[3] | 1 | # module pyparsing.py
|
---|
| 2 | #
|
---|
| 3 | # Copyright (c) 2003-2006 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 | # Todo:
|
---|
| 25 | # - add pprint() - pretty-print output of defined BNF
|
---|
| 26 | #
|
---|
| 27 | #from __future__ import generators
|
---|
| 28 |
|
---|
| 29 | __doc__ = \
|
---|
| 30 | """
|
---|
| 31 | pyparsing module - Classes and methods to define and execute parsing grammars
|
---|
| 32 |
|
---|
| 33 | The pyparsing module is an alternative approach to creating and executing simple grammars,
|
---|
| 34 | vs. the traditional lex/yacc approach, or the use of regular expressions. With pyparsing, you
|
---|
| 35 | don't need to learn a new syntax for defining grammars or matching expressions - the parsing module
|
---|
| 36 | provides a library of classes that you use to construct the grammar directly in Python.
|
---|
| 37 |
|
---|
| 38 | Here is a program to parse "Hello, World!" (or any greeting of the form "<salutation>, <addressee>!")::
|
---|
| 39 |
|
---|
| 40 | from pyparsing import Word, alphas
|
---|
| 41 |
|
---|
| 42 | # define grammar of a greeting
|
---|
| 43 | greet = Word( alphas ) + "," + Word( alphas ) + "!"
|
---|
| 44 |
|
---|
| 45 | hello = "Hello, World!"
|
---|
| 46 | print hello, "->", greet.parseString( hello )
|
---|
| 47 |
|
---|
| 48 | The program outputs the following::
|
---|
| 49 |
|
---|
| 50 | Hello, World! -> ['Hello', ',', 'World', '!']
|
---|
| 51 |
|
---|
| 52 | The Python representation of the grammar is quite readable, owing to the self-explanatory
|
---|
| 53 | class names, and the use of '+', '|' and '^' operators.
|
---|
| 54 |
|
---|
| 55 | The parsed results returned from parseString() can be accessed as a nested list, a dictionary, or an
|
---|
| 56 | object with named attributes.
|
---|
| 57 |
|
---|
| 58 | The pyparsing module handles some of the problems that are typically vexing when writing text parsers:
|
---|
| 59 | - extra or missing whitespace (the above program will also handle "Hello,World!", "Hello , World !", etc.)
|
---|
| 60 | - quoted strings
|
---|
| 61 | - embedded comments
|
---|
| 62 | """
|
---|
| 63 | __version__ = "1.4.1"
|
---|
| 64 | __versionTime__ = "05 February 2006 12:24"
|
---|
| 65 | __author__ = "Paul McGuire <ptmcg@users.sourceforge.net>"
|
---|
| 66 |
|
---|
| 67 | import string
|
---|
| 68 | import copy,sys
|
---|
| 69 | import warnings
|
---|
| 70 | import re
|
---|
| 71 | #~ sys.stderr.write( "testing pyparsing module, version %s, %s\n" % (__version__,__versionTime__ ) )
|
---|
| 72 |
|
---|
| 73 | def _ustr(obj):
|
---|
| 74 | """Drop-in replacement for str(obj) that tries to be Unicode friendly. It first tries
|
---|
| 75 | str(obj). If that fails with a UnicodeEncodeError, then it tries unicode(obj). It
|
---|
| 76 | then < returns the unicode object | encodes it with the default encoding | ... >.
|
---|
| 77 | """
|
---|
| 78 | try:
|
---|
| 79 | # If this works, then _ustr(obj) has the same behaviour as str(obj), so
|
---|
| 80 | # it won't break any existing code.
|
---|
| 81 | return str(obj)
|
---|
| 82 |
|
---|
| 83 | except UnicodeEncodeError, e:
|
---|
| 84 | # The Python docs (http://docs.python.org/ref/customization.html#l2h-182)
|
---|
| 85 | # state that "The return value must be a string object". However, does a
|
---|
| 86 | # unicode object (being a subclass of basestring) count as a "string
|
---|
| 87 | # object"?
|
---|
| 88 | # If so, then return a unicode object:
|
---|
| 89 | return unicode(obj)
|
---|
| 90 | # Else encode it... but how? There are many choices... :)
|
---|
| 91 | # Replace unprintables with escape codes?
|
---|
| 92 | #return unicode(obj).encode(sys.getdefaultencoding(), 'backslashreplace_errors')
|
---|
| 93 | # Replace unprintables with question marks?
|
---|
| 94 | #return unicode(obj).encode(sys.getdefaultencoding(), 'replace')
|
---|
| 95 | # ...
|
---|
| 96 |
|
---|
| 97 | def _str2dict(strg):
|
---|
| 98 | return dict( [(c,0) for c in strg] )
|
---|
| 99 |
|
---|
| 100 | alphas = string.lowercase + string.uppercase
|
---|
| 101 | nums = string.digits
|
---|
| 102 | hexnums = nums + "ABCDEFabcdef"
|
---|
| 103 | alphanums = alphas + nums
|
---|
| 104 |
|
---|
| 105 | class ParseBaseException(Exception):
|
---|
| 106 | """base exception class for all parsing runtime exceptions"""
|
---|
| 107 | __slots__ = ( "loc","msg","pstr","parserElement" )
|
---|
| 108 | # Performance tuning: we construct a *lot* of these, so keep this
|
---|
| 109 | # constructor as small and fast as possible
|
---|
| 110 | def __init__( self, pstr, loc, msg, elem=None ):
|
---|
| 111 | self.loc = loc
|
---|
| 112 | self.msg = msg
|
---|
| 113 | self.pstr = pstr
|
---|
| 114 | self.parserElement = elem
|
---|
| 115 |
|
---|
| 116 | def __getattr__( self, aname ):
|
---|
| 117 | """supported attributes by name are:
|
---|
| 118 | - lineno - returns the line number of the exception text
|
---|
| 119 | - col - returns the column number of the exception text
|
---|
| 120 | - line - returns the line containing the exception text
|
---|
| 121 | """
|
---|
| 122 | if( aname == "lineno" ):
|
---|
| 123 | return lineno( self.loc, self.pstr )
|
---|
| 124 | elif( aname in ("col", "column") ):
|
---|
| 125 | return col( self.loc, self.pstr )
|
---|
| 126 | elif( aname == "line" ):
|
---|
| 127 | return line( self.loc, self.pstr )
|
---|
| 128 | else:
|
---|
| 129 | raise AttributeError, aname
|
---|
| 130 |
|
---|
| 131 | def __str__( self ):
|
---|
| 132 | return "%s (at char %d), (line:%d, col:%d)" % ( self.msg, self.loc, self.lineno, self.column )
|
---|
| 133 | def __repr__( self ):
|
---|
| 134 | return _ustr(self)
|
---|
| 135 | def markInputline( self, markerString = ">!<" ):
|
---|
| 136 | """Extracts the exception line from the input string, and marks
|
---|
| 137 | the location of the exception with a special symbol.
|
---|
| 138 | """
|
---|
| 139 | line_str = self.line
|
---|
| 140 | line_column = self.column - 1
|
---|
| 141 | if markerString:
|
---|
| 142 | line_str = "".join( [line_str[:line_column], markerString, line_str[line_column:]])
|
---|
| 143 | return line_str.strip()
|
---|
| 144 |
|
---|
| 145 | class ParseException(ParseBaseException):
|
---|
| 146 | """exception thrown when parse expressions don't match class"""
|
---|
| 147 | """supported attributes by name are:
|
---|
| 148 | - lineno - returns the line number of the exception text
|
---|
| 149 | - col - returns the column number of the exception text
|
---|
| 150 | - line - returns the line containing the exception text
|
---|
| 151 | """
|
---|
| 152 | pass
|
---|
| 153 |
|
---|
| 154 | class ParseFatalException(ParseBaseException):
|
---|
| 155 | """user-throwable exception thrown when inconsistent parse content
|
---|
| 156 | is found; stops all parsing immediately"""
|
---|
| 157 | pass
|
---|
| 158 |
|
---|
| 159 | class RecursiveGrammarException(Exception):
|
---|
| 160 | """exception thrown by validate() if the grammar could be improperly recursive"""
|
---|
| 161 | def __init__( self, parseElementList ):
|
---|
| 162 | self.parseElementTrace = parseElementList
|
---|
| 163 |
|
---|
| 164 | def __str__( self ):
|
---|
| 165 | return "RecursiveGrammarException: %s" % self.parseElementTrace
|
---|
| 166 |
|
---|
| 167 | class ParseResults(object):
|
---|
| 168 | """Structured parse results, to provide multiple means of access to the parsed data:
|
---|
| 169 | - as a list (len(results))
|
---|
| 170 | - by list index (results[0], results[1], etc.)
|
---|
| 171 | - by attribute (results.<resultsName>)
|
---|
| 172 | """
|
---|
| 173 | __slots__ = ( "__toklist", "__tokdict", "__doinit", "__name", "__parent", "__modal" )
|
---|
| 174 | def __new__(cls, toklist, name=None, asList=True, modal=True ):
|
---|
| 175 | if isinstance(toklist, cls):
|
---|
| 176 | return toklist
|
---|
| 177 | retobj = object.__new__(cls)
|
---|
| 178 | retobj.__doinit = True
|
---|
| 179 | return retobj
|
---|
| 180 |
|
---|
| 181 | # Performance tuning: we construct a *lot* of these, so keep this
|
---|
| 182 | # constructor as small and fast as possible
|
---|
| 183 | def __init__( self, toklist, name=None, asList=True, modal=True ):
|
---|
| 184 | if self.__doinit:
|
---|
| 185 | self.__doinit = False
|
---|
| 186 | self.__name = None
|
---|
| 187 | self.__parent = None
|
---|
| 188 | self.__modal = modal
|
---|
| 189 | if isinstance(toklist, list):
|
---|
| 190 | self.__toklist = toklist[:]
|
---|
| 191 | else:
|
---|
| 192 | self.__toklist = [toklist]
|
---|
| 193 | self.__tokdict = dict()
|
---|
| 194 |
|
---|
| 195 | if name:
|
---|
| 196 | if not self.__name:
|
---|
| 197 | self.__modal = self.__modal and modal
|
---|
| 198 | if isinstance(name,int):
|
---|
| 199 | name = _ustr(name) # will always return a str, but use _ustr for consistency
|
---|
| 200 | self.__name = name
|
---|
| 201 | if not toklist in (None,'',[]):
|
---|
| 202 | if isinstance(toklist,basestring):
|
---|
| 203 | toklist = [ toklist ]
|
---|
| 204 | if asList:
|
---|
| 205 | if isinstance(toklist,ParseResults):
|
---|
| 206 | self[name] = (toklist.copy(),-1)
|
---|
| 207 | else:
|
---|
| 208 | self[name] = (ParseResults(toklist[0]),-1)
|
---|
| 209 | self[name].__name = name
|
---|
| 210 | else:
|
---|
| 211 | try:
|
---|
| 212 | self[name] = toklist[0]
|
---|
| 213 | except TypeError:
|
---|
| 214 | self[name] = toklist
|
---|
| 215 |
|
---|
| 216 | def __getitem__( self, i ):
|
---|
| 217 | if isinstance( i, (int,slice) ):
|
---|
| 218 | return self.__toklist[i]
|
---|
| 219 | else:
|
---|
| 220 | if self.__modal:
|
---|
| 221 | return self.__tokdict[i][-1][0]
|
---|
| 222 | else:
|
---|
| 223 | return ParseResults([ v[0] for v in self.__tokdict[i] ])
|
---|
| 224 |
|
---|
| 225 | def __setitem__( self, k, v ):
|
---|
| 226 | if isinstance(v,tuple):
|
---|
| 227 | self.__tokdict[k] = self.__tokdict.get(k,list()) + [v]
|
---|
| 228 | sub = v[0]
|
---|
| 229 | else:
|
---|
| 230 | self.__tokdict[k] = self.__tokdict.get(k,list()) + [(v,0)]
|
---|
| 231 | sub = v
|
---|
| 232 | if isinstance(sub,ParseResults):
|
---|
| 233 | sub.__parent = self
|
---|
| 234 |
|
---|
| 235 | def __delitem__( self, i ):
|
---|
| 236 | del self.__toklist[i]
|
---|
| 237 |
|
---|
| 238 | def __contains__( self, k ):
|
---|
| 239 | return self.__tokdict.has_key(k)
|
---|
| 240 |
|
---|
| 241 | def __len__( self ): return len( self.__toklist )
|
---|
| 242 | def __iter__( self ): return iter( self.__toklist )
|
---|
| 243 | def keys( self ):
|
---|
| 244 | """Returns all named result keys."""
|
---|
| 245 | return self.__tokdict.keys()
|
---|
| 246 |
|
---|
| 247 | def items( self ):
|
---|
| 248 | """Returns all named result keys and values as a list of tuples."""
|
---|
| 249 | return [(k,v[-1][0]) for k,v in self.__tokdict.items()]
|
---|
| 250 |
|
---|
| 251 | def values( self ):
|
---|
| 252 | """Returns all named result values."""
|
---|
| 253 | return [ v[-1][0] for v in self.__tokdict.values() ]
|
---|
| 254 |
|
---|
| 255 | def __getattr__( self, name ):
|
---|
| 256 | if name not in self.__slots__:
|
---|
| 257 | if self.__tokdict.has_key( name ):
|
---|
| 258 | if self.__modal:
|
---|
| 259 | return self.__tokdict[name][-1][0]
|
---|
| 260 | else:
|
---|
| 261 | return ParseResults([ v[0] for v in self.__tokdict[name] ])
|
---|
| 262 | else:
|
---|
| 263 | return ""
|
---|
| 264 | return None
|
---|
| 265 |
|
---|
| 266 | def __iadd__( self, other ):
|
---|
| 267 | if other.__tokdict:
|
---|
| 268 | offset = len(self.__toklist)
|
---|
| 269 | addoffset = ( lambda a: (a<0 and offset) or (a+offset) )
|
---|
| 270 | otherdictitems = [(k,(v[0],addoffset(v[1])) ) for (k,vlist) in other.__tokdict.items() for v in vlist]
|
---|
| 271 | for k,v in otherdictitems:
|
---|
| 272 | self[k] = v
|
---|
| 273 | if isinstance(v[0],ParseResults):
|
---|
| 274 | v[0].__parent = self
|
---|
| 275 | self.__toklist += other.__toklist
|
---|
| 276 | del other
|
---|
| 277 | return self
|
---|
| 278 |
|
---|
| 279 | def __repr__( self ):
|
---|
| 280 | return "(%s, %s)" % ( repr( self.__toklist ), repr( self.__tokdict ) )
|
---|
| 281 |
|
---|
| 282 | def __str__( self ):
|
---|
| 283 | out = "["
|
---|
| 284 | sep = ""
|
---|
| 285 | for i in self.__toklist:
|
---|
| 286 | if isinstance(i, ParseResults):
|
---|
| 287 | out += sep + _ustr(i)
|
---|
| 288 | else:
|
---|
| 289 | out += sep + repr(i)
|
---|
| 290 | sep = ", "
|
---|
| 291 | out += "]"
|
---|
| 292 | return out
|
---|
| 293 |
|
---|
| 294 | def _asStringList( self, sep='' ):
|
---|
| 295 | out = []
|
---|
| 296 | for item in self.__toklist:
|
---|
| 297 | if out and sep:
|
---|
| 298 | out.append(sep)
|
---|
| 299 | if isinstance( item, ParseResults ):
|
---|
| 300 | out += item._asStringList()
|
---|
| 301 | else:
|
---|
| 302 | out.append( _ustr(item) )
|
---|
| 303 | return out
|
---|
| 304 |
|
---|
| 305 | def asList( self ):
|
---|
| 306 | """Returns the parse results as a nested list of matching tokens, all converted to strings."""
|
---|
| 307 | out = []
|
---|
| 308 | for res in self.__toklist:
|
---|
| 309 | if isinstance(res,ParseResults):
|
---|
| 310 | out.append( res.asList() )
|
---|
| 311 | else:
|
---|
| 312 | out.append( res )
|
---|
| 313 | return out
|
---|
| 314 |
|
---|
| 315 | def asDict( self ):
|
---|
| 316 | """Returns the named parse results as dictionary."""
|
---|
| 317 | return dict( self.items() )
|
---|
| 318 |
|
---|
| 319 | def copy( self ):
|
---|
| 320 | """Returns a new copy of a ParseResults object."""
|
---|
| 321 | ret = ParseResults( self.__toklist )
|
---|
| 322 | ret.__tokdict = self.__tokdict.copy()
|
---|
| 323 | ret.__parent = self.__parent
|
---|
| 324 | ret.__modal = self.__modal
|
---|
| 325 | ret.__name = self.__name
|
---|
| 326 | return ret
|
---|
| 327 |
|
---|
| 328 | def asXML( self, doctag=None, namedItemsOnly=False, indent="", formatted=True ):
|
---|
| 329 | """Returns the parse results as XML. Tags are created for tokens and lists that have defined results names."""
|
---|
| 330 | nl = "\n"
|
---|
| 331 | out = []
|
---|
| 332 | namedItems = dict( [ (v[1],k) for (k,vlist) in self.__tokdict.items() for v in vlist ] )
|
---|
| 333 | nextLevelIndent = indent + " "
|
---|
| 334 |
|
---|
| 335 | # collapse out indents if formatting is not desired
|
---|
| 336 | if not formatted:
|
---|
| 337 | indent = ""
|
---|
| 338 | nextLevelIndent = ""
|
---|
| 339 | nl = ""
|
---|
| 340 |
|
---|
| 341 | selfTag = None
|
---|
| 342 | if doctag is not None:
|
---|
| 343 | selfTag = doctag
|
---|
| 344 | else:
|
---|
| 345 | if self.__name:
|
---|
| 346 | selfTag = self.__name
|
---|
| 347 |
|
---|
| 348 | if not selfTag:
|
---|
| 349 | if namedItemsOnly:
|
---|
| 350 | return ""
|
---|
| 351 | else:
|
---|
| 352 | selfTag = "ITEM"
|
---|
| 353 |
|
---|
| 354 | out += [ nl, indent, "<", selfTag, ">" ]
|
---|
| 355 |
|
---|
| 356 | worklist = self.__toklist
|
---|
| 357 | for i,res in enumerate(worklist):
|
---|
| 358 | if isinstance(res,ParseResults):
|
---|
| 359 | if i in namedItems:
|
---|
| 360 | out += [ res.asXML(namedItems[i], namedItemsOnly and doctag is None, nextLevelIndent,formatted)]
|
---|
| 361 | else:
|
---|
| 362 | out += [ res.asXML(None, namedItemsOnly and doctag is None, nextLevelIndent,formatted)]
|
---|
| 363 | else:
|
---|
| 364 | # individual token, see if there is a name for it
|
---|
| 365 | resTag = None
|
---|
| 366 | if i in namedItems:
|
---|
| 367 | resTag = namedItems[i]
|
---|
| 368 | if not resTag:
|
---|
| 369 | if namedItemsOnly:
|
---|
| 370 | continue
|
---|
| 371 | else:
|
---|
| 372 | resTag = "ITEM"
|
---|
| 373 | out += [ nl, nextLevelIndent, "<", resTag, ">", _ustr(res), "</", resTag, ">" ]
|
---|
| 374 |
|
---|
| 375 | out += [ nl, indent, "</", selfTag, ">" ]
|
---|
| 376 | return "".join(out)
|
---|
| 377 |
|
---|
| 378 |
|
---|
| 379 | def __lookup(self,sub):
|
---|
| 380 | for k,vlist in self.__tokdict.items():
|
---|
| 381 | for v,loc in vlist:
|
---|
| 382 | if sub is v:
|
---|
| 383 | return k
|
---|
| 384 | return None
|
---|
| 385 |
|
---|
| 386 | def getName(self):
|
---|
| 387 | """Returns the results name for this token expression."""
|
---|
| 388 | if self.__name:
|
---|
| 389 | return self.__name
|
---|
| 390 | elif self.__parent:
|
---|
| 391 | par = self.__parent
|
---|
| 392 | if par:
|
---|
| 393 | return par.__lookup(self)
|
---|
| 394 | else:
|
---|
| 395 | return None
|
---|
| 396 | elif (len(self) == 1 and
|
---|
| 397 | len(self.__tokdict) == 1 and
|
---|
| 398 | self.__tokdict.values()[0][0][1] in (0,-1)):
|
---|
| 399 | return self.__tokdict.keys()[0]
|
---|
| 400 | else:
|
---|
| 401 | return None
|
---|
| 402 |
|
---|
| 403 | def col (loc,strg):
|
---|
| 404 | """Returns current column within a string, counting newlines as line separators.
|
---|
| 405 | The first column is number 1.
|
---|
| 406 | """
|
---|
| 407 | return loc - strg.rfind("\n", 0, loc)
|
---|
| 408 |
|
---|
| 409 | def lineno(loc,strg):
|
---|
| 410 | """Returns current line number within a string, counting newlines as line separators.
|
---|
| 411 | The first line is number 1.
|
---|
| 412 | """
|
---|
| 413 | return strg.count("\n",0,loc) + 1
|
---|
| 414 |
|
---|
| 415 | def line( loc, strg ):
|
---|
| 416 | """Returns the line of text containing loc within a string, counting newlines as line separators.
|
---|
| 417 | """
|
---|
| 418 | lastCR = strg.rfind("\n", 0, loc)
|
---|
| 419 | nextCR = strg.find("\n", loc)
|
---|
| 420 | if nextCR > 0:
|
---|
| 421 | return strg[lastCR+1:nextCR]
|
---|
| 422 | else:
|
---|
| 423 | return strg[lastCR+1:]
|
---|
| 424 |
|
---|
| 425 | def _defaultStartDebugAction( instring, loc, expr ):
|
---|
| 426 | print "Match",expr,"at loc",loc,"(%d,%d)" % ( lineno(loc,instring), col(loc,instring) )
|
---|
| 427 |
|
---|
| 428 | def _defaultSuccessDebugAction( instring, startloc, endloc, expr, toks ):
|
---|
| 429 | print "Matched",expr,"->",toks.asList()
|
---|
| 430 |
|
---|
| 431 | def _defaultExceptionDebugAction( instring, loc, expr, exc ):
|
---|
| 432 | print "Exception raised:", exc
|
---|
| 433 |
|
---|
| 434 | def nullDebugAction(*args):
|
---|
| 435 | """'Do-nothing' debug action, to suppress debugging output during parsing."""
|
---|
| 436 | pass
|
---|
| 437 |
|
---|
| 438 | class ParserElement(object):
|
---|
| 439 | """Abstract base level parser element class."""
|
---|
| 440 | DEFAULT_WHITE_CHARS = " \n\t\r"
|
---|
| 441 |
|
---|
| 442 | def setDefaultWhitespaceChars( chars ):
|
---|
| 443 | """Overrides the default whitespace chars
|
---|
| 444 | """
|
---|
| 445 | ParserElement.DEFAULT_WHITE_CHARS = chars
|
---|
| 446 | setDefaultWhitespaceChars = staticmethod(setDefaultWhitespaceChars)
|
---|
| 447 |
|
---|
| 448 | def __init__( self, savelist=False ):
|
---|
| 449 | self.parseAction = None
|
---|
| 450 | #~ self.name = "<unknown>" # don't define self.name, let subclasses try/except upcall
|
---|
| 451 | self.strRepr = None
|
---|
| 452 | self.resultsName = None
|
---|
| 453 | self.saveAsList = savelist
|
---|
| 454 | self.skipWhitespace = True
|
---|
| 455 | self.whiteChars = ParserElement.DEFAULT_WHITE_CHARS
|
---|
| 456 | self.mayReturnEmpty = False
|
---|
| 457 | self.keepTabs = False
|
---|
| 458 | self.ignoreExprs = []
|
---|
| 459 | self.debug = False
|
---|
| 460 | self.streamlined = False
|
---|
| 461 | self.mayIndexError = True
|
---|
| 462 | self.errmsg = ""
|
---|
| 463 | self.modalResults = True
|
---|
| 464 | self.debugActions = ( None, None, None )
|
---|
| 465 | self.re = None
|
---|
| 466 |
|
---|
| 467 | def copy( self ):
|
---|
| 468 | """Make a copy of this ParserElement. Useful for defining different parse actions
|
---|
| 469 | for the same parsing pattern, using copies of the original parse element."""
|
---|
| 470 | cpy = copy.copy( self )
|
---|
| 471 | cpy.whiteChars = ParserElement.DEFAULT_WHITE_CHARS
|
---|
| 472 | return cpy
|
---|
| 473 |
|
---|
| 474 | def setName( self, name ):
|
---|
| 475 | """Define name for this expression, for use in debugging."""
|
---|
| 476 | self.name = name
|
---|
| 477 | self.errmsg = "Expected " + self.name
|
---|
| 478 | return self
|
---|
| 479 |
|
---|
| 480 | def setResultsName( self, name, listAllMatches=False ):
|
---|
| 481 | """Define name for referencing matching tokens as a nested attribute
|
---|
| 482 | of the returned parse results.
|
---|
| 483 | NOTE: this returns a *copy* of the original ParserElement object;
|
---|
| 484 | this is so that the client can define a basic element, such as an
|
---|
| 485 | integer, and reference it in multiple places with different names.
|
---|
| 486 | """
|
---|
| 487 | newself = self.copy()
|
---|
| 488 | newself.resultsName = name
|
---|
| 489 | newself.modalResults = not listAllMatches
|
---|
| 490 | return newself
|
---|
| 491 |
|
---|
| 492 | def setParseAction( self, fn ):
|
---|
| 493 | """Define action to perform when successfully matching parse element definition.
|
---|
| 494 | Parse action fn is a callable method with the arguments (s, loc, toks) where:
|
---|
| 495 | - s = the original string being parsed
|
---|
| 496 | - loc = the location of the matching substring
|
---|
| 497 | - toks = a list of the matched tokens, packaged as a ParseResults object
|
---|
| 498 | If the function fn modifies the tokens, it can return them as the return
|
---|
| 499 | value from fn, and the modified list of tokens will replace the original.
|
---|
| 500 | Otherwise, fn does not need to return any value.
|
---|
| 501 | """
|
---|
| 502 | self.parseAction = fn
|
---|
| 503 | return self
|
---|
| 504 |
|
---|
| 505 | def skipIgnorables( self, instring, loc ):
|
---|
| 506 | exprsFound = True
|
---|
| 507 | while exprsFound:
|
---|
| 508 | exprsFound = False
|
---|
| 509 | for e in self.ignoreExprs:
|
---|
| 510 | try:
|
---|
| 511 | while 1:
|
---|
| 512 | loc,dummy = e.parse( instring, loc )
|
---|
| 513 | exprsFound = True
|
---|
| 514 | except ParseException:
|
---|
| 515 | pass
|
---|
| 516 | return loc
|
---|
| 517 |
|
---|
| 518 | def preParse( self, instring, loc ):
|
---|
| 519 | if self.ignoreExprs:
|
---|
| 520 | loc = self.skipIgnorables( instring, loc )
|
---|
| 521 |
|
---|
| 522 | if self.skipWhitespace:
|
---|
| 523 | wt = self.whiteChars
|
---|
| 524 | instrlen = len(instring)
|
---|
| 525 | while loc < instrlen and instring[loc] in wt:
|
---|
| 526 | loc += 1
|
---|
| 527 |
|
---|
| 528 | return loc
|
---|
| 529 |
|
---|
| 530 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 531 | return loc, []
|
---|
| 532 |
|
---|
| 533 | def postParse( self, instring, loc, tokenlist ):
|
---|
| 534 | return tokenlist
|
---|
| 535 |
|
---|
| 536 | #~ @profile
|
---|
| 537 | def parse( self, instring, loc, doActions=True, callPreParse=True ):
|
---|
| 538 | debugging = ( self.debug ) #and doActions )
|
---|
| 539 |
|
---|
| 540 | if debugging:
|
---|
| 541 | #~ print "Match",self,"at loc",loc,"(%d,%d)" % ( lineno(loc,instring), col(loc,instring) )
|
---|
| 542 | if (self.debugActions[0] ):
|
---|
| 543 | self.debugActions[0]( instring, loc, self )
|
---|
| 544 | if callPreParse:
|
---|
| 545 | loc = self.preParse( instring, loc )
|
---|
| 546 | tokensStart = loc
|
---|
| 547 | try:
|
---|
| 548 | try:
|
---|
| 549 | loc,tokens = self.parseImpl( instring, loc, doActions )
|
---|
| 550 | except IndexError:
|
---|
| 551 | raise ParseException, ( instring, len(instring), self.errmsg, self )
|
---|
| 552 | except ParseException, err:
|
---|
| 553 | #~ print "Exception raised:", err
|
---|
| 554 | if (self.debugActions[2] ):
|
---|
| 555 | self.debugActions[2]( instring, tokensStart, self, err )
|
---|
| 556 | raise
|
---|
| 557 | else:
|
---|
| 558 | if callPreParse:
|
---|
| 559 | loc = self.preParse( instring, loc )
|
---|
| 560 | tokensStart = loc
|
---|
| 561 | if self.mayIndexError or loc >= len(instring):
|
---|
| 562 | try:
|
---|
| 563 | loc,tokens = self.parseImpl( instring, loc, doActions )
|
---|
| 564 | except IndexError:
|
---|
| 565 | raise ParseException, ( instring, len(instring), self.errmsg, self )
|
---|
| 566 | else:
|
---|
| 567 | loc,tokens = self.parseImpl( instring, loc, doActions )
|
---|
| 568 |
|
---|
| 569 | tokens = self.postParse( instring, loc, tokens )
|
---|
| 570 |
|
---|
| 571 | retTokens = ParseResults( tokens, self.resultsName, asList=self.saveAsList, modal=self.modalResults )
|
---|
| 572 | if self.parseAction and doActions:
|
---|
| 573 | if debugging:
|
---|
| 574 | try:
|
---|
| 575 | tokens = self.parseAction( instring, tokensStart, retTokens )
|
---|
| 576 | if tokens is not None:
|
---|
| 577 | if isinstance(tokens,tuple):
|
---|
| 578 | tokens = tokens[1]
|
---|
| 579 | retTokens = ParseResults( tokens,
|
---|
| 580 | self.resultsName,
|
---|
| 581 | asList=self.saveAsList and isinstance(tokens,(ParseResults,list)),
|
---|
| 582 | modal=self.modalResults )
|
---|
| 583 | except ParseException, err:
|
---|
| 584 | #~ print "Exception raised in user parse action:", err
|
---|
| 585 | if (self.debugActions[2] ):
|
---|
| 586 | self.debugActions[2]( instring, tokensStart, self, err )
|
---|
| 587 | raise
|
---|
| 588 | else:
|
---|
| 589 | tokens = self.parseAction( instring, tokensStart, retTokens )
|
---|
| 590 | if tokens is not None:
|
---|
| 591 | if isinstance(tokens,tuple):
|
---|
| 592 | tokens = tokens[1]
|
---|
| 593 | retTokens = ParseResults( tokens,
|
---|
| 594 | self.resultsName,
|
---|
| 595 | asList=self.saveAsList and isinstance(tokens,(ParseResults,list)),
|
---|
| 596 | modal=self.modalResults )
|
---|
| 597 |
|
---|
| 598 | if debugging:
|
---|
| 599 | #~ print "Matched",self,"->",retTokens.asList()
|
---|
| 600 | if (self.debugActions[1] ):
|
---|
| 601 | self.debugActions[1]( instring, tokensStart, loc, self, retTokens )
|
---|
| 602 |
|
---|
| 603 | return loc, retTokens
|
---|
| 604 |
|
---|
| 605 | def tryParse( self, instring, loc ):
|
---|
| 606 | return self.parse( instring, loc, doActions=False )[0]
|
---|
| 607 |
|
---|
| 608 | def parseString( self, instring ):
|
---|
| 609 | """Execute the parse expression with the given string.
|
---|
| 610 | This is the main interface to the client code, once the complete
|
---|
| 611 | expression has been built.
|
---|
| 612 | """
|
---|
| 613 | if not self.streamlined:
|
---|
| 614 | self.streamline()
|
---|
| 615 | self.saveAsList = True
|
---|
| 616 | for e in self.ignoreExprs:
|
---|
| 617 | e.streamline()
|
---|
| 618 | if self.keepTabs:
|
---|
| 619 | loc, tokens = self.parse( instring, 0 )
|
---|
| 620 | else:
|
---|
| 621 | loc, tokens = self.parse( instring.expandtabs(), 0 )
|
---|
| 622 | return tokens
|
---|
| 623 |
|
---|
| 624 | def scanString( self, instring ):
|
---|
| 625 | """Scan the input string for expression matches. Each match will return the matching tokens, start location, and end location."""
|
---|
| 626 | if not self.streamlined:
|
---|
| 627 | self.streamline()
|
---|
| 628 | for e in self.ignoreExprs:
|
---|
| 629 | e.streamline()
|
---|
| 630 |
|
---|
| 631 | if not self.keepTabs:
|
---|
| 632 | instring = instring.expandtabs()
|
---|
| 633 | instrlen = len(instring)
|
---|
| 634 | loc = 0
|
---|
| 635 | preparseFn = self.preParse
|
---|
| 636 | parseFn = self.parse
|
---|
| 637 | while loc < instrlen:
|
---|
| 638 | try:
|
---|
| 639 | loc = preparseFn( instring, loc )
|
---|
| 640 | nextLoc,tokens = parseFn( instring, loc, callPreParse=False )
|
---|
| 641 | except ParseException:
|
---|
| 642 | loc += 1
|
---|
| 643 | else:
|
---|
| 644 | yield tokens, loc, nextLoc
|
---|
| 645 | loc = nextLoc
|
---|
| 646 |
|
---|
| 647 | def transformString( self, instring ):
|
---|
| 648 | """Extension to scanString, to modify matching text with modified tokens that may
|
---|
| 649 | be returned from a parse action. To use transformString, define a grammar and
|
---|
| 650 | attach a parse action to it that modifies the returned token list.
|
---|
| 651 | Invoking transformString() on a target string will then scan for matches,
|
---|
| 652 | and replace the matched text patterns according to the logic in the parse
|
---|
| 653 | action. transformString() returns the resulting transformed string."""
|
---|
| 654 | out = []
|
---|
| 655 | lastE = 0
|
---|
| 656 | # force preservation of <TAB>s, to minimize unwanted transformation of string, and to
|
---|
| 657 | # keep string locs straight between transformString and scanString
|
---|
| 658 | self.keepTabs = True
|
---|
| 659 | for t,s,e in self.scanString( instring ):
|
---|
| 660 | out.append( instring[lastE:s] )
|
---|
| 661 | if t:
|
---|
| 662 | if isinstance(t,ParseResults):
|
---|
| 663 | out += t.asList()
|
---|
| 664 | elif isinstance(t,list):
|
---|
| 665 | out += t
|
---|
| 666 | else:
|
---|
| 667 | out.append(t)
|
---|
| 668 | lastE = e
|
---|
| 669 | out.append(instring[lastE:])
|
---|
| 670 | return "".join(out)
|
---|
| 671 |
|
---|
| 672 | def searchString( self, instring ):
|
---|
| 673 | """Another extension to scanString, simplifying the access to the tokens found
|
---|
| 674 | to match the given parse expression.
|
---|
| 675 | """
|
---|
| 676 | return [ t[0] for t,s,e in self.scanString( instring ) ]
|
---|
| 677 |
|
---|
| 678 | def __add__(self, other ):
|
---|
| 679 | """Implementation of + operator - returns And"""
|
---|
| 680 | if isinstance( other, basestring ):
|
---|
| 681 | other = Literal( other )
|
---|
| 682 | if not isinstance( other, ParserElement ):
|
---|
| 683 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 684 | SyntaxWarning, stacklevel=2)
|
---|
| 685 | return And( [ self, other ] )
|
---|
| 686 |
|
---|
| 687 | def __radd__(self, other ):
|
---|
| 688 | """Implementation of += operator"""
|
---|
| 689 | if isinstance( other, basestring ):
|
---|
| 690 | other = Literal( other )
|
---|
| 691 | if not isinstance( other, ParserElement ):
|
---|
| 692 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 693 | SyntaxWarning, stacklevel=2)
|
---|
| 694 | return other + self
|
---|
| 695 |
|
---|
| 696 | def __or__(self, other ):
|
---|
| 697 | """Implementation of | operator - returns MatchFirst"""
|
---|
| 698 | if isinstance( other, basestring ):
|
---|
| 699 | other = Literal( other )
|
---|
| 700 | if not isinstance( other, ParserElement ):
|
---|
| 701 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 702 | SyntaxWarning, stacklevel=2)
|
---|
| 703 | return MatchFirst( [ self, other ] )
|
---|
| 704 |
|
---|
| 705 | def __ror__(self, other ):
|
---|
| 706 | """Implementation of |= operator"""
|
---|
| 707 | if isinstance( other, basestring ):
|
---|
| 708 | other = Literal( other )
|
---|
| 709 | if not isinstance( other, ParserElement ):
|
---|
| 710 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 711 | SyntaxWarning, stacklevel=2)
|
---|
| 712 | return other | self
|
---|
| 713 |
|
---|
| 714 | def __xor__(self, other ):
|
---|
| 715 | """Implementation of ^ operator - returns Or"""
|
---|
| 716 | if isinstance( other, basestring ):
|
---|
| 717 | other = Literal( other )
|
---|
| 718 | if not isinstance( other, ParserElement ):
|
---|
| 719 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 720 | SyntaxWarning, stacklevel=2)
|
---|
| 721 | return Or( [ self, other ] )
|
---|
| 722 |
|
---|
| 723 | def __rxor__(self, other ):
|
---|
| 724 | """Implementation of ^= operator"""
|
---|
| 725 | if isinstance( other, basestring ):
|
---|
| 726 | other = Literal( other )
|
---|
| 727 | if not isinstance( other, ParserElement ):
|
---|
| 728 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 729 | SyntaxWarning, stacklevel=2)
|
---|
| 730 | return other ^ self
|
---|
| 731 |
|
---|
| 732 | def __and__(self, other ):
|
---|
| 733 | """Implementation of & operator - returns Each"""
|
---|
| 734 | if isinstance( other, basestring ):
|
---|
| 735 | other = Literal( other )
|
---|
| 736 | if not isinstance( other, ParserElement ):
|
---|
| 737 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 738 | SyntaxWarning, stacklevel=2)
|
---|
| 739 | return Each( [ self, other ] )
|
---|
| 740 |
|
---|
| 741 | def __rand__(self, other ):
|
---|
| 742 | """Implementation of right-& operator"""
|
---|
| 743 | if isinstance( other, basestring ):
|
---|
| 744 | other = Literal( other )
|
---|
| 745 | if not isinstance( other, ParserElement ):
|
---|
| 746 | warnings.warn("Cannot add element of type %s to ParserElement" % type(other),
|
---|
| 747 | SyntaxWarning, stacklevel=2)
|
---|
| 748 | return other & self
|
---|
| 749 |
|
---|
| 750 | def __invert__( self ):
|
---|
| 751 | """Implementation of ~ operator - returns NotAny"""
|
---|
| 752 | return NotAny( self )
|
---|
| 753 |
|
---|
| 754 | def suppress( self ):
|
---|
| 755 | """Suppresses the output of this ParserElement; useful to keep punctuation from
|
---|
| 756 | cluttering up returned output.
|
---|
| 757 | """
|
---|
| 758 | return Suppress( self )
|
---|
| 759 |
|
---|
| 760 | def leaveWhitespace( self ):
|
---|
| 761 | """Disables the skipping of whitespace before matching the characters in the
|
---|
| 762 | ParserElement's defined pattern. This is normally only used internally by
|
---|
| 763 | the pyparsing module, but may be needed in some whitespace-sensitive grammars.
|
---|
| 764 | """
|
---|
| 765 | self.skipWhitespace = False
|
---|
| 766 | return self
|
---|
| 767 |
|
---|
| 768 | def setWhitespaceChars( self, chars ):
|
---|
| 769 | """Overrides the default whitespace chars
|
---|
| 770 | """
|
---|
| 771 | self.skipWhitespace = True
|
---|
| 772 | self.whiteChars = chars
|
---|
| 773 |
|
---|
| 774 | def parseWithTabs( self ):
|
---|
| 775 | """Overrides default behavior to expand <TAB>s to spaces before parsing the input string.
|
---|
| 776 | Must be called before parseString when the input grammar contains elements that
|
---|
| 777 | match <TAB> characters."""
|
---|
| 778 | self.keepTabs = True
|
---|
| 779 | return self
|
---|
| 780 |
|
---|
| 781 | def ignore( self, other ):
|
---|
| 782 | """Define expression to be ignored (e.g., comments) while doing pattern
|
---|
| 783 | matching; may be called repeatedly, to define multiple comment or other
|
---|
| 784 | ignorable patterns.
|
---|
| 785 | """
|
---|
| 786 | if isinstance( other, Suppress ):
|
---|
| 787 | if other not in self.ignoreExprs:
|
---|
| 788 | self.ignoreExprs.append( other )
|
---|
| 789 | else:
|
---|
| 790 | self.ignoreExprs.append( Suppress( other ) )
|
---|
| 791 | return self
|
---|
| 792 |
|
---|
| 793 | def setDebugActions( self, startAction, successAction, exceptionAction ):
|
---|
| 794 | """Enable display of debugging messages while doing pattern matching."""
|
---|
| 795 | self.debugActions = (startAction or _defaultStartDebugAction,
|
---|
| 796 | successAction or _defaultSuccessDebugAction,
|
---|
| 797 | exceptionAction or _defaultExceptionDebugAction)
|
---|
| 798 | self.debug = True
|
---|
| 799 | return self
|
---|
| 800 |
|
---|
| 801 | def setDebug( self, flag=True ):
|
---|
| 802 | """Enable display of debugging messages while doing pattern matching."""
|
---|
| 803 | if flag:
|
---|
| 804 | self.setDebugActions( _defaultStartDebugAction, _defaultSuccessDebugAction, _defaultExceptionDebugAction )
|
---|
| 805 | else:
|
---|
| 806 | self.debug = False
|
---|
| 807 | return self
|
---|
| 808 |
|
---|
| 809 | def __str__( self ):
|
---|
| 810 | return self.name
|
---|
| 811 |
|
---|
| 812 | def __repr__( self ):
|
---|
| 813 | return _ustr(self)
|
---|
| 814 |
|
---|
| 815 | def streamline( self ):
|
---|
| 816 | self.streamlined = True
|
---|
| 817 | self.strRepr = None
|
---|
| 818 | return self
|
---|
| 819 |
|
---|
| 820 | def checkRecursion( self, parseElementList ):
|
---|
| 821 | pass
|
---|
| 822 |
|
---|
| 823 | def validate( self, validateTrace=[] ):
|
---|
| 824 | """Check defined expressions for valid structure, check for infinite recursive definitions."""
|
---|
| 825 | self.checkRecursion( [] )
|
---|
| 826 |
|
---|
| 827 | def parseFile( self, file_or_filename ):
|
---|
| 828 | """Execute the parse expression on the given file or filename.
|
---|
| 829 | If a filename is specified (instead of a file object),
|
---|
| 830 | the entire file is opened, read, and closed before parsing.
|
---|
| 831 | """
|
---|
| 832 | try:
|
---|
| 833 | file_contents = file_or_filename.read()
|
---|
| 834 | except AttributeError:
|
---|
| 835 | f = open(file_or_filename, "rb")
|
---|
| 836 | file_contents = f.read()
|
---|
| 837 | f.close()
|
---|
| 838 | return self.parseString(file_contents)
|
---|
| 839 |
|
---|
| 840 |
|
---|
| 841 | class Token(ParserElement):
|
---|
| 842 | """Abstract ParserElement subclass, for defining atomic matching patterns."""
|
---|
| 843 | def __init__( self ):
|
---|
| 844 | super(Token,self).__init__( savelist=False )
|
---|
| 845 | self.myException = ParseException("",0,"",self)
|
---|
| 846 |
|
---|
| 847 | def setName(self, name):
|
---|
| 848 | s = super(Token,self).setName(name)
|
---|
| 849 | self.errmsg = "Expected " + self.name
|
---|
| 850 | s.myException.msg = self.errmsg
|
---|
| 851 | return s
|
---|
| 852 |
|
---|
| 853 |
|
---|
| 854 | class Empty(Token):
|
---|
| 855 | """An empty token, will always match."""
|
---|
| 856 | def __init__( self ):
|
---|
| 857 | super(Empty,self).__init__()
|
---|
| 858 | self.name = "Empty"
|
---|
| 859 | self.mayReturnEmpty = True
|
---|
| 860 | self.mayIndexError = False
|
---|
| 861 |
|
---|
| 862 |
|
---|
| 863 | class NoMatch(Token):
|
---|
| 864 | """A token that will never match."""
|
---|
| 865 | def __init__( self ):
|
---|
| 866 | super(NoMatch,self).__init__()
|
---|
| 867 | self.name = "NoMatch"
|
---|
| 868 | self.mayReturnEmpty = True
|
---|
| 869 | self.mayIndexError = False
|
---|
| 870 | self.errmsg = "Unmatchable token"
|
---|
| 871 | self.myException.msg = self.errmsg
|
---|
| 872 |
|
---|
| 873 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 874 | exc = self.myException
|
---|
| 875 | exc.loc = loc
|
---|
| 876 | exc.pstr = instring
|
---|
| 877 | raise exc
|
---|
| 878 |
|
---|
| 879 |
|
---|
| 880 | class Literal(Token):
|
---|
| 881 | """Token to exactly match a specified string."""
|
---|
| 882 | def __init__( self, matchString ):
|
---|
| 883 | super(Literal,self).__init__()
|
---|
| 884 | self.match = matchString
|
---|
| 885 | self.matchLen = len(matchString)
|
---|
| 886 | try:
|
---|
| 887 | self.firstMatchChar = matchString[0]
|
---|
| 888 | except IndexError:
|
---|
| 889 | warnings.warn("null string passed to Literal; use Empty() instead",
|
---|
| 890 | SyntaxWarning, stacklevel=2)
|
---|
| 891 | self.__class__ = Empty
|
---|
| 892 | self.name = '"%s"' % self.match
|
---|
| 893 | self.errmsg = "Expected " + self.name
|
---|
| 894 | self.mayReturnEmpty = False
|
---|
| 895 | self.myException.msg = self.errmsg
|
---|
| 896 | self.mayIndexError = False
|
---|
| 897 |
|
---|
| 898 | # Performance tuning: this routine gets called a *lot*
|
---|
| 899 | # if this is a single character match string and the first character matches,
|
---|
| 900 | # short-circuit as quickly as possible, and avoid calling startswith
|
---|
| 901 | #~ @profile
|
---|
| 902 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 903 | if (instring[loc] == self.firstMatchChar and
|
---|
| 904 | (self.matchLen==1 or instring.startswith(self.match,loc)) ):
|
---|
| 905 | return loc+self.matchLen, self.match
|
---|
| 906 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 907 | exc = self.myException
|
---|
| 908 | exc.loc = loc
|
---|
| 909 | exc.pstr = instring
|
---|
| 910 | raise exc
|
---|
| 911 |
|
---|
| 912 | class Keyword(Token):
|
---|
| 913 | """Token to exactly match a specified string as a keyword, that is, it must be
|
---|
| 914 | immediately followed by a non-keyword character. Compare with Literal::
|
---|
| 915 | Literal("if") will match the leading 'if' in 'ifAndOnlyIf'.
|
---|
| 916 | Keyword("if") will not; it will only match the leading 'if in 'if x=1', or 'if(y==2)'
|
---|
| 917 | Accepts two optional constructor arguments in addition to the keyword string:
|
---|
| 918 | identChars is a string of characters that would be valid identifier characters,
|
---|
| 919 | defaulting to all alphanumerics + "_" and "$"; caseless allows case-insensitive
|
---|
| 920 | matching, default is False.
|
---|
| 921 | """
|
---|
| 922 | DEFAULT_KEYWORD_CHARS = alphanums+"_$"
|
---|
| 923 |
|
---|
| 924 | def __init__( self, matchString, identChars=DEFAULT_KEYWORD_CHARS, caseless=False ):
|
---|
| 925 | super(Keyword,self).__init__()
|
---|
| 926 | self.match = matchString
|
---|
| 927 | self.matchLen = len(matchString)
|
---|
| 928 | try:
|
---|
| 929 | self.firstMatchChar = matchString[0]
|
---|
| 930 | except IndexError:
|
---|
| 931 | warnings.warn("null string passed to Keyword; use Empty() instead",
|
---|
| 932 | SyntaxWarning, stacklevel=2)
|
---|
| 933 | self.name = '"%s"' % self.match
|
---|
| 934 | self.errmsg = "Expected " + self.name
|
---|
| 935 | self.mayReturnEmpty = False
|
---|
| 936 | self.myException.msg = self.errmsg
|
---|
| 937 | self.mayIndexError = False
|
---|
| 938 | self.caseless = caseless
|
---|
| 939 | if caseless:
|
---|
| 940 | self.caselessmatch = matchString.upper()
|
---|
| 941 | identChars = identChars.upper()
|
---|
| 942 | self.identChars = _str2dict(identChars)
|
---|
| 943 |
|
---|
| 944 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 945 | if self.caseless:
|
---|
| 946 | if ( (instring[ loc:loc+self.matchLen ].upper() == self.caselessmatch) and
|
---|
| 947 | (loc >= len(instring)-self.matchLen or instring[loc+self.matchLen].upper() not in self.identChars) and
|
---|
| 948 | (loc == 0 or instring[loc-1].upper() not in self.identChars) ):
|
---|
| 949 | return loc+self.matchLen, self.match
|
---|
| 950 | else:
|
---|
| 951 | if (instring[loc] == self.firstMatchChar and
|
---|
| 952 | (self.matchLen==1 or instring.startswith(self.match,loc)) and
|
---|
| 953 | (loc >= len(instring)-self.matchLen or instring[loc+self.matchLen] not in self.identChars) and
|
---|
| 954 | (loc == 0 or instring[loc-1] not in self.identChars) ):
|
---|
| 955 | return loc+self.matchLen, self.match
|
---|
| 956 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 957 | exc = self.myException
|
---|
| 958 | exc.loc = loc
|
---|
| 959 | exc.pstr = instring
|
---|
| 960 | raise exc
|
---|
| 961 |
|
---|
| 962 | def copy(self):
|
---|
| 963 | c = super(Keyword,self).copy()
|
---|
| 964 | c.identChars = Keyword.DEFAULT_KEYWORD_CHARS
|
---|
| 965 | return c
|
---|
| 966 |
|
---|
| 967 | def setDefaultKeywordChars( chars ):
|
---|
| 968 | """Overrides the default Keyword chars
|
---|
| 969 | """
|
---|
| 970 | Keyword.DEFAULT_KEYWORD_CHARS = chars
|
---|
| 971 | setDefaultKeywordChars = staticmethod(setDefaultKeywordChars)
|
---|
| 972 |
|
---|
| 973 |
|
---|
| 974 | class CaselessLiteral(Literal):
|
---|
| 975 | """Token to match a specified string, ignoring case of letters.
|
---|
| 976 | Note: the matched results will always be in the case of the given
|
---|
| 977 | match string, NOT the case of the input text.
|
---|
| 978 | """
|
---|
| 979 | def __init__( self, matchString ):
|
---|
| 980 | super(CaselessLiteral,self).__init__( matchString.upper() )
|
---|
| 981 | # Preserve the defining literal.
|
---|
| 982 | self.returnString = matchString
|
---|
| 983 | self.name = "'%s'" % self.returnString
|
---|
| 984 | self.errmsg = "Expected " + self.name
|
---|
| 985 | self.myException.msg = self.errmsg
|
---|
| 986 |
|
---|
| 987 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 988 | if instring[ loc:loc+self.matchLen ].upper() == self.match:
|
---|
| 989 | return loc+self.matchLen, self.returnString
|
---|
| 990 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 991 | exc = self.myException
|
---|
| 992 | exc.loc = loc
|
---|
| 993 | exc.pstr = instring
|
---|
| 994 | raise exc
|
---|
| 995 |
|
---|
| 996 | class CaselessKeyword(Keyword):
|
---|
| 997 | def __init__( self, matchString, identChars=Keyword.DEFAULT_KEYWORD_CHARS ):
|
---|
| 998 | super(CaselessKeyword,self).__init__( matchString, identCars, caseless=True )
|
---|
| 999 |
|
---|
| 1000 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1001 | if ( (instring[ loc:loc+self.matchLen ].upper() == self.caselessmatch) and
|
---|
| 1002 | (loc >= len(instring)-self.matchLen or instring[loc+self.matchLen].upper() not in self.identChars) ):
|
---|
| 1003 | return loc+self.matchLen, self.match
|
---|
| 1004 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1005 | exc = self.myException
|
---|
| 1006 | exc.loc = loc
|
---|
| 1007 | exc.pstr = instring
|
---|
| 1008 | raise exc
|
---|
| 1009 |
|
---|
| 1010 | class Word(Token):
|
---|
| 1011 | """Token for matching words composed of allowed character sets.
|
---|
| 1012 | Defined with string containing all allowed initial characters,
|
---|
| 1013 | an optional string containing allowed body characters (if omitted,
|
---|
| 1014 | defaults to the initial character set), and an optional minimum,
|
---|
| 1015 | maximum, and/or exact length.
|
---|
| 1016 | """
|
---|
| 1017 | def __init__( self, initChars, bodyChars=None, min=1, max=0, exact=0 ):
|
---|
| 1018 | super(Word,self).__init__()
|
---|
| 1019 | self.initCharsOrig = initChars
|
---|
| 1020 | self.initChars = _str2dict(initChars)
|
---|
| 1021 | if bodyChars :
|
---|
| 1022 | self.bodyCharsOrig = bodyChars
|
---|
| 1023 | self.bodyChars = _str2dict(bodyChars)
|
---|
| 1024 | else:
|
---|
| 1025 | self.bodyCharsOrig = initChars
|
---|
| 1026 | self.bodyChars = _str2dict(initChars)
|
---|
| 1027 |
|
---|
| 1028 | self.maxSpecified = max > 0
|
---|
| 1029 |
|
---|
| 1030 | self.minLen = min
|
---|
| 1031 |
|
---|
| 1032 | if max > 0:
|
---|
| 1033 | self.maxLen = max
|
---|
| 1034 | else:
|
---|
| 1035 | self.maxLen = sys.maxint
|
---|
| 1036 |
|
---|
| 1037 | if exact > 0:
|
---|
| 1038 | self.maxLen = exact
|
---|
| 1039 | self.minLen = exact
|
---|
| 1040 |
|
---|
| 1041 | self.name = _ustr(self)
|
---|
| 1042 | self.errmsg = "Expected " + self.name
|
---|
| 1043 | self.myException.msg = self.errmsg
|
---|
| 1044 | self.mayIndexError = False
|
---|
| 1045 |
|
---|
| 1046 | if ' ' not in self.initCharsOrig+self.bodyCharsOrig and (min==1 and max==0 and exact==0):
|
---|
| 1047 | if self.bodyCharsOrig == self.initCharsOrig:
|
---|
| 1048 | self.reString = "[%s]+" % _escapeRegexRangeChars(self.initCharsOrig)
|
---|
| 1049 | elif len(self.bodyCharsOrig) == 1:
|
---|
| 1050 | self.reString = "%s[%s]*" % \
|
---|
| 1051 | (re.escape(self.initCharsOrig),
|
---|
| 1052 | _escapeRegexRangeChars(self.bodyCharsOrig),)
|
---|
| 1053 | else:
|
---|
| 1054 | self.reString = "[%s][%s]*" % \
|
---|
| 1055 | (_escapeRegexRangeChars(self.initCharsOrig),
|
---|
| 1056 | _escapeRegexRangeChars(self.bodyCharsOrig),)
|
---|
| 1057 | try:
|
---|
| 1058 | self.re = re.compile( self.reString )
|
---|
| 1059 | except:
|
---|
| 1060 | self.re = None
|
---|
| 1061 |
|
---|
| 1062 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1063 | if self.re:
|
---|
| 1064 | result = self.re.match(instring,loc)
|
---|
| 1065 | if not result:
|
---|
| 1066 | exc = self.myException
|
---|
| 1067 | exc.loc = loc
|
---|
| 1068 | exc.pstr = instring
|
---|
| 1069 | raise exc
|
---|
| 1070 |
|
---|
| 1071 | loc = result.end()
|
---|
| 1072 | return loc,result.group()
|
---|
| 1073 |
|
---|
| 1074 | if not(instring[ loc ] in self.initChars):
|
---|
| 1075 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1076 | exc = self.myException
|
---|
| 1077 | exc.loc = loc
|
---|
| 1078 | exc.pstr = instring
|
---|
| 1079 | raise exc
|
---|
| 1080 | start = loc
|
---|
| 1081 | loc += 1
|
---|
| 1082 | instrlen = len(instring)
|
---|
| 1083 | bodychars = self.bodyChars
|
---|
| 1084 | maxloc = start + self.maxLen
|
---|
| 1085 | maxloc = min( maxloc, instrlen )
|
---|
| 1086 | while loc < maxloc and instring[loc] in bodychars:
|
---|
| 1087 | loc += 1
|
---|
| 1088 |
|
---|
| 1089 | throwException = False
|
---|
| 1090 | if loc - start < self.minLen:
|
---|
| 1091 | throwException = True
|
---|
| 1092 | if self.maxSpecified and loc < instrlen and instring[loc] in bodychars:
|
---|
| 1093 | throwException = True
|
---|
| 1094 |
|
---|
| 1095 | if throwException:
|
---|
| 1096 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1097 | exc = self.myException
|
---|
| 1098 | exc.loc = loc
|
---|
| 1099 | exc.pstr = instring
|
---|
| 1100 | raise exc
|
---|
| 1101 |
|
---|
| 1102 | return loc, instring[start:loc]
|
---|
| 1103 |
|
---|
| 1104 | def __str__( self ):
|
---|
| 1105 | try:
|
---|
| 1106 | return super(Word,self).__str__()
|
---|
| 1107 | except:
|
---|
| 1108 | pass
|
---|
| 1109 |
|
---|
| 1110 |
|
---|
| 1111 | if self.strRepr is None:
|
---|
| 1112 |
|
---|
| 1113 | def charsAsStr(s):
|
---|
| 1114 | if len(s)>4:
|
---|
| 1115 | return s[:4]+"..."
|
---|
| 1116 | else:
|
---|
| 1117 | return s
|
---|
| 1118 |
|
---|
| 1119 | if ( self.initCharsOrig != self.bodyCharsOrig ):
|
---|
| 1120 | self.strRepr = "W:(%s,%s)" % ( charsAsStr(self.initCharsOrig), charsAsStr(self.bodyCharsOrig) )
|
---|
| 1121 | else:
|
---|
| 1122 | self.strRepr = "W:(%s)" % charsAsStr(self.initCharsOrig)
|
---|
| 1123 |
|
---|
| 1124 | return self.strRepr
|
---|
| 1125 |
|
---|
| 1126 |
|
---|
| 1127 | class Regex(Token):
|
---|
| 1128 | """Token for matching strings that match a given regular expression.
|
---|
| 1129 | Defined with string specifying the regular expression in a form recognized by the inbuilt Python re module.
|
---|
| 1130 | """
|
---|
| 1131 | def __init__( self, pattern, flags=0):
|
---|
| 1132 | """The parameters pattern and flags are passed to the re.compile() function as-is. See the Python re module for an explanation of the acceptable patterns and flags."""
|
---|
| 1133 | super(Regex,self).__init__()
|
---|
| 1134 |
|
---|
| 1135 | if len(pattern) == 0:
|
---|
| 1136 | warnings.warn("null string passed to Regex; use Empty() instead",
|
---|
| 1137 | SyntaxWarning, stacklevel=2)
|
---|
| 1138 |
|
---|
| 1139 | self.pattern = pattern
|
---|
| 1140 | self.flags = flags
|
---|
| 1141 |
|
---|
| 1142 | try:
|
---|
| 1143 | self.re = re.compile(self.pattern, self.flags)
|
---|
| 1144 | self.reString = self.pattern
|
---|
| 1145 | except Exception,e:
|
---|
| 1146 | warnings.warn("invalid pattern (%s) passed to Regex" % pattern,
|
---|
| 1147 | SyntaxWarning, stacklevel=2)
|
---|
| 1148 | raise
|
---|
| 1149 |
|
---|
| 1150 | self.name = _ustr(self)
|
---|
| 1151 | self.errmsg = "Expected " + self.name
|
---|
| 1152 | self.myException.msg = self.errmsg
|
---|
| 1153 | self.mayIndexError = False
|
---|
| 1154 | self.mayReturnEmpty = True
|
---|
| 1155 |
|
---|
| 1156 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1157 | result = self.re.match(instring,loc)
|
---|
| 1158 | if not result:
|
---|
| 1159 | exc = self.myException
|
---|
| 1160 | exc.loc = loc
|
---|
| 1161 | exc.pstr = instring
|
---|
| 1162 | raise exc
|
---|
| 1163 |
|
---|
| 1164 | loc = result.end()
|
---|
| 1165 | d = result.groupdict()
|
---|
| 1166 | ret = ParseResults(result.group())
|
---|
| 1167 | if d:
|
---|
| 1168 | for k in d.keys():
|
---|
| 1169 | ret[k] = d[k]
|
---|
| 1170 | return loc,ret
|
---|
| 1171 |
|
---|
| 1172 | def __str__( self ):
|
---|
| 1173 | try:
|
---|
| 1174 | return super(Regex,self).__str__()
|
---|
| 1175 | except:
|
---|
| 1176 | pass
|
---|
| 1177 |
|
---|
| 1178 | if self.strRepr is None:
|
---|
| 1179 | self.strRepr = "Re:(%s)" % repr(self.pattern)
|
---|
| 1180 |
|
---|
| 1181 | return self.strRepr
|
---|
| 1182 |
|
---|
| 1183 |
|
---|
| 1184 | class QuotedString(Token):
|
---|
| 1185 | """Token for matching strings that are delimited by quoting characters.
|
---|
| 1186 | """
|
---|
| 1187 | def __init__( self, quoteChar, escChar=None, escQuote=None, multiline=False, unquoteResults=True):
|
---|
| 1188 | """
|
---|
| 1189 | Defined with the following parameters:
|
---|
| 1190 | - quoteCharacter - string of one or more characters defining the quote delimiting string
|
---|
| 1191 | - escapeCharacter - character to escape quotes, typically backslash (default=None)
|
---|
| 1192 | - escapedQuote - special quote sequence to escape an embedded quote string (such as SQL's "" to escape and embedded ") (default=None)
|
---|
| 1193 | - multiline - boolean indicating whether quotes can span multiple lines (default=False)
|
---|
| 1194 | - unquoteResults - boolean indicating whether the matched text should be unquoted (default=True)
|
---|
| 1195 | """
|
---|
| 1196 | super(QuotedString,self).__init__()
|
---|
| 1197 |
|
---|
| 1198 | # remove white space from quote char - wont work anyway
|
---|
| 1199 | quoteChar = quoteChar.strip()
|
---|
| 1200 | if len(quoteChar) == 0:
|
---|
| 1201 | warnings.warn("quoteChar cannot be the empty string",SyntaxWarning,stacklevel=2)
|
---|
| 1202 | raise SyntaxError()
|
---|
| 1203 |
|
---|
| 1204 | self.quoteChar = quoteChar
|
---|
| 1205 | self.quoteCharLen = len(quoteChar)
|
---|
| 1206 | self.firstQuoteChar = quoteChar[0]
|
---|
| 1207 | self.escChar = escChar
|
---|
| 1208 | self.escQuote = escQuote
|
---|
| 1209 | self.unquoteResults = unquoteResults
|
---|
| 1210 |
|
---|
| 1211 | if multiline:
|
---|
| 1212 | self.flags = re.MULTILINE | re.DOTALL
|
---|
| 1213 | self.pattern = r'%s([^%s%s]' % \
|
---|
| 1214 | ( re.escape(self.quoteChar),
|
---|
| 1215 | _escapeRegexRangeChars(self.quoteChar[0]),
|
---|
| 1216 | (escChar is not None and _escapeRegexRangeChars(escChar) or '') )
|
---|
| 1217 | else:
|
---|
| 1218 | self.flags = 0
|
---|
| 1219 | self.pattern = r'%s([^%s\n\r%s]' % \
|
---|
| 1220 | ( re.escape(self.quoteChar),
|
---|
| 1221 | _escapeRegexRangeChars(self.quoteChar[0]),
|
---|
| 1222 | (escChar is not None and _escapeRegexRangeChars(escChar) or '') )
|
---|
| 1223 | if len(self.quoteChar) > 1:
|
---|
| 1224 | self.pattern += (
|
---|
| 1225 | '|(' + ')|('.join(["%s[^%s]" % (re.escape(self.quoteChar[:i]),
|
---|
| 1226 | _escapeRegexRangeChars(self.quoteChar[i]))
|
---|
| 1227 | for i in range(len(self.quoteChar)-1,0,-1)]) + ')'
|
---|
| 1228 | )
|
---|
| 1229 | if escQuote:
|
---|
| 1230 | self.pattern += (r'|(%s)' % re.escape(escQuote))
|
---|
| 1231 | if escChar:
|
---|
| 1232 | self.pattern += (r'|(%s.)' % re.escape(escChar))
|
---|
| 1233 | self.escCharReplacePattern = re.escape(self.escChar)+"(.)"
|
---|
| 1234 | self.pattern += (r')*%s' % re.escape(self.quoteChar))
|
---|
| 1235 |
|
---|
| 1236 | try:
|
---|
| 1237 | self.re = re.compile(self.pattern, self.flags)
|
---|
| 1238 | self.reString = self.pattern
|
---|
| 1239 | except Exception,e:
|
---|
| 1240 | warnings.warn("invalid pattern (%s) passed to Regex" % self.pattern,
|
---|
| 1241 | SyntaxWarning, stacklevel=2)
|
---|
| 1242 | raise
|
---|
| 1243 |
|
---|
| 1244 | self.name = _ustr(self)
|
---|
| 1245 | self.errmsg = "Expected " + self.name
|
---|
| 1246 | self.myException.msg = self.errmsg
|
---|
| 1247 | self.mayIndexError = False
|
---|
| 1248 | self.mayReturnEmpty = True
|
---|
| 1249 |
|
---|
| 1250 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1251 | result = instring[loc] == self.firstQuoteChar and self.re.match(instring,loc) or None
|
---|
| 1252 | if not result:
|
---|
| 1253 | exc = self.myException
|
---|
| 1254 | exc.loc = loc
|
---|
| 1255 | exc.pstr = instring
|
---|
| 1256 | raise exc
|
---|
| 1257 |
|
---|
| 1258 | loc = result.end()
|
---|
| 1259 | ret = result.group()
|
---|
| 1260 |
|
---|
| 1261 | if self.unquoteResults:
|
---|
| 1262 |
|
---|
| 1263 | # strip off quotes
|
---|
| 1264 | ret = ret[self.quoteCharLen:-self.quoteCharLen]
|
---|
| 1265 |
|
---|
| 1266 | if isinstance(ret,basestring):
|
---|
| 1267 | # replace escaped characters
|
---|
| 1268 | if self.escChar:
|
---|
| 1269 | ret = re.sub(self.escCharReplacePattern,"\g<1>",ret)
|
---|
| 1270 |
|
---|
| 1271 | # replace escaped quotes
|
---|
| 1272 | if self.escQuote:
|
---|
| 1273 | ret = ret.replace(self.escQuote, self.quoteChar)
|
---|
| 1274 |
|
---|
| 1275 | return loc, ret
|
---|
| 1276 |
|
---|
| 1277 | def __str__( self ):
|
---|
| 1278 | try:
|
---|
| 1279 | return super(QuotedString,self).__str__()
|
---|
| 1280 | except:
|
---|
| 1281 | pass
|
---|
| 1282 |
|
---|
| 1283 | if self.strRepr is None:
|
---|
| 1284 | self.strRepr = "quoted string, delimited by %s characters" % self.quoteChar
|
---|
| 1285 |
|
---|
| 1286 | return self.strRepr
|
---|
| 1287 |
|
---|
| 1288 |
|
---|
| 1289 | class CharsNotIn(Token):
|
---|
| 1290 | """Token for matching words composed of characters *not* in a given set.
|
---|
| 1291 | Defined with string containing all disallowed characters, and an optional
|
---|
| 1292 | minimum, maximum, and/or exact length.
|
---|
| 1293 | """
|
---|
| 1294 | def __init__( self, notChars, min=1, max=0, exact=0 ):
|
---|
| 1295 | super(CharsNotIn,self).__init__()
|
---|
| 1296 | self.skipWhitespace = False
|
---|
| 1297 | self.notChars = notChars
|
---|
| 1298 |
|
---|
| 1299 | self.minLen = min
|
---|
| 1300 |
|
---|
| 1301 | if max > 0:
|
---|
| 1302 | self.maxLen = max
|
---|
| 1303 | else:
|
---|
| 1304 | self.maxLen = sys.maxint
|
---|
| 1305 |
|
---|
| 1306 | if exact > 0:
|
---|
| 1307 | self.maxLen = exact
|
---|
| 1308 | self.minLen = exact
|
---|
| 1309 |
|
---|
| 1310 | self.name = _ustr(self)
|
---|
| 1311 | self.errmsg = "Expected " + self.name
|
---|
| 1312 | self.mayReturnEmpty = ( self.minLen == 0 )
|
---|
| 1313 | self.myException.msg = self.errmsg
|
---|
| 1314 | self.mayIndexError = False
|
---|
| 1315 |
|
---|
| 1316 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1317 | if instring[loc] in self.notChars:
|
---|
| 1318 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1319 | exc = self.myException
|
---|
| 1320 | exc.loc = loc
|
---|
| 1321 | exc.pstr = instring
|
---|
| 1322 | raise exc
|
---|
| 1323 |
|
---|
| 1324 | start = loc
|
---|
| 1325 | loc += 1
|
---|
| 1326 | notchars = self.notChars
|
---|
| 1327 | maxlen = min( start+self.maxLen, len(instring) )
|
---|
| 1328 | while loc < maxlen and \
|
---|
| 1329 | (instring[loc] not in notchars):
|
---|
| 1330 | loc += 1
|
---|
| 1331 |
|
---|
| 1332 | if loc - start < self.minLen:
|
---|
| 1333 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1334 | exc = self.myException
|
---|
| 1335 | exc.loc = loc
|
---|
| 1336 | exc.pstr = instring
|
---|
| 1337 | raise exc
|
---|
| 1338 |
|
---|
| 1339 | return loc, instring[start:loc]
|
---|
| 1340 |
|
---|
| 1341 | def __str__( self ):
|
---|
| 1342 | try:
|
---|
| 1343 | return super(CharsNotIn, self).__str__()
|
---|
| 1344 | except:
|
---|
| 1345 | pass
|
---|
| 1346 |
|
---|
| 1347 | if self.strRepr is None:
|
---|
| 1348 | if len(self.notChars) > 4:
|
---|
| 1349 | self.strRepr = "!W:(%s...)" % self.notChars[:4]
|
---|
| 1350 | else:
|
---|
| 1351 | self.strRepr = "!W:(%s)" % self.notChars
|
---|
| 1352 |
|
---|
| 1353 | return self.strRepr
|
---|
| 1354 |
|
---|
| 1355 | class White(Token):
|
---|
| 1356 | """Special matching class for matching whitespace. Normally, whitespace is ignored
|
---|
| 1357 | by pyparsing grammars. This class is included when some whitespace structures
|
---|
| 1358 | are significant. Define with a string containing the whitespace characters to be
|
---|
| 1359 | matched; default is " \\t\\n". Also takes optional min, max, and exact arguments,
|
---|
| 1360 | as defined for the Word class."""
|
---|
| 1361 | whiteStrs = {
|
---|
| 1362 | " " : "<SPC>",
|
---|
| 1363 | "\t": "<TAB>",
|
---|
| 1364 | "\n": "<LF>",
|
---|
| 1365 | "\r": "<CR>",
|
---|
| 1366 | "\f": "<FF>",
|
---|
| 1367 | }
|
---|
| 1368 | def __init__(self, ws=" \t\r\n", min=1, max=0, exact=0):
|
---|
| 1369 | super(White,self).__init__()
|
---|
| 1370 | self.matchWhite = ws
|
---|
| 1371 | self.whiteChars = "".join([c for c in self.whiteChars if c not in self.matchWhite])
|
---|
| 1372 | #~ self.leaveWhitespace()
|
---|
| 1373 | self.name = ("".join([White.whiteStrs[c] for c in self.matchWhite]))
|
---|
| 1374 | self.mayReturnEmpty = True
|
---|
| 1375 | self.errmsg = "Expected " + self.name
|
---|
| 1376 | self.myException.msg = self.errmsg
|
---|
| 1377 |
|
---|
| 1378 | self.minLen = min
|
---|
| 1379 |
|
---|
| 1380 | if max > 0:
|
---|
| 1381 | self.maxLen = max
|
---|
| 1382 | else:
|
---|
| 1383 | self.maxLen = sys.maxint
|
---|
| 1384 |
|
---|
| 1385 | if exact > 0:
|
---|
| 1386 | self.maxLen = exact
|
---|
| 1387 | self.minLen = exact
|
---|
| 1388 |
|
---|
| 1389 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1390 | if not(instring[ loc ] in self.matchWhite):
|
---|
| 1391 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1392 | exc = self.myException
|
---|
| 1393 | exc.loc = loc
|
---|
| 1394 | exc.pstr = instring
|
---|
| 1395 | raise exc
|
---|
| 1396 | start = loc
|
---|
| 1397 | loc += 1
|
---|
| 1398 | maxloc = start + self.maxLen
|
---|
| 1399 | maxloc = min( maxloc, len(instring) )
|
---|
| 1400 | while loc < maxloc and instring[loc] in self.matchWhite:
|
---|
| 1401 | loc += 1
|
---|
| 1402 |
|
---|
| 1403 | if loc - start < self.minLen:
|
---|
| 1404 | #~ raise ParseException, ( instring, loc, self.errmsg )
|
---|
| 1405 | exc = self.myException
|
---|
| 1406 | exc.loc = loc
|
---|
| 1407 | exc.pstr = instring
|
---|
| 1408 | raise exc
|
---|
| 1409 |
|
---|
| 1410 | return loc, instring[start:loc]
|
---|
| 1411 |
|
---|
| 1412 |
|
---|
| 1413 | class PositionToken(Token):
|
---|
| 1414 | def __init__( self ):
|
---|
| 1415 | super(PositionToken,self).__init__()
|
---|
| 1416 | self.name=self.__class__.__name__
|
---|
| 1417 | self.mayReturnEmpty = True
|
---|
| 1418 |
|
---|
| 1419 | class GoToColumn(PositionToken):
|
---|
| 1420 | """Token to advance to a specific column of input text; useful for tabular report scraping."""
|
---|
| 1421 | def __init__( self, colno ):
|
---|
| 1422 | super(GoToColumn,self).__init__()
|
---|
| 1423 | self.col = colno
|
---|
| 1424 |
|
---|
| 1425 | def preParse( self, instring, loc ):
|
---|
| 1426 | if col(loc,instring) != self.col:
|
---|
| 1427 | instrlen = len(instring)
|
---|
| 1428 | if self.ignoreExprs:
|
---|
| 1429 | loc = self.skipIgnorables( instring, loc )
|
---|
| 1430 | while loc < instrlen and instring[loc].isspace() and col( loc, instring ) != self.col :
|
---|
| 1431 | loc += 1
|
---|
| 1432 | return loc
|
---|
| 1433 |
|
---|
| 1434 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1435 | thiscol = col( loc, instring )
|
---|
| 1436 | if thiscol > self.col:
|
---|
| 1437 | raise ParseException, ( instring, loc, "Text not in expected column", self )
|
---|
| 1438 | newloc = loc + self.col - thiscol
|
---|
| 1439 | ret = instring[ loc: newloc ]
|
---|
| 1440 | return newloc, ret
|
---|
| 1441 |
|
---|
| 1442 | class LineStart(PositionToken):
|
---|
| 1443 | """Matches if current position is at the beginning of a line within the parse string"""
|
---|
| 1444 | def __init__( self ):
|
---|
| 1445 | super(LineStart,self).__init__()
|
---|
| 1446 | self.whiteChars = " \t"
|
---|
| 1447 | self.errmsg = "Expected start of line"
|
---|
| 1448 | self.myException.msg = self.errmsg
|
---|
| 1449 |
|
---|
| 1450 | def preParse( self, instring, loc ):
|
---|
| 1451 | loc = super(LineStart,self).preParse(instring,loc)
|
---|
| 1452 | if instring[loc] == "\n":
|
---|
| 1453 | loc += 1
|
---|
| 1454 | return loc
|
---|
| 1455 |
|
---|
| 1456 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1457 | if not( loc==0 or ( loc<len(instring) and instring[loc-1] == "\n" ) ): #col(loc, instring) != 1:
|
---|
| 1458 | #~ raise ParseException, ( instring, loc, "Expected start of line" )
|
---|
| 1459 | exc = self.myException
|
---|
| 1460 | exc.loc = loc
|
---|
| 1461 | exc.pstr = instring
|
---|
| 1462 | raise exc
|
---|
| 1463 | return loc, []
|
---|
| 1464 |
|
---|
| 1465 | class LineEnd(PositionToken):
|
---|
| 1466 | """Matches if current position is at the end of a line within the parse string"""
|
---|
| 1467 | def __init__( self ):
|
---|
| 1468 | super(LineEnd,self).__init__()
|
---|
| 1469 | self.whiteChars = " \t"
|
---|
| 1470 | self.errmsg = "Expected end of line"
|
---|
| 1471 | self.myException.msg = self.errmsg
|
---|
| 1472 |
|
---|
| 1473 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1474 | if loc<len(instring):
|
---|
| 1475 | if instring[loc] == "\n":
|
---|
| 1476 | return loc+1, "\n"
|
---|
| 1477 | else:
|
---|
| 1478 | #~ raise ParseException, ( instring, loc, "Expected end of line" )
|
---|
| 1479 | exc = self.myException
|
---|
| 1480 | exc.loc = loc
|
---|
| 1481 | exc.pstr = instring
|
---|
| 1482 | raise exc
|
---|
| 1483 | else:
|
---|
| 1484 | return loc, []
|
---|
| 1485 |
|
---|
| 1486 | class StringStart(PositionToken):
|
---|
| 1487 | """Matches if current position is at the beginning of the parse string"""
|
---|
| 1488 | def __init__( self ):
|
---|
| 1489 | super(StringStart,self).__init__()
|
---|
| 1490 | self.errmsg = "Expected start of text"
|
---|
| 1491 | self.myException.msg = self.errmsg
|
---|
| 1492 |
|
---|
| 1493 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1494 | if loc != 0:
|
---|
| 1495 | # see if entire string up to here is just whitespace and ignoreables
|
---|
| 1496 | if loc != self.preParse( instring, 0 ):
|
---|
| 1497 | #~ raise ParseException, ( instring, loc, "Expected start of text" )
|
---|
| 1498 | exc = self.myException
|
---|
| 1499 | exc.loc = loc
|
---|
| 1500 | exc.pstr = instring
|
---|
| 1501 | raise exc
|
---|
| 1502 | return loc, []
|
---|
| 1503 |
|
---|
| 1504 | class StringEnd(PositionToken):
|
---|
| 1505 | """Matches if current position is at the end of the parse string"""
|
---|
| 1506 | def __init__( self ):
|
---|
| 1507 | super(StringEnd,self).__init__()
|
---|
| 1508 | self.errmsg = "Expected end of text"
|
---|
| 1509 | self.myException.msg = self.errmsg
|
---|
| 1510 |
|
---|
| 1511 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1512 | if loc < len(instring):
|
---|
| 1513 | #~ raise ParseException, ( instring, loc, "Expected end of text" )
|
---|
| 1514 | exc = self.myException
|
---|
| 1515 | exc.loc = loc
|
---|
| 1516 | exc.pstr = instring
|
---|
| 1517 | raise exc
|
---|
| 1518 | return loc, []
|
---|
| 1519 |
|
---|
| 1520 |
|
---|
| 1521 | class ParseExpression(ParserElement):
|
---|
| 1522 | """Abstract subclass of ParserElement, for combining and post-processing parsed tokens."""
|
---|
| 1523 | def __init__( self, exprs, savelist = False ):
|
---|
| 1524 | super(ParseExpression,self).__init__(savelist)
|
---|
| 1525 | if isinstance( exprs, list ):
|
---|
| 1526 | self.exprs = exprs
|
---|
| 1527 | elif isinstance( exprs, basestring ):
|
---|
| 1528 | self.exprs = [ Literal( exprs ) ]
|
---|
| 1529 | else:
|
---|
| 1530 | self.exprs = [ exprs ]
|
---|
| 1531 |
|
---|
| 1532 | def __getitem__( self, i ):
|
---|
| 1533 | return self.exprs[i]
|
---|
| 1534 |
|
---|
| 1535 | def append( self, other ):
|
---|
| 1536 | self.exprs.append( other )
|
---|
| 1537 | self.strRepr = None
|
---|
| 1538 | return self
|
---|
| 1539 |
|
---|
| 1540 | def leaveWhitespace( self ):
|
---|
| 1541 | """Extends leaveWhitespace defined in base class, and also invokes leaveWhitespace on
|
---|
| 1542 | all contained expressions."""
|
---|
| 1543 | self.skipWhitespace = False
|
---|
| 1544 | self.exprs = [ copy.copy(e) for e in self.exprs ]
|
---|
| 1545 | for e in self.exprs:
|
---|
| 1546 | e.leaveWhitespace()
|
---|
| 1547 | return self
|
---|
| 1548 |
|
---|
| 1549 | def ignore( self, other ):
|
---|
| 1550 | if isinstance( other, Suppress ):
|
---|
| 1551 | if other not in self.ignoreExprs:
|
---|
| 1552 | super( ParseExpression, self).ignore( other )
|
---|
| 1553 | for e in self.exprs:
|
---|
| 1554 | e.ignore( self.ignoreExprs[-1] )
|
---|
| 1555 | else:
|
---|
| 1556 | super( ParseExpression, self).ignore( other )
|
---|
| 1557 | for e in self.exprs:
|
---|
| 1558 | e.ignore( self.ignoreExprs[-1] )
|
---|
| 1559 | return self
|
---|
| 1560 |
|
---|
| 1561 | def __str__( self ):
|
---|
| 1562 | try:
|
---|
| 1563 | return super(ParseExpression,self).__str__()
|
---|
| 1564 | except:
|
---|
| 1565 | pass
|
---|
| 1566 |
|
---|
| 1567 | if self.strRepr is None:
|
---|
| 1568 | self.strRepr = "%s:(%s)" % ( self.__class__.__name__, _ustr(self.exprs) )
|
---|
| 1569 | return self.strRepr
|
---|
| 1570 |
|
---|
| 1571 | def streamline( self ):
|
---|
| 1572 | super(ParseExpression,self).streamline()
|
---|
| 1573 |
|
---|
| 1574 | for e in self.exprs:
|
---|
| 1575 | e.streamline()
|
---|
| 1576 |
|
---|
| 1577 | # collapse nested And's of the form And( And( And( a,b), c), d) to And( a,b,c,d )
|
---|
| 1578 | # but only if there are no parse actions or resultsNames on the nested And's
|
---|
| 1579 | # (likewise for Or's and MatchFirst's)
|
---|
| 1580 | if ( len(self.exprs) == 2 ):
|
---|
| 1581 | other = self.exprs[0]
|
---|
| 1582 | if ( isinstance( other, self.__class__ ) and
|
---|
| 1583 | other.parseAction is None and
|
---|
| 1584 | other.resultsName is None and
|
---|
| 1585 | not other.debug ):
|
---|
| 1586 | self.exprs = other.exprs[:] + [ self.exprs[1] ]
|
---|
| 1587 | self.strRepr = None
|
---|
| 1588 |
|
---|
| 1589 | other = self.exprs[-1]
|
---|
| 1590 | if ( isinstance( other, self.__class__ ) and
|
---|
| 1591 | other.parseAction is None and
|
---|
| 1592 | other.resultsName is None and
|
---|
| 1593 | not other.debug ):
|
---|
| 1594 | self.exprs = self.exprs[:-1] + other.exprs[:]
|
---|
| 1595 | self.strRepr = None
|
---|
| 1596 |
|
---|
| 1597 | return self
|
---|
| 1598 |
|
---|
| 1599 | def setResultsName( self, name, listAllMatches=False ):
|
---|
| 1600 | ret = super(ParseExpression,self).setResultsName(name,listAllMatches)
|
---|
| 1601 | #~ ret.saveAsList = True
|
---|
| 1602 | return ret
|
---|
| 1603 |
|
---|
| 1604 | def validate( self, validateTrace=[] ):
|
---|
| 1605 | tmp = validateTrace[:]+[self]
|
---|
| 1606 | for e in self.exprs:
|
---|
| 1607 | e.validate(tmp)
|
---|
| 1608 | self.checkRecursion( [] )
|
---|
| 1609 |
|
---|
| 1610 |
|
---|
| 1611 | class And(ParseExpression):
|
---|
| 1612 | """Requires all given ParseExpressions to be found in the given order.
|
---|
| 1613 | Expressions may be separated by whitespace.
|
---|
| 1614 | May be constructed using the '+' operator.
|
---|
| 1615 | """
|
---|
| 1616 | def __init__( self, exprs, savelist = True ):
|
---|
| 1617 | super(And,self).__init__(exprs, savelist)
|
---|
| 1618 | self.mayReturnEmpty = True
|
---|
| 1619 | for e in exprs:
|
---|
| 1620 | if not e.mayReturnEmpty:
|
---|
| 1621 | self.mayReturnEmpty = False
|
---|
| 1622 | break
|
---|
| 1623 | self.skipWhitespace = exprs[0].skipWhitespace
|
---|
| 1624 | self.whiteChars = exprs[0].whiteChars
|
---|
| 1625 |
|
---|
| 1626 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1627 | loc, resultlist = self.exprs[0].parse( instring, loc, doActions )
|
---|
| 1628 | for e in self.exprs[1:]:
|
---|
| 1629 | loc, exprtokens = e.parse( instring, loc, doActions )
|
---|
| 1630 | if exprtokens or exprtokens.keys():
|
---|
| 1631 | resultlist += exprtokens
|
---|
| 1632 | return loc, resultlist
|
---|
| 1633 |
|
---|
| 1634 | def __iadd__(self, other ):
|
---|
| 1635 | if isinstance( other, basestring ):
|
---|
| 1636 | other = Literal( other )
|
---|
| 1637 | return self.append( other ) #And( [ self, other ] )
|
---|
| 1638 |
|
---|
| 1639 | def checkRecursion( self, parseElementList ):
|
---|
| 1640 | subRecCheckList = parseElementList[:] + [ self ]
|
---|
| 1641 | for e in self.exprs:
|
---|
| 1642 | e.checkRecursion( subRecCheckList )
|
---|
| 1643 | if not e.mayReturnEmpty:
|
---|
| 1644 | break
|
---|
| 1645 |
|
---|
| 1646 | def __str__( self ):
|
---|
| 1647 | if hasattr(self,"name"):
|
---|
| 1648 | return self.name
|
---|
| 1649 |
|
---|
| 1650 | if self.strRepr is None:
|
---|
| 1651 | self.strRepr = "{" + " ".join( [ _ustr(e) for e in self.exprs ] ) + "}"
|
---|
| 1652 |
|
---|
| 1653 | return self.strRepr
|
---|
| 1654 |
|
---|
| 1655 |
|
---|
| 1656 | class Or(ParseExpression):
|
---|
| 1657 | """Requires that at least one ParseExpression is found.
|
---|
| 1658 | If two expressions match, the expression that matches the longest string will be used.
|
---|
| 1659 | May be constructed using the '^' operator.
|
---|
| 1660 | """
|
---|
| 1661 | def __init__( self, exprs, savelist = False ):
|
---|
| 1662 | super(Or,self).__init__(exprs, savelist)
|
---|
| 1663 | self.mayReturnEmpty = False
|
---|
| 1664 | for e in exprs:
|
---|
| 1665 | if e.mayReturnEmpty:
|
---|
| 1666 | self.mayReturnEmpty = True
|
---|
| 1667 | break
|
---|
| 1668 |
|
---|
| 1669 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1670 | maxExcLoc = -1
|
---|
| 1671 | maxMatchLoc = -1
|
---|
| 1672 | for e in self.exprs:
|
---|
| 1673 | try:
|
---|
| 1674 | loc2 = e.tryParse( instring, loc )
|
---|
| 1675 | except ParseException, err:
|
---|
| 1676 | if err.loc > maxExcLoc:
|
---|
| 1677 | maxException = err
|
---|
| 1678 | maxExcLoc = err.loc
|
---|
| 1679 | except IndexError, err:
|
---|
| 1680 | if len(instring) > maxExcLoc:
|
---|
| 1681 | maxException = ParseException(instring,len(instring),e.errmsg,self)
|
---|
| 1682 | maxExcLoc = len(instring)
|
---|
| 1683 | else:
|
---|
| 1684 | if loc2 > maxMatchLoc:
|
---|
| 1685 | maxMatchLoc = loc2
|
---|
| 1686 | maxMatchExp = e
|
---|
| 1687 |
|
---|
| 1688 | if maxMatchLoc < 0:
|
---|
| 1689 | if self.exprs:
|
---|
| 1690 | raise maxException
|
---|
| 1691 | else:
|
---|
| 1692 | raise ParseException(instring, loc, "no defined alternatives to match", self)
|
---|
| 1693 |
|
---|
| 1694 | return maxMatchExp.parse( instring, loc, doActions )
|
---|
| 1695 |
|
---|
| 1696 | def __ixor__(self, other ):
|
---|
| 1697 | if isinstance( other, basestring ):
|
---|
| 1698 | other = Literal( other )
|
---|
| 1699 | return self.append( other ) #Or( [ self, other ] )
|
---|
| 1700 |
|
---|
| 1701 | def __str__( self ):
|
---|
| 1702 | if hasattr(self,"name"):
|
---|
| 1703 | return self.name
|
---|
| 1704 |
|
---|
| 1705 | if self.strRepr is None:
|
---|
| 1706 | self.strRepr = "{" + " ^ ".join( [ _ustr(e) for e in self.exprs ] ) + "}"
|
---|
| 1707 |
|
---|
| 1708 | return self.strRepr
|
---|
| 1709 |
|
---|
| 1710 | def checkRecursion( self, parseElementList ):
|
---|
| 1711 | subRecCheckList = parseElementList[:] + [ self ]
|
---|
| 1712 | for e in self.exprs:
|
---|
| 1713 | e.checkRecursion( subRecCheckList )
|
---|
| 1714 |
|
---|
| 1715 |
|
---|
| 1716 | class MatchFirst(ParseExpression):
|
---|
| 1717 | """Requires that at least one ParseExpression is found.
|
---|
| 1718 | If two expressions match, the first one listed is the one that will match.
|
---|
| 1719 | May be constructed using the '|' operator.
|
---|
| 1720 | """
|
---|
| 1721 | def __init__( self, exprs, savelist = False ):
|
---|
| 1722 | super(MatchFirst,self).__init__(exprs, savelist)
|
---|
| 1723 | if exprs:
|
---|
| 1724 | self.mayReturnEmpty = False
|
---|
| 1725 | for e in exprs:
|
---|
| 1726 | if e.mayReturnEmpty:
|
---|
| 1727 | self.mayReturnEmpty = True
|
---|
| 1728 | break
|
---|
| 1729 | else:
|
---|
| 1730 | self.mayReturnEmpty = True
|
---|
| 1731 |
|
---|
| 1732 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1733 | maxExcLoc = -1
|
---|
| 1734 | for e in self.exprs:
|
---|
| 1735 | try:
|
---|
| 1736 | ret = e.parse( instring, loc, doActions )
|
---|
| 1737 | return ret
|
---|
| 1738 | except ParseException, err:
|
---|
| 1739 | if err.loc > maxExcLoc:
|
---|
| 1740 | maxException = err
|
---|
| 1741 | maxExcLoc = err.loc
|
---|
| 1742 | except IndexError, err:
|
---|
| 1743 | if len(instring) > maxExcLoc:
|
---|
| 1744 | maxException = ParseException(instring,len(instring),e.errmsg,self)
|
---|
| 1745 | maxExcLoc = len(instring)
|
---|
| 1746 |
|
---|
| 1747 | # only got here if no expression matched, raise exception for match that made it the furthest
|
---|
| 1748 | else:
|
---|
| 1749 | if self.exprs:
|
---|
| 1750 | raise maxException
|
---|
| 1751 | else:
|
---|
| 1752 | raise ParseException(instring, loc, "no defined alternatives to match", self)
|
---|
| 1753 |
|
---|
| 1754 | def __ior__(self, other ):
|
---|
| 1755 | if isinstance( other, basestring ):
|
---|
| 1756 | other = Literal( other )
|
---|
| 1757 | return self.append( other ) #MatchFirst( [ self, other ] )
|
---|
| 1758 |
|
---|
| 1759 | def __str__( self ):
|
---|
| 1760 | if hasattr(self,"name"):
|
---|
| 1761 | return self.name
|
---|
| 1762 |
|
---|
| 1763 | if self.strRepr is None:
|
---|
| 1764 | self.strRepr = "{" + " | ".join( [ _ustr(e) for e in self.exprs ] ) + "}"
|
---|
| 1765 |
|
---|
| 1766 | return self.strRepr
|
---|
| 1767 |
|
---|
| 1768 | def checkRecursion( self, parseElementList ):
|
---|
| 1769 | subRecCheckList = parseElementList[:] + [ self ]
|
---|
| 1770 | for e in self.exprs:
|
---|
| 1771 | e.checkRecursion( subRecCheckList )
|
---|
| 1772 |
|
---|
| 1773 | class Each(ParseExpression):
|
---|
| 1774 | """Requires all given ParseExpressions to be found, but in any order.
|
---|
| 1775 | Expressions may be separated by whitespace.
|
---|
| 1776 | May be constructed using the '&' operator.
|
---|
| 1777 | """
|
---|
| 1778 | def __init__( self, exprs, savelist = True ):
|
---|
| 1779 | super(Each,self).__init__(exprs, savelist)
|
---|
| 1780 | self.mayReturnEmpty = True
|
---|
| 1781 | for e in exprs:
|
---|
| 1782 | if not e.mayReturnEmpty:
|
---|
| 1783 | self.mayReturnEmpty = False
|
---|
| 1784 | break
|
---|
| 1785 | self.skipWhitespace = True
|
---|
| 1786 | self.optionals = [ e.expr for e in exprs if isinstance(e,Optional) ]
|
---|
| 1787 | self.multioptionals = [ e.expr for e in exprs if isinstance(e,ZeroOrMore) ]
|
---|
| 1788 | self.multirequired = [ e.expr for e in exprs if isinstance(e,OneOrMore) ]
|
---|
| 1789 | self.required = [ e for e in exprs if not isinstance(e,(Optional,ZeroOrMore,OneOrMore)) ]
|
---|
| 1790 | self.required += self.multirequired
|
---|
| 1791 |
|
---|
| 1792 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1793 | tmpLoc = loc
|
---|
| 1794 | tmpReqd = self.required[:]
|
---|
| 1795 | tmpOpt = self.optionals[:]
|
---|
| 1796 | matchOrder = []
|
---|
| 1797 |
|
---|
| 1798 | keepMatching = True
|
---|
| 1799 | while keepMatching:
|
---|
| 1800 | tmpExprs = tmpReqd + tmpOpt + self.multioptionals + self.multirequired
|
---|
| 1801 | failed = []
|
---|
| 1802 | for e in tmpExprs:
|
---|
| 1803 | try:
|
---|
| 1804 | tmpLoc = e.tryParse( instring, tmpLoc )
|
---|
| 1805 | except ParseException:
|
---|
| 1806 | failed.append(e)
|
---|
| 1807 | else:
|
---|
| 1808 | matchOrder.append(e)
|
---|
| 1809 | if e in tmpReqd:
|
---|
| 1810 | tmpReqd.remove(e)
|
---|
| 1811 | elif e in tmpOpt:
|
---|
| 1812 | tmpOpt.remove(e)
|
---|
| 1813 | if len(failed) == len(tmpExprs):
|
---|
| 1814 | keepMatching = False
|
---|
| 1815 |
|
---|
| 1816 | if tmpReqd:
|
---|
| 1817 | missing = ", ".join( [ str(e) for e in tmpReqd ] )
|
---|
| 1818 | raise ParseException(instring,loc,"Missing one or more required elements (%s)" % missing )
|
---|
| 1819 |
|
---|
| 1820 | resultlist = []
|
---|
| 1821 | for e in matchOrder:
|
---|
| 1822 | loc,results = e.parse(instring,loc,doActions)
|
---|
| 1823 | resultlist.append(results)
|
---|
| 1824 |
|
---|
| 1825 | finalResults = ParseResults([])
|
---|
| 1826 | for r in resultlist:
|
---|
| 1827 | dups = {}
|
---|
| 1828 | for k in r.keys():
|
---|
| 1829 | if k in finalResults.keys():
|
---|
| 1830 | tmp = ParseResults(finalResults[k])
|
---|
| 1831 | tmp += ParseResults(r[k])
|
---|
| 1832 | dups[k] = tmp
|
---|
| 1833 | finalResults += ParseResults(r)
|
---|
| 1834 | for k,v in dups.items():
|
---|
| 1835 | finalResults[k] = v
|
---|
| 1836 | return loc, finalResults
|
---|
| 1837 |
|
---|
| 1838 | def __str__( self ):
|
---|
| 1839 | if hasattr(self,"name"):
|
---|
| 1840 | return self.name
|
---|
| 1841 |
|
---|
| 1842 | if self.strRepr is None:
|
---|
| 1843 | self.strRepr = "{" + " & ".join( [ _ustr(e) for e in self.exprs ] ) + "}"
|
---|
| 1844 |
|
---|
| 1845 | return self.strRepr
|
---|
| 1846 |
|
---|
| 1847 | def checkRecursion( self, parseElementList ):
|
---|
| 1848 | subRecCheckList = parseElementList[:] + [ self ]
|
---|
| 1849 | for e in self.exprs:
|
---|
| 1850 | e.checkRecursion( subRecCheckList )
|
---|
| 1851 |
|
---|
| 1852 |
|
---|
| 1853 | class ParseElementEnhance(ParserElement):
|
---|
| 1854 | """Abstract subclass of ParserElement, for combining and post-processing parsed tokens."""
|
---|
| 1855 | def __init__( self, expr, savelist=False ):
|
---|
| 1856 | super(ParseElementEnhance,self).__init__(savelist)
|
---|
| 1857 | if isinstance( expr, basestring ):
|
---|
| 1858 | expr = Literal(expr)
|
---|
| 1859 | self.expr = expr
|
---|
| 1860 | self.strRepr = None
|
---|
| 1861 | if expr is not None:
|
---|
| 1862 | self.mayIndexError = expr.mayIndexError
|
---|
| 1863 | self.skipWhitespace = expr.skipWhitespace
|
---|
| 1864 | self.whiteChars = expr.whiteChars
|
---|
| 1865 |
|
---|
| 1866 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1867 | if self.expr is not None:
|
---|
| 1868 | return self.expr.parse( instring, loc, doActions )
|
---|
| 1869 | else:
|
---|
| 1870 | raise ParseException(instring,loc,"",self)
|
---|
| 1871 |
|
---|
| 1872 | def leaveWhitespace( self ):
|
---|
| 1873 | self.skipWhitespace = False
|
---|
| 1874 | self.expr = copy.copy(self.expr)
|
---|
| 1875 | if self.expr is not None:
|
---|
| 1876 | self.expr.leaveWhitespace()
|
---|
| 1877 | return self
|
---|
| 1878 |
|
---|
| 1879 | def ignore( self, other ):
|
---|
| 1880 | if isinstance( other, Suppress ):
|
---|
| 1881 | if other not in self.ignoreExprs:
|
---|
| 1882 | super( ParseElementEnhance, self).ignore( other )
|
---|
| 1883 | if self.expr is not None:
|
---|
| 1884 | self.expr.ignore( self.ignoreExprs[-1] )
|
---|
| 1885 | else:
|
---|
| 1886 | super( ParseElementEnhance, self).ignore( other )
|
---|
| 1887 | if self.expr is not None:
|
---|
| 1888 | self.expr.ignore( self.ignoreExprs[-1] )
|
---|
| 1889 | return self
|
---|
| 1890 |
|
---|
| 1891 | def streamline( self ):
|
---|
| 1892 | super(ParseElementEnhance,self).streamline()
|
---|
| 1893 | if self.expr is not None:
|
---|
| 1894 | self.expr.streamline()
|
---|
| 1895 | return self
|
---|
| 1896 |
|
---|
| 1897 | def checkRecursion( self, parseElementList ):
|
---|
| 1898 | if self in parseElementList:
|
---|
| 1899 | raise RecursiveGrammarException( parseElementList+[self] )
|
---|
| 1900 | subRecCheckList = parseElementList[:] + [ self ]
|
---|
| 1901 | if self.expr is not None:
|
---|
| 1902 | self.expr.checkRecursion( subRecCheckList )
|
---|
| 1903 |
|
---|
| 1904 | def validate( self, validateTrace=[] ):
|
---|
| 1905 | tmp = validateTrace[:]+[self]
|
---|
| 1906 | if self.expr is not None:
|
---|
| 1907 | self.expr.validate(tmp)
|
---|
| 1908 | self.checkRecursion( [] )
|
---|
| 1909 |
|
---|
| 1910 | def __str__( self ):
|
---|
| 1911 | try:
|
---|
| 1912 | return super(ParseElementEnhance,self).__str__()
|
---|
| 1913 | except:
|
---|
| 1914 | pass
|
---|
| 1915 |
|
---|
| 1916 | if self.strRepr is None and self.expr is not None:
|
---|
| 1917 | self.strRepr = "%s:(%s)" % ( self.__class__.__name__, _ustr(self.expr) )
|
---|
| 1918 | return self.strRepr
|
---|
| 1919 |
|
---|
| 1920 |
|
---|
| 1921 | class FollowedBy(ParseElementEnhance):
|
---|
| 1922 | """Lookahead matching of the given parse expression. FollowedBy
|
---|
| 1923 | does *not* advance the parsing position within the input string, it only
|
---|
| 1924 | verifies that the specified parse expression matches at the current
|
---|
| 1925 | position. FollowedBy always returns a null token list."""
|
---|
| 1926 | def __init__( self, expr ):
|
---|
| 1927 | super(FollowedBy,self).__init__(expr)
|
---|
| 1928 | self.mayReturnEmpty = True
|
---|
| 1929 |
|
---|
| 1930 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1931 | self.expr.tryParse( instring, loc )
|
---|
| 1932 | return loc, []
|
---|
| 1933 |
|
---|
| 1934 |
|
---|
| 1935 | class NotAny(ParseElementEnhance):
|
---|
| 1936 | """Lookahead to disallow matching with the given parse expression. NotAny
|
---|
| 1937 | does *not* advance the parsing position within the input string, it only
|
---|
| 1938 | verifies that the specified parse expression does *not* match at the current
|
---|
| 1939 | position. Also, NotAny does *not* skip over leading whitespace. NotAny
|
---|
| 1940 | always returns a null token list. May be constructed using the '~' operator."""
|
---|
| 1941 | def __init__( self, expr ):
|
---|
| 1942 | super(NotAny,self).__init__(expr)
|
---|
| 1943 | #~ self.leaveWhitespace()
|
---|
| 1944 | self.skipWhitespace = False # do NOT use self.leaveWhitespace(), don't want to propagate to exprs
|
---|
| 1945 | self.mayReturnEmpty = True
|
---|
| 1946 | self.errmsg = "Found unexpected token, "+_ustr(self.expr)
|
---|
| 1947 | self.myException = ParseException("",0,self.errmsg,self)
|
---|
| 1948 |
|
---|
| 1949 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1950 | try:
|
---|
| 1951 | self.expr.tryParse( instring, loc )
|
---|
| 1952 | except (ParseException,IndexError):
|
---|
| 1953 | pass
|
---|
| 1954 | else:
|
---|
| 1955 | #~ raise ParseException, (instring, loc, self.errmsg )
|
---|
| 1956 | exc = self.myException
|
---|
| 1957 | exc.loc = loc
|
---|
| 1958 | exc.pstr = instring
|
---|
| 1959 | raise exc
|
---|
| 1960 | return loc, []
|
---|
| 1961 |
|
---|
| 1962 | def __str__( self ):
|
---|
| 1963 | if hasattr(self,"name"):
|
---|
| 1964 | return self.name
|
---|
| 1965 |
|
---|
| 1966 | if self.strRepr is None:
|
---|
| 1967 | self.strRepr = "~{" + _ustr(self.expr) + "}"
|
---|
| 1968 |
|
---|
| 1969 | return self.strRepr
|
---|
| 1970 |
|
---|
| 1971 |
|
---|
| 1972 | class ZeroOrMore(ParseElementEnhance):
|
---|
| 1973 | """Optional repetition of zero or more of the given expression."""
|
---|
| 1974 | def __init__( self, expr ):
|
---|
| 1975 | super(ZeroOrMore,self).__init__(expr)
|
---|
| 1976 | self.mayReturnEmpty = True
|
---|
| 1977 |
|
---|
| 1978 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 1979 | tokens = []
|
---|
| 1980 | try:
|
---|
| 1981 | loc, tokens = self.expr.parse( instring, loc, doActions )
|
---|
| 1982 | hasIgnoreExprs = ( len(self.ignoreExprs) > 0 )
|
---|
| 1983 | while 1:
|
---|
| 1984 | if hasIgnoreExprs:
|
---|
| 1985 | loc = self.skipIgnorables( instring, loc )
|
---|
| 1986 | loc, tmptokens = self.expr.parse( instring, loc, doActions )
|
---|
| 1987 | if tmptokens or tmptokens.keys():
|
---|
| 1988 | tokens += tmptokens
|
---|
| 1989 | except (ParseException,IndexError):
|
---|
| 1990 | pass
|
---|
| 1991 |
|
---|
| 1992 | return loc, tokens
|
---|
| 1993 |
|
---|
| 1994 | def __str__( self ):
|
---|
| 1995 | if hasattr(self,"name"):
|
---|
| 1996 | return self.name
|
---|
| 1997 |
|
---|
| 1998 | if self.strRepr is None:
|
---|
| 1999 | self.strRepr = "[" + _ustr(self.expr) + "]..."
|
---|
| 2000 |
|
---|
| 2001 | return self.strRepr
|
---|
| 2002 |
|
---|
| 2003 | def setResultsName( self, name, listAllMatches=False ):
|
---|
| 2004 | ret = super(ZeroOrMore,self).setResultsName(name,listAllMatches)
|
---|
| 2005 | ret.saveAsList = True
|
---|
| 2006 | return ret
|
---|
| 2007 |
|
---|
| 2008 |
|
---|
| 2009 | class OneOrMore(ParseElementEnhance):
|
---|
| 2010 | """Repetition of one or more of the given expression."""
|
---|
| 2011 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 2012 | # must be at least one
|
---|
| 2013 | loc, tokens = self.expr.parse( instring, loc, doActions )
|
---|
| 2014 | try:
|
---|
| 2015 | hasIgnoreExprs = ( len(self.ignoreExprs) > 0 )
|
---|
| 2016 | while 1:
|
---|
| 2017 | if hasIgnoreExprs:
|
---|
| 2018 | loc = self.skipIgnorables( instring, loc )
|
---|
| 2019 | loc, tmptokens = self.expr.parse( instring, loc, doActions )
|
---|
| 2020 | if tmptokens or tmptokens.keys():
|
---|
| 2021 | tokens += tmptokens
|
---|
| 2022 | except (ParseException,IndexError):
|
---|
| 2023 | pass
|
---|
| 2024 |
|
---|
| 2025 | return loc, tokens
|
---|
| 2026 |
|
---|
| 2027 | def __str__( self ):
|
---|
| 2028 | if hasattr(self,"name"):
|
---|
| 2029 | return self.name
|
---|
| 2030 |
|
---|
| 2031 | if self.strRepr is None:
|
---|
| 2032 | self.strRepr = "{" + _ustr(self.expr) + "}..."
|
---|
| 2033 |
|
---|
| 2034 | return self.strRepr
|
---|
| 2035 |
|
---|
| 2036 | def setResultsName( self, name, listAllMatches=False ):
|
---|
| 2037 | ret = super(OneOrMore,self).setResultsName(name,listAllMatches)
|
---|
| 2038 | ret.saveAsList = True
|
---|
| 2039 | return ret
|
---|
| 2040 |
|
---|
| 2041 |
|
---|
| 2042 | class Optional(ParseElementEnhance):
|
---|
| 2043 | """Optional matching of the given expression.
|
---|
| 2044 | A default return string can also be specified, if the optional expression
|
---|
| 2045 | is not found.
|
---|
| 2046 | """
|
---|
| 2047 | def __init__( self, exprs, default=None ):
|
---|
| 2048 | super(Optional,self).__init__( exprs, savelist=False )
|
---|
| 2049 | self.defaultValue = default
|
---|
| 2050 | self.mayReturnEmpty = True
|
---|
| 2051 |
|
---|
| 2052 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 2053 | try:
|
---|
| 2054 | loc, tokens = self.expr.parse( instring, loc, doActions )
|
---|
| 2055 | except (ParseException,IndexError):
|
---|
| 2056 | if self.defaultValue is not None:
|
---|
| 2057 | tokens = [ self.defaultValue ]
|
---|
| 2058 | else:
|
---|
| 2059 | tokens = []
|
---|
| 2060 |
|
---|
| 2061 | return loc, tokens
|
---|
| 2062 |
|
---|
| 2063 | def __str__( self ):
|
---|
| 2064 | if hasattr(self,"name"):
|
---|
| 2065 | return self.name
|
---|
| 2066 |
|
---|
| 2067 | if self.strRepr is None:
|
---|
| 2068 | self.strRepr = "[" + _ustr(self.expr) + "]"
|
---|
| 2069 |
|
---|
| 2070 | return self.strRepr
|
---|
| 2071 |
|
---|
| 2072 |
|
---|
| 2073 | class SkipTo(ParseElementEnhance):
|
---|
| 2074 | """Token for skipping over all undefined text until the matched expression is found.
|
---|
| 2075 | If include is set to true, the matched expression is also consumed. The ignore
|
---|
| 2076 | argument is used to define grammars (typically quoted strings and comments) that
|
---|
| 2077 | might contain false matches.
|
---|
| 2078 | """
|
---|
| 2079 | def __init__( self, other, include=False, ignore=None ):
|
---|
| 2080 | super( SkipTo, self ).__init__( other )
|
---|
| 2081 | if ignore is not None:
|
---|
| 2082 | self.expr = copy.copy( self.expr )
|
---|
| 2083 | self.expr.ignore(ignore)
|
---|
| 2084 | self.mayReturnEmpty = True
|
---|
| 2085 | self.mayIndexError = False
|
---|
| 2086 | self.includeMatch = include
|
---|
| 2087 | self.errmsg = "No match found for "+_ustr(self.expr)
|
---|
| 2088 | self.myException = ParseException("",0,self.errmsg,self)
|
---|
| 2089 |
|
---|
| 2090 | def parseImpl( self, instring, loc, doActions=True ):
|
---|
| 2091 | startLoc = loc
|
---|
| 2092 | instrlen = len(instring)
|
---|
| 2093 | expr = self.expr
|
---|
| 2094 | while loc < instrlen:
|
---|
| 2095 | try:
|
---|
| 2096 | loc = expr.skipIgnorables( instring, loc )
|
---|
| 2097 | expr.parse( instring, loc, doActions=False, callPreParse=False )
|
---|
| 2098 | if self.includeMatch:
|
---|
| 2099 | skipText = instring[startLoc:loc]
|
---|
| 2100 | loc,mat = expr.parse(instring,loc)
|
---|
| 2101 | if mat:
|
---|
| 2102 | return loc, [ skipText, mat ]
|
---|
| 2103 | else:
|
---|
| 2104 | return loc, [ skipText ]
|
---|
| 2105 | else:
|
---|
| 2106 | return loc, [ instring[startLoc:loc] ]
|
---|
| 2107 | except (ParseException,IndexError):
|
---|
| 2108 | loc += 1
|
---|
| 2109 | exc = self.myException
|
---|
| 2110 | exc.loc = loc
|
---|
| 2111 | exc.pstr = instring
|
---|
| 2112 | raise exc
|
---|
| 2113 |
|
---|
| 2114 | class Forward(ParseElementEnhance):
|
---|
| 2115 | """Forward declaration of an expression to be defined later -
|
---|
| 2116 | used for recursive grammars, such as algebraic infix notation.
|
---|
| 2117 | When the expression is known, it is assigned to the Forward variable using the '<<' operator.
|
---|
| 2118 |
|
---|
| 2119 | Note: take care when assigning to Forward to not overlook precedence of operators.
|
---|
| 2120 | Specifically, '|' has a lower precedence than '<<', so that::
|
---|
| 2121 | fwdExpr << a | b | c
|
---|
| 2122 | will actually be evaluated as::
|
---|
| 2123 | (fwdExpr << a) | b | c
|
---|
| 2124 | thereby leaving b and c out as parseable alternatives. It is recommended that you
|
---|
| 2125 | explicitly group the values inserted into the Forward::
|
---|
| 2126 | fwdExpr << (a | b | c)
|
---|
| 2127 | """
|
---|
| 2128 | def __init__( self, other=None ):
|
---|
| 2129 | super(Forward,self).__init__( other, savelist=False )
|
---|
| 2130 |
|
---|
| 2131 | def __lshift__( self, other ):
|
---|
| 2132 | self.expr = other
|
---|
| 2133 | self.mayReturnEmpty = other.mayReturnEmpty
|
---|
| 2134 | self.strRepr = None
|
---|
| 2135 | return self
|
---|
| 2136 |
|
---|
| 2137 | def leaveWhitespace( self ):
|
---|
| 2138 | self.skipWhitespace = False
|
---|
| 2139 | return self
|
---|
| 2140 |
|
---|
| 2141 | def streamline( self ):
|
---|
| 2142 | if not self.streamlined:
|
---|
| 2143 | self.streamlined = True
|
---|
| 2144 | if self.expr is not None:
|
---|
| 2145 | self.expr.streamline()
|
---|
| 2146 | return self
|
---|
| 2147 |
|
---|
| 2148 | def validate( self, validateTrace=[] ):
|
---|
| 2149 | if self not in validateTrace:
|
---|
| 2150 | tmp = validateTrace[:]+[self]
|
---|
| 2151 | if self.expr is not None:
|
---|
| 2152 | self.expr.validate(tmp)
|
---|
| 2153 | self.checkRecursion([])
|
---|
| 2154 |
|
---|
| 2155 | def __str__( self ):
|
---|
| 2156 | if hasattr(self,"name"):
|
---|
| 2157 | return self.name
|
---|
| 2158 |
|
---|
| 2159 | self.__class__ = _ForwardNoRecurse
|
---|
| 2160 | try:
|
---|
| 2161 | if self.expr is not None:
|
---|
| 2162 | retString = _ustr(self.expr)
|
---|
| 2163 | else:
|
---|
| 2164 | retString = "None"
|
---|
| 2165 | finally:
|
---|
| 2166 | self.__class__ = Forward
|
---|
| 2167 | return "Forward: "+retString
|
---|
| 2168 |
|
---|
| 2169 | class _ForwardNoRecurse(Forward):
|
---|
| 2170 | def __str__( self ):
|
---|
| 2171 | return "..."
|
---|
| 2172 |
|
---|
| 2173 | class TokenConverter(ParseElementEnhance):
|
---|
| 2174 | """Abstract subclass of ParseExpression, for converting parsed results."""
|
---|
| 2175 | def __init__( self, expr, savelist=False ):
|
---|
| 2176 | super(TokenConverter,self).__init__( expr )#, savelist )
|
---|
| 2177 |
|
---|
| 2178 |
|
---|
| 2179 | class Upcase(TokenConverter):
|
---|
| 2180 | """Converter to upper case all matching tokens."""
|
---|
| 2181 | def __init__(self, *args):
|
---|
| 2182 | super(Upcase,self).__init__(*args)
|
---|
| 2183 | warnings.warn("Upcase class is deprecated, use upcaseTokens parse action instead",
|
---|
| 2184 | DeprecationWarning,stacklevel=2)
|
---|
| 2185 |
|
---|
| 2186 | def postParse( self, instring, loc, tokenlist ):
|
---|
| 2187 | return map( string.upper, tokenlist )
|
---|
| 2188 |
|
---|
| 2189 |
|
---|
| 2190 | class Combine(TokenConverter):
|
---|
| 2191 | """Converter to concatenate all matching tokens to a single string.
|
---|
| 2192 | By default, the matching patterns must also be contiguous in the input string;
|
---|
| 2193 | this can be disabled by specifying 'adjacent=False' in the constructor.
|
---|
| 2194 | """
|
---|
| 2195 | def __init__( self, expr, joinString="", adjacent=True ):
|
---|
| 2196 | super(Combine,self).__init__( expr )
|
---|
| 2197 | # suppress whitespace-stripping in contained parse expressions, but re-enable it on the Combine itself
|
---|
| 2198 | if adjacent:
|
---|
| 2199 | self.leaveWhitespace()
|
---|
| 2200 | self.adjacent = adjacent
|
---|
| 2201 | self.skipWhitespace = True
|
---|
| 2202 | self.joinString = joinString
|
---|
| 2203 |
|
---|
| 2204 | def ignore( self, other ):
|
---|
| 2205 | if self.adjacent:
|
---|
| 2206 | ParserElement.ignore(self, other)
|
---|
| 2207 | else:
|
---|
| 2208 | super( Combine, self).ignore( other )
|
---|
| 2209 | return self
|
---|
| 2210 |
|
---|
| 2211 | def postParse( self, instring, loc, tokenlist ):
|
---|
| 2212 | retToks = tokenlist.copy()
|
---|
| 2213 | del retToks[:]
|
---|
| 2214 | retToks += ParseResults([ "".join(tokenlist._asStringList(self.joinString)) ], modal=self.modalResults)
|
---|
| 2215 |
|
---|
| 2216 | if self.resultsName and len(retToks.keys())>0:
|
---|
| 2217 | return [ retToks ]
|
---|
| 2218 | else:
|
---|
| 2219 | return retToks
|
---|
| 2220 |
|
---|
| 2221 | class Group(TokenConverter):
|
---|
| 2222 | """Converter to return the matched tokens as a list - useful for returning tokens of ZeroOrMore and OneOrMore expressions."""
|
---|
| 2223 | def __init__( self, expr ):
|
---|
| 2224 | super(Group,self).__init__( expr )
|
---|
| 2225 | self.saveAsList = True
|
---|
| 2226 |
|
---|
| 2227 | def postParse( self, instring, loc, tokenlist ):
|
---|
| 2228 | return [ tokenlist ]
|
---|
| 2229 |
|
---|
| 2230 | class Dict(TokenConverter):
|
---|
| 2231 | """Converter to return a repetitive expression as a list, but also as a dictionary.
|
---|
| 2232 | Each element can also be referenced using the first token in the expression as its key.
|
---|
| 2233 | Useful for tabular report scraping when the first column can be used as a item key.
|
---|
| 2234 | """
|
---|
| 2235 | def __init__( self, exprs ):
|
---|
| 2236 | super(Dict,self).__init__( exprs )
|
---|
| 2237 | self.saveAsList = True
|
---|
| 2238 |
|
---|
| 2239 | def postParse( self, instring, loc, tokenlist ):
|
---|
| 2240 | for i,tok in enumerate(tokenlist):
|
---|
| 2241 | ikey = _ustr(tok[0]).strip()
|
---|
| 2242 | if len(tok)==1:
|
---|
| 2243 | tokenlist[ikey] = ("",i)
|
---|
| 2244 | elif len(tok)==2 and not isinstance(tok[1],ParseResults):
|
---|
| 2245 | tokenlist[ikey] = (tok[1],i)
|
---|
| 2246 | else:
|
---|
| 2247 | dictvalue = tok.copy() #ParseResults(i)
|
---|
| 2248 | del dictvalue[0]
|
---|
| 2249 | if len(dictvalue)!= 1 or (isinstance(dictvalue,ParseResults) and dictvalue.keys()):
|
---|
| 2250 | tokenlist[ikey] = (dictvalue,i)
|
---|
| 2251 | else:
|
---|
| 2252 | tokenlist[ikey] = (dictvalue[0],i)
|
---|
| 2253 |
|
---|
| 2254 | if self.resultsName:
|
---|
| 2255 | return [ tokenlist ]
|
---|
| 2256 | else:
|
---|
| 2257 | return tokenlist
|
---|
| 2258 |
|
---|
| 2259 |
|
---|
| 2260 | class Suppress(TokenConverter):
|
---|
| 2261 | """Converter for ignoring the results of a parsed expression."""
|
---|
| 2262 | def postParse( self, instring, loc, tokenlist ):
|
---|
| 2263 | return []
|
---|
| 2264 |
|
---|
| 2265 | def suppress( self ):
|
---|
| 2266 | return self
|
---|
| 2267 |
|
---|
| 2268 | #
|
---|
| 2269 | # global helpers
|
---|
| 2270 | #
|
---|
| 2271 | def delimitedList( expr, delim=",", combine=False ):
|
---|
| 2272 | """Helper to define a delimited list of expressions - the delimiter defaults to ','.
|
---|
| 2273 | By default, the list elements and delimiters can have intervening whitespace, and
|
---|
| 2274 | comments, but this can be overridden by passing 'combine=True' in the constructor.
|
---|
| 2275 | If combine is set to True, the matching tokens are returned as a single token
|
---|
| 2276 | string, with the delimiters included; otherwise, the matching tokens are returned
|
---|
| 2277 | as a list of tokens, with the delimiters suppressed.
|
---|
| 2278 | """
|
---|
| 2279 | if combine:
|
---|
| 2280 | return Combine( expr + ZeroOrMore( delim + expr ) ).setName(_ustr(expr)+_ustr(delim)+"...")
|
---|
| 2281 | else:
|
---|
| 2282 | return ( expr + ZeroOrMore( Suppress( delim ) + expr ) ).setName(_ustr(expr)+_ustr(delim)+"...")
|
---|
| 2283 |
|
---|
| 2284 | def _escapeRegexRangeChars(s):
|
---|
| 2285 | #~ escape these chars: ^-]
|
---|
| 2286 | for c in r"\^-]":
|
---|
| 2287 | s = s.replace(c,"\\"+c)
|
---|
| 2288 | s = s.replace("\n",r"\n")
|
---|
| 2289 | s = s.replace("\t",r"\t")
|
---|
| 2290 | return _ustr(s)
|
---|
| 2291 |
|
---|
| 2292 | def oneOf( strs, caseless=False, useRegex=True ):
|
---|
| 2293 | """Helper to quickly define a set of alternative Literals, and makes sure to do
|
---|
| 2294 | longest-first testing when there is a conflict, regardless of the input order,
|
---|
| 2295 | but returns a MatchFirst for best performance.
|
---|
| 2296 |
|
---|
| 2297 | Parameters:
|
---|
| 2298 | - strs - a string of space-delimited literals, or a list of string literals
|
---|
| 2299 | - caseless - (default=False) - treat all literals as caseless
|
---|
| 2300 | - useRegex - (default=True) - as an optimization, will generate a Regex
|
---|
| 2301 | object; otherwise, will generate a MatchFirst object (if caseless=True, or
|
---|
| 2302 | if creating a Regex raises an exception)
|
---|
| 2303 | """
|
---|
| 2304 | if caseless:
|
---|
| 2305 | isequal = ( lambda a,b: a.upper() == b.upper() )
|
---|
| 2306 | masks = ( lambda a,b: b.upper().startswith(a.upper()) )
|
---|
| 2307 | parseElementClass = CaselessLiteral
|
---|
| 2308 | else:
|
---|
| 2309 | isequal = ( lambda a,b: a == b )
|
---|
| 2310 | masks = ( lambda a,b: b.startswith(a) )
|
---|
| 2311 | parseElementClass = Literal
|
---|
| 2312 |
|
---|
| 2313 | if isinstance(strs,list):
|
---|
| 2314 | symbols = strs[:]
|
---|
| 2315 | elif isinstance(strs,basestring):
|
---|
| 2316 | symbols = strs.split()
|
---|
| 2317 | else:
|
---|
| 2318 | warnings.warn("Invalid argument to oneOf, expected string or list",
|
---|
| 2319 | SyntaxWarning, stacklevel=2)
|
---|
| 2320 |
|
---|
| 2321 | i = 0
|
---|
| 2322 | while i < len(symbols)-1:
|
---|
| 2323 | cur = symbols[i]
|
---|
| 2324 | for j,other in enumerate(symbols[i+1:]):
|
---|
| 2325 | if ( isequal(other, cur) ):
|
---|
| 2326 | del symbols[i+j+1]
|
---|
| 2327 | break
|
---|
| 2328 | elif ( masks(cur, other) ):
|
---|
| 2329 | del symbols[i+j+1]
|
---|
| 2330 | symbols.insert(i,other)
|
---|
| 2331 | cur = other
|
---|
| 2332 | break
|
---|
| 2333 | else:
|
---|
| 2334 | i += 1
|
---|
| 2335 |
|
---|
| 2336 | if not caseless and useRegex:
|
---|
| 2337 | #~ print strs,"->", "|".join( [ _escapeRegexChars(sym) for sym in symbols] )
|
---|
| 2338 | try:
|
---|
| 2339 | if len(symbols)==len("".join(symbols)):
|
---|
| 2340 | return Regex( "[%s]" % "".join( [ _escapeRegexRangeChars(sym) for sym in symbols] ) )
|
---|
| 2341 | else:
|
---|
| 2342 | return Regex( "|".join( [ re.escape(sym) for sym in symbols] ) )
|
---|
| 2343 | except:
|
---|
| 2344 | warnings.warn("Exception creating Regex for oneOf, building MatchFirst",
|
---|
| 2345 | SyntaxWarning, stacklevel=2)
|
---|
| 2346 |
|
---|
| 2347 |
|
---|
| 2348 | # last resort, just use MatchFirst
|
---|
| 2349 | return MatchFirst( [ parseElementClass(sym) for sym in symbols ] )
|
---|
| 2350 |
|
---|
| 2351 | def dictOf( key, value ):
|
---|
| 2352 | """Helper to easily and clearly define a dictionary by specifying the respective patterns
|
---|
| 2353 | for the key and value. Takes care of defining the Dict, ZeroOrMore, and Group tokens
|
---|
| 2354 | in the proper order. The key pattern can include delimiting markers or punctuation,
|
---|
| 2355 | as long as they are suppressed, thereby leaving the significant key text. The value
|
---|
| 2356 | pattern can include named results, so that the Dict results can include named token
|
---|
| 2357 | fields.
|
---|
| 2358 | """
|
---|
| 2359 | return Dict( ZeroOrMore( Group ( key + value ) ) )
|
---|
| 2360 |
|
---|
| 2361 | _bslash = "\\"
|
---|
| 2362 | printables = "".join( [ c for c in string.printable if c not in string.whitespace ] )
|
---|
| 2363 |
|
---|
| 2364 | # convenience constants for positional expressions
|
---|
| 2365 | empty = Empty().setName("empty")
|
---|
| 2366 | lineStart = LineStart().setName("lineStart")
|
---|
| 2367 | lineEnd = LineEnd().setName("lineEnd")
|
---|
| 2368 | stringStart = StringStart().setName("stringStart")
|
---|
| 2369 | stringEnd = StringEnd().setName("stringEnd")
|
---|
| 2370 |
|
---|
| 2371 | _escapedPunc = Word( _bslash, r"\[]-*.$+^?()~ ", exact=2 ).setParseAction(lambda s,l,t:t[0][1])
|
---|
| 2372 | _printables_less_backslash = "".join([ c for c in printables if c not in r"\]" ])
|
---|
| 2373 | _escapedHexChar = Combine( Suppress(_bslash + "0x") + Word(hexnums) ).setParseAction(lambda s,l,t:unichr(int(t[0],16)))
|
---|
| 2374 | _escapedOctChar = Combine( Suppress(_bslash) + Word("0","01234567") ).setParseAction(lambda s,l,t:unichr(int(t[0],8)))
|
---|
| 2375 | _singleChar = _escapedPunc | _escapedHexChar | _escapedOctChar | Word(_printables_less_backslash,exact=1)
|
---|
| 2376 | _charRange = Group(_singleChar + Suppress("-") + _singleChar)
|
---|
| 2377 | _reBracketExpr = "[" + Optional("^").setResultsName("negate") + Group( OneOrMore( _charRange | _singleChar ) ).setResultsName("body") + "]"
|
---|
| 2378 |
|
---|
| 2379 | _expanded = lambda p: (isinstance(p,ParseResults) and ''.join([ unichr(c) for c in range(ord(p[0]),ord(p[1])+1) ]) or p)
|
---|
| 2380 |
|
---|
| 2381 | def srange(s):
|
---|
| 2382 | r"""Helper to easily define string ranges for use in Word construction. Borrows
|
---|
| 2383 | syntax from regexp '[]' string range definitions::
|
---|
| 2384 | srange("[0-9]") -> "0123456789"
|
---|
| 2385 | srange("[a-z]") -> "abcdefghijklmnopqrstuvwxyz"
|
---|
| 2386 | srange("[a-z$_]") -> "abcdefghijklmnopqrstuvwxyz$_"
|
---|
| 2387 | The input string must be enclosed in []'s, and the returned string is the expanded
|
---|
| 2388 | character set joined into a single string.
|
---|
| 2389 | The values enclosed in the []'s may be::
|
---|
| 2390 | a single character
|
---|
| 2391 | an escaped character with a leading backslash (such as \- or \])
|
---|
| 2392 | an escaped hex character with a leading '\0x' (\0x21, which is a '!' character)
|
---|
| 2393 | an escaped octal character with a leading '\0' (\041, which is a '!' character)
|
---|
| 2394 | a range of any of the above, separated by a dash ('a-z', etc.)
|
---|
| 2395 | any combination of the above ('aeiouy', 'a-zA-Z0-9_$', etc.)
|
---|
| 2396 | """
|
---|
| 2397 | try:
|
---|
| 2398 | return "".join([_expanded(part) for part in _reBracketExpr.parseString(s).body])
|
---|
| 2399 | except:
|
---|
| 2400 | return ""
|
---|
| 2401 |
|
---|
| 2402 | def replaceWith(replStr):
|
---|
| 2403 | """Helper method for common parse actions that simply return a literal value. Especially
|
---|
| 2404 | useful when used with transformString().
|
---|
| 2405 | """
|
---|
| 2406 | def _replFunc(*args):
|
---|
| 2407 | return [replStr]
|
---|
| 2408 | return _replFunc
|
---|
| 2409 |
|
---|
| 2410 | def removeQuotes(s,l,t):
|
---|
| 2411 | """Helper parse action for removing quotation marks from parsed quoted strings.
|
---|
| 2412 | To use, add this parse action to quoted string using::
|
---|
| 2413 | quotedString.setParseAction( removeQuotes )
|
---|
| 2414 | """
|
---|
| 2415 | return t[0][1:-1]
|
---|
| 2416 |
|
---|
| 2417 | def upcaseTokens(s,l,t):
|
---|
| 2418 | """Helper parse action to convert tokens to upper case."""
|
---|
| 2419 | return map( str.upper, t )
|
---|
| 2420 |
|
---|
| 2421 | def downcaseTokens(s,l,t):
|
---|
| 2422 | """Helper parse action to convert tokens to lower case."""
|
---|
| 2423 | return map( str.lower, t )
|
---|
| 2424 |
|
---|
| 2425 | def _makeTags(tagStr, xml):
|
---|
| 2426 | """Internal helper to construct opening and closing tag expressions, given a tag name"""
|
---|
| 2427 | tagAttrName = Word(alphanums)
|
---|
| 2428 | if (xml):
|
---|
| 2429 | tagAttrValue = dblQuotedString.copy().setParseAction( removeQuotes )
|
---|
| 2430 | openTag = Suppress("<") + Keyword(tagStr) + \
|
---|
| 2431 | Dict(ZeroOrMore(Group( tagAttrName + Suppress("=") + tagAttrValue ))) + \
|
---|
| 2432 | Optional("/",default=[False]).setResultsName("empty").setParseAction(lambda s,l,t:t[0]=='/') + Suppress(">")
|
---|
| 2433 | else:
|
---|
| 2434 | printablesLessRAbrack = "".join( [ c for c in printables if c not in ">" ] )
|
---|
| 2435 | tagAttrValue = quotedString.copy().setParseAction( removeQuotes ) | Word(printablesLessRAbrack)
|
---|
| 2436 | openTag = Suppress("<") + Keyword(tagStr,caseless=True) + \
|
---|
| 2437 | Dict(ZeroOrMore(Group( tagAttrName.setParseAction(downcaseTokens) + \
|
---|
| 2438 | Suppress("=") + tagAttrValue ))) + \
|
---|
| 2439 | Optional("/",default=[False]).setResultsName("empty").setParseAction(lambda s,l,t:t[0]=='/') + Suppress(">")
|
---|
| 2440 | closeTag = Combine("</" + Keyword(tagStr,caseless=not xml) + ">")
|
---|
| 2441 |
|
---|
| 2442 | openTag = openTag.setResultsName("start"+"".join(tagStr.replace(":"," ").title().split())).setName("<%s>" % tagStr)
|
---|
| 2443 | closeTag = closeTag.setResultsName("end"+"".join(tagStr.replace(":"," ").title().split())).setName("</%s>" % tagStr)
|
---|
| 2444 |
|
---|
| 2445 | return openTag, closeTag
|
---|
| 2446 |
|
---|
| 2447 | def makeHTMLTags(tagStr):
|
---|
| 2448 | """Helper to construct opening and closing tag expressions for HTML, given a tag name"""
|
---|
| 2449 | return _makeTags( tagStr, False )
|
---|
| 2450 |
|
---|
| 2451 | def makeXMLTags(tagStr):
|
---|
| 2452 | """Helper to construct opening and closing tag expressions for XML, given a tag name"""
|
---|
| 2453 | return _makeTags( tagStr, True )
|
---|
| 2454 |
|
---|
| 2455 | alphas8bit = srange(r"[\0xc0-\0xd6\0xd8-\0xf6\0xf8-\0xfe]")
|
---|
| 2456 |
|
---|
| 2457 | _escapedChar = Regex(r"\\.")
|
---|
| 2458 | dblQuotedString = Regex(r'"([^"\n\r\\]|("")|(\\.))*"').setName("string enclosed in double quotes")
|
---|
| 2459 | sglQuotedString = Regex(r"'([^'\n\r\\]|('')|(\\.))*'").setName("string enclosed in single quotes")
|
---|
| 2460 | quotedString = Regex(r'''("([^"\n\r\\]|("")|(\\.))*")|('([^'\n\r\\]|('')|(\\.))*')''').setName("quotedString using single or double quotes")
|
---|
| 2461 |
|
---|
| 2462 | # it's easy to get these comment structures wrong - they're very common, so may as well make them available
|
---|
| 2463 | cStyleComment = Regex(r"\/\*[\s\S]*?\*\/").setName("C style comment")
|
---|
| 2464 | htmlComment = Regex(r"<!--[\s\S]*?-->")
|
---|
| 2465 | restOfLine = Regex(r".*").leaveWhitespace()
|
---|
| 2466 | dblSlashComment = Regex(r"\/\/.*").setName("// comment")
|
---|
| 2467 | cppStyleComment = Regex(r"(\/\*[\s\S]*?\*\/)|(\/\/.*)").setName("C++ style comment")
|
---|
| 2468 | javaStyleComment = cppStyleComment
|
---|
| 2469 | pythonStyleComment = Regex(r"#.*").setName("Python style comment")
|
---|
| 2470 | _noncomma = "".join( [ c for c in printables if c != "," ] )
|
---|
| 2471 | _commasepitem = Combine(OneOrMore(Word(_noncomma) +
|
---|
| 2472 | Optional( Word(" \t") +
|
---|
| 2473 | ~Literal(",") + ~LineEnd() ) ) ).streamline().setName("commaItem")
|
---|
| 2474 | commaSeparatedList = delimitedList( Optional( quotedString | _commasepitem, default="") ).setName("commaSeparatedList")
|
---|
| 2475 |
|
---|
| 2476 |
|
---|
| 2477 | if __name__ == "__main__":
|
---|
| 2478 |
|
---|
| 2479 | def test( teststring ):
|
---|
| 2480 | print teststring,"->",
|
---|
| 2481 | try:
|
---|
| 2482 | tokens = simpleSQL.parseString( teststring )
|
---|
| 2483 | tokenlist = tokens.asList()
|
---|
| 2484 | print tokenlist
|
---|
| 2485 | print "tokens = ", tokens
|
---|
| 2486 | print "tokens.columns =", tokens.columns
|
---|
| 2487 | print "tokens.tables =", tokens.tables
|
---|
| 2488 | print tokens.asXML("SQL",True)
|
---|
| 2489 | except ParseException, err:
|
---|
| 2490 | print err.line
|
---|
| 2491 | print " "*(err.column-1) + "^"
|
---|
| 2492 | print err
|
---|
| 2493 | print
|
---|
| 2494 |
|
---|
| 2495 | selectToken = CaselessLiteral( "select" )
|
---|
| 2496 | fromToken = CaselessLiteral( "from" )
|
---|
| 2497 |
|
---|
| 2498 | ident = Word( alphas, alphanums + "_$" )
|
---|
| 2499 | columnName = delimitedList( ident, ".", combine=True ).setParseAction( upcaseTokens )
|
---|
| 2500 | columnNameList = Group( delimitedList( columnName ) )#.setName("columns")
|
---|
| 2501 | tableName = delimitedList( ident, ".", combine=True ).setParseAction( upcaseTokens )
|
---|
| 2502 | tableNameList = Group( delimitedList( tableName ) )#.setName("tables")
|
---|
| 2503 | simpleSQL = ( selectToken + \
|
---|
| 2504 | ( '*' | columnNameList ).setResultsName( "columns" ) + \
|
---|
| 2505 | fromToken + \
|
---|
| 2506 | tableNameList.setResultsName( "tables" ) )
|
---|
| 2507 |
|
---|
| 2508 | test( "SELECT * from XYZZY, ABC" )
|
---|
| 2509 | test( "select * from SYS.XYZZY" )
|
---|
| 2510 | test( "Select A from Sys.dual" )
|
---|
| 2511 | test( "Select AA,BB,CC from Sys.dual" )
|
---|
| 2512 | test( "Select A, B, C from Sys.dual" )
|
---|
| 2513 | test( "Select A, B, C from Sys.dual" )
|
---|
| 2514 | test( "Xelect A, B, C from Sys.dual" )
|
---|
| 2515 | test( "Select A, B, C frox Sys.dual" )
|
---|
| 2516 | test( "Select" )
|
---|
| 2517 | test( "Select ^^^ frox Sys.dual" )
|
---|
| 2518 | test( "Select A, B, C from Sys.dual, Table2 " )
|
---|