
3477   - return _makeTags( tagStr, True )
3478   -
3479   -def withAttribute(*args,**attrDict):
3480   - """Helper to create a validating parse action to be used with start tags created
3481   - with C{L{makeXMLTags}} or C{L{makeHTMLTags}}. Use C{withAttribute} to qualify a starting tag
3482   - with a required attribute value, to avoid false matches on common tags such as
3483   - C{<TD>} or C{<DIV>}.
3484   -
3485   - Call C{withAttribute} with a series of attribute names and values. Specify the list
3486   - of filter attributes names and values as:
3487   - - keyword arguments, as in C{(align="right")}, or
3488   - - as an explicit dict with C{**} operator, when an attribute name is also a Python
3489   - reserved word, as in C{**{"class":"Customer", "align":"right"}}
3490   - - a list of name-value tuples, as in ( ("ns1:class", "Customer"), ("ns2:align","right") )
3491   - For attribute names with a namespace prefix, you must use the second form. Attribute
3492   - names are matched insensitive to upper/lower case.
3493   -
3494   - To verify that the attribute exists, but without specifying a value, pass
3495   - C{withAttribute.ANY_VALUE} as the value.
3496   - """
3497   - if args:
3498   - attrs = args[:]
3499   - else:
3500   - attrs = attrDict.items()
3501   - attrs = [(k,v) for k,v in attrs]
3502   - def pa(s,l,tokens):
3503   - for attrName,attrValue in attrs:
3504   - if attrName not in tokens:
3505   - raise ParseException(s,l,"no matching attribute " + attrName)
3506   - if attrValue != withAttribute.ANY_VALUE and tokens[attrName] != attrValue:
3507   - raise ParseException(s,l,"attribute '%s' has value '%s', must be '%s'" %
3508   - (attrName, tokens[attrName], attrValue))
3509   - return pa
3510   -withAttribute.ANY_VALUE = object()
3511   -
3512   -opAssoc = _Constants()
3513   -opAssoc.LEFT = object()
3514   -opAssoc.RIGHT = object()
3515   -
3516   -def infixNotation( baseExpr, opList, lpar=Suppress('('), rpar=Suppress(')') ):
3517   - """Helper method for constructing grammars of expressions made up of
3518   - operators working in a precedence hierarchy. Operators may be unary or
3519   - binary, left- or right-associative. Parse actions can also be attached
3520   - to operator expressions.
3521   -
3522   - Parameters:
3523   - - baseExpr - expression representing the most basic element for the nested
3524   - - opList - list of tuples, one for each operator precedence level in the
3525   - expression grammar; each tuple is of the form
3526   - (opExpr, numTerms, rightLeftAssoc, parseAction), where:
3527   - - opExpr is the pyparsing expression for the operator;
3528   - may also be a string, which will be converted to a Literal;
3529   - if numTerms is 3, opExpr is a tuple of two expressions, for the
3530   - two operators separating the 3 terms
3531   - - numTerms is the number of terms for this operator (must
3532   - be 1, 2, or 3)
3533   - - rightLeftAssoc is the indicator whether the operator is
3534   - right or left associative, using the pyparsing-defined
3535   - constants C{opAssoc.RIGHT} and C{opAssoc.LEFT}.
3536   - - parseAction is the parse action to be associated with
3537   - expressions matching this operator expression (the
3538   - parse action tuple member may be omitted)
3539   - - lpar - expression for matching left-parentheses (default=Suppress('('))
3540   - - rpar - expression for matching right-parentheses (default=Suppress(')'))
3541   - """
3542   - ret = Forward()
3543   - lastExpr = baseExpr | ( lpar + ret + rpar )
3544   - for i,operDef in enumerate(opList):
3545   - opExpr,arity,rightLeftAssoc,pa = (operDef + (None,))[:4]
3546   - if arity == 3:
3547   - if opExpr is None or len(opExpr) != 2:
3548   - raise ValueError("if numterms=3, opExpr must be a tuple or list of two expressions")
3549   - opExpr1, opExpr2 = opExpr
3550   - thisExpr = Forward()#.setName("expr%d" % i)
3551   - if rightLeftAssoc == opAssoc.LEFT:
3552   - if arity == 1:
3553   - matchExpr = FollowedBy(lastExpr + opExpr) + Group( lastExpr + OneOrMore( opExpr ) )
3554   - elif arity == 2:
3555   - if opExpr is not None:
3556   - matchExpr = FollowedBy(lastExpr + opExpr + lastExpr) + Group( lastExpr + OneOrMore( opExpr + lastExpr ) )
3557   - else:
3558   - matchExpr = FollowedBy(lastExpr+lastExpr) + Group( lastExpr + OneOrMore(lastExpr) )
3559   - elif arity == 3:
3560   - matchExpr = FollowedBy(lastExpr + opExpr1 + lastExpr + opExpr2 + lastExpr) + \
3561   - Group( lastExpr + opExpr1 + lastExpr + opExpr2 + lastExpr )
3562   - else:
3563   - raise ValueError("operator must be unary (1), binary (2), or ternary (3)")
3564   - elif rightLeftAssoc == opAssoc.RIGHT:
3565   - if arity == 1:
3566   - # try to avoid LR with this extra test
3567   - if not isinstance(opExpr, Optional):
3568   - opExpr = Optional(opExpr)
3569   - matchExpr = FollowedBy(opExpr.expr + thisExpr) + Group( opExpr + thisExpr )
3570   - elif arity == 2:
3571   - if opExpr is not None:
3572   - matchExpr = FollowedBy(lastExpr + opExpr + thisExpr) + Group( lastExpr + OneOrMore( opExpr + thisExpr ) )
3573   - else:
3574   - matchExpr = FollowedBy(lastExpr + thisExpr) + Group( lastExpr + OneOrMore( thisExpr ) )
3575   - elif arity == 3:
3576   - matchExpr = FollowedBy(lastExpr + opExpr1 + thisExpr + opExpr2 + thisExpr) + \
3577   - Group( lastExpr + opExpr1 + thisExpr + opExpr2 + thisExpr )
3578   - else:
3579   - raise ValueError("operator must be unary (1), binary (2), or ternary (3)")
3580   - else:
3581   - raise ValueError("operator must indicate right or left associativity")
3582   - if pa:
3583   - matchExpr.setParseAction( pa )
3584   - thisExpr <<= ( matchExpr | lastExpr )
3585   - lastExpr = thisExpr
3586   - ret <<= lastExpr
3587   - return ret
3588   -operatorPrecedence = infixNotation
3589   -
3590   -dblQuotedString = Regex(r'"(?:[^"\n\r\\]|(?:"")|(?:\\x[0-9a-fA-F]+)|(?:\\.))*"').setName("string enclosed in double quotes")
3591   -sglQuotedString = Regex(r"'(?:[^'\n\r\\]|(?:'')|(?:\\x[0-9a-fA-F]+)|(?:\\.))*'").setName("string enclosed in single quotes")
3592   -quotedString = Regex(r'''(?:"(?:[^"\n\r\\]|(?:"")|(?:\\x[0-9a-fA-F]+)|(?:\\.))*")|(?:'(?:[^'\n\r\\]|(?:'')|(?:\\x[0-9a-fA-F]+)|(?:\\.))*')''').setName("quotedString using single or double quotes")
3593   -unicodeString = Combine(_L('u') + quotedString.copy())
3594   -
3595   -def nestedExpr(opener="(", closer=")", content=None, ignoreExpr=quotedString.copy()):
3596   - """Helper method for defining nested lists enclosed in opening and closing
3597   - delimiters ("(" and ")" are the default).
3598   -
3599   - Parameters:
3600   - - opener - opening character for a nested list (default="("); can also be a pyparsing expression
3601   - - closer - closing character for a nested list (default=")"); can also be a pyparsing expression
3602   - - content - expression for items within the nested lists (default=None)
3603   - - ignoreExpr - expression for ignoring opening and closing delimiters (default=quotedString)
3604   -
3605   - If an expression is not provided for the content argument, the nested
3606   - expression will capture all whitespace-delimited content between delimiters
3607   - as a list of separate values.
3608   -
3609   - Use the C{ignoreExpr} argument to define expressions that may contain
3610   - opening or closing characters that should not be treated as opening
3611   - or closing characters for nesting, such as quotedString or a comment
3612   - expression. Specify multiple expressions using an C{L{Or}} or C{L{MatchFirst}}.
3613   - The default is L{quotedString}, but if no expressions are to be ignored,
3614   - then pass C{None} for this argument.
3615   - """
3616   - if opener == closer:
3617   - raise ValueError("opening and closing strings cannot be the same")
3618   - if content is None:
3619   - if isinstance(opener,basestring) and isinstance(closer,basestring):
3620   - if len(opener) == 1 and len(closer)==1:
3621   - if ignoreExpr is not None:
3622   - content = (Combine(OneOrMore(~ignoreExpr +
3623   - CharsNotIn(opener+closer+ParserElement.DEFAULT_WHITE_CHARS,exact=1))
3624   - ).setParseAction(lambda t:t[0].strip()))
3625   - else:
3626   - content = (empty.copy()+CharsNotIn(opener+closer+ParserElement.DEFAULT_WHITE_CHARS
3627   - ).setParseAction(lambda t:t[0].strip()))
3628   - else:
3629   - if ignoreExpr is not None:
3630   - content = (Combine(OneOrMore(~ignoreExpr +
3631   - ~Literal(opener) + ~Literal(closer) +
3632   - CharsNotIn(ParserElement.DEFAULT_WHITE_CHARS,exact=1))
3633   - ).setParseAction(lambda t:t[0].strip()))
3634   - else:
3635   - content = (Combine(OneOrMore(~Literal(opener) + ~Literal(closer) +
3636   - CharsNotIn(ParserElement.DEFAULT_WHITE_CHARS,exact=1))
3637   - ).setParseAction(lambda t:t[0].strip()))
3638   - else:
3639   - raise ValueError("opening and closing arguments must be strings if no content expression is given")
3640   - ret = Forward()
3641   - if ignoreExpr is not None:
3642   - ret <<= Group( Suppress(opener) + ZeroOrMore( ignoreExpr | ret | content ) + Suppress(closer) )
3643   - else:
3644   - ret <<= Group( Suppress(opener) + ZeroOrMore( ret | content ) + Suppress(closer) )
3645   - return ret
3646   -
3647   -def indentedBlock(blockStatementExpr, indentStack, indent=True):
3648   - """Helper method for defining space-delimited indentation blocks, such as
3649   - those used to define block statements in Python source code.
3650   -
3651   - Parameters:
3652   - - blockStatementExpr - expression defining syntax of statement that
3653   - is repeated within the indented block
3654   - - indentStack - list created by caller to manage indentation stack
3655   - (multiple statementWithIndentedBlock expressions within a single grammar
3656   - should share a common indentStack)
3657   - - indent - boolean indicating whether block must be indented beyond the
3658   - the current level; set to False for block of left-most statements
3659   - (default=True)
3660   -
3661   - A valid block must contain at least one C{blockStatement}.
3662   - """
3663   - def checkPeerIndent(s,l,t):
3664   - if l >= len(s): return
3665   - curCol = col(l,s)
3666   - if curCol != indentStack[-1]:
3667   - if curCol > indentStack[-1]:
3668   - raise ParseFatalException(s,l,"illegal nesting")
3669   - raise ParseException(s,l,"not a peer entry")
3670   -
3671   - def checkSubIndent(s,l,t):
3672   - curCol = col(l,s)
3673   - if curCol > indentStack[-1]:
3674   - indentStack.append( curCol )
3675   - else:
3676   - raise ParseException(s,l,"not a subentry")
3677   -
3678   - def checkUnindent(s,l,t):
3679   - if l >= len(s): return
3680   - curCol = col(l,s)
3681   - if not(indentStack and curCol < indentStack[-1] and curCol <= indentStack[-2]):
3682   - raise ParseException(s,l,"not an unindent")
3683   - indentStack.pop()
3684   -
3685   - NL = OneOrMore(LineEnd().setWhitespaceChars("\t ").suppress())
3686   - INDENT = Empty() + Empty().setParseAction(checkSubIndent)
3687   - PEER = Empty().setParseAction(checkPeerIndent)
3688   - UNDENT = Empty().setParseAction(checkUnindent)
3689   - if indent:
3690   - smExpr = Group( Optional(NL) +
3691   - #~ FollowedBy(blockStatementExpr) +
3692   - INDENT + (OneOrMore( PEER + Group(blockStatementExpr) + Optional(NL) )) + UNDENT)
3693   - else:
3694   - smExpr = Group( Optional(NL) +
3695   - (OneOrMore( PEER + Group(blockStatementExpr) + Optional(NL) )) )
3696   - blockStatementExpr.ignore(_bslash + LineEnd())
3697   - return smExpr
3698   -
3699   -alphas8bit = srange(r"[\0xc0-\0xd6\0xd8-\0xf6\0xf8-\0xff]")
3700   -punc8bit = srange(r"[\0xa1-\0xbf\0xd7\0xf7]")
3701   -
3702   -anyOpenTag,anyCloseTag = makeHTMLTags(Word(alphas,alphanums+"_:"))
3703   -commonHTMLEntity = Combine(_L("&") + oneOf("gt lt amp nbsp quot").setResultsName("entity") +";").streamline()
3704   -_htmlEntityMap = dict(zip("gt lt amp nbsp quot".split(),'><& "'))
3705   -replaceHTMLEntity = lambda t : t.entity in _htmlEntityMap and _htmlEntityMap[t.entity] or None
3706   -
3707   -# it's easy to get these comment structures wrong - they're very common, so may as well make them available
3708   -cStyleComment = Regex(r"/\*(?:[^*]*\*+)+?/").setName("C style comment")
3709   -
3710   -htmlComment = Regex(r"<!--[\s\S]*?-->")
3711   -restOfLine = Regex(r".*").leaveWhitespace()
3712   -dblSlashComment = Regex(r"\/\/(\\\n|.)*").setName("// comment")
3713   -cppStyleComment = Regex(r"/(?:\*(?:[^*]*\*+)+?/|/[^\n]*(?:\n[^\n]*)*?(?:(?<!\\)|\Z))").setName("C++ style comment")
3714   -
3715   -javaStyleComment = cppStyleComment
3716   -pythonStyleComment = Regex(r"#.*").setName("Python style comment")
3717   -_commasepitem = Combine(OneOrMore(Word(printables, excludeChars=',') +
3718   - Optional( Word(" \t") +
3719   - ~Literal(",") + ~LineEnd() ) ) ).streamline().setName("commaItem")
3720   -commaSeparatedList = delimitedList( Optional( quotedString.copy() | _commasepitem, default="") ).setName("commaSeparatedList")
3721   -
3722   -
3723   -if __name__ == "__main__":
3724   -
3725   - def test( teststring ):
3726   - try:
3727   - tokens = simpleSQL.parseString( teststring )
3728   - tokenlist = tokens.asList()
3729   - print (teststring + "->" + str(tokenlist))
3730   - print ("tokens = " + str(tokens))
3731   - print ("tokens.columns = " + str(tokens.columns))
3732   - print ("tokens.tables = " + str(tokens.tables))
3733   - print (tokens.asXML("SQL",True))
3734   - except ParseBaseException as err:
3735   - print (teststring + "->")
3736   - print (err.line)
3737   - print (" "*(err.column-1) + "^")
3738   - print (err)
3739   - print()
3740   -
3741   - selectToken = CaselessLiteral( "select" )
3742   - fromToken = CaselessLiteral( "from" )
3743   -
3744   - ident = Word( alphas, alphanums + "_$" )
3745   - columnName = delimitedList( ident, ".", combine=True ).setParseAction( upcaseTokens )
3746   - columnNameList = Group( delimitedList( columnName ) )#.setName("columns")
3747   - tableName = delimitedList( ident, ".", combine=True ).setParseAction( upcaseTokens )
3748   - tableNameList = Group( delimitedList( tableName ) )#.setName("tables")
3749   - simpleSQL = ( selectToken + \
3750   - ( '*' | columnNameList ).setResultsName( "columns" ) + \
3751   - fromToken + \
3752   - tableNameList.setResultsName( "tables" ) )
3753   -
3754   - test( "SELECT * from XYZZY, ABC" )
3755   - test( "select * from SYS.XYZZY" )
3756   - test( "Select A from Sys.dual" )
3757   - test( "Select AA,BB,CC from Sys.dual" )
3758   - test( "Select A, B, C from Sys.dual" )
3759   - test( "Select A, B, C from Sys.dual" )
3760   - test( "Xelect A, B, C from Sys.dual" )
3761   - test( "Select A, B, C frox Sys.dual" )
3762   - test( "Select" )
3763   - test( "Select ^^^ frox Sys.dual" )
3764   - test( "Select A, B, C from Sys.dual, Table2 " )
setup.py
  View file @2ea8bb4
... ... @@ -23,6 +23,7 @@ to install this package.
23 23 # 2016-09-05 PL: - added more entry points
24 24 # 2017-01-18 v0.51 PL: - added package zipfile27 (issue #121)
25 25 # 2017-10-18 v0.52 PL: - added msodde
  26 +# 2018-03-19 v0.52.3 PL: - added install_requires, removed thirdparty.pyparsing
26 27  
27 28 #--- TODO ---------------------------------------------------------------------
28 29  
... ... @@ -42,7 +43,7 @@ import os, fnmatch
42 43 #--- METADATA -----------------------------------------------------------------
43 44  
44 45 name = "oletools"
45   -version = '0.52.2'
  46 +version = '0.52.3'
46 47 desc = "Python tools to analyze security characteristics of MS Office and OLE files (also called Structured Storage, Compound File Binary Format or Compound Document File Format), for Malware Analysis and Incident Response #DFIR"
47 48 long_desc = open('oletools/README.rst').read()
48 49 author = "Philippe Lagadec"
... ... @@ -83,7 +84,6 @@ packages=[
83 84 'oletools.thirdparty.prettytable',
84 85 'oletools.thirdparty.xglob',
85 86 'oletools.thirdparty.DridexUrlDecoder',
86   - 'oletools.thirdparty.pyparsing',
87 87 'oletools.thirdparty.colorclass',
88 88 'oletools.thirdparty.tablestream',
89 89 'oletools.thirdparty.zipfile27',
... ... @@ -181,9 +181,6 @@ package_data={
181 181 'oletools.thirdparty.DridexUrlDecoder': [
182 182 'LICENSE.txt',
183 183 ],
184   - 'oletools.thirdparty.pyparsing': [
185   - 'LICENSE', 'README',
186   - ],
187 184 'oletools.thirdparty.colorclass': [
188 185 'LICENSE.txt',
189 186 ],
... ... @@ -322,6 +319,7 @@ def main():
322 319 entry_points=entry_points,
323 320 test_suite="tests",
324 321 # scripts=scripts,
  322 + install_requires=["pyparsing"],
325 323 )
326 324  
327 325  
... ...