Source

ply / ply / lex.py

The default branch has multiple heads

Full commit
  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
#-----------------------------------------------------------------------------
# ply: lex.py
#
# Author: David M. Beazley (dave@dabeaz.com)
#
# Copyright (C) 2001-2006, David M. Beazley
#
# This library is free software; you can redistribute it and/or
# modify it under the terms of the GNU Lesser General Public
# License as published by the Free Software Foundation; either
# version 2.1 of the License, or (at your option) any later version.
# 
# This library is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
# Lesser General Public License for more details.
# 
# You should have received a copy of the GNU Lesser General Public
# License along with this library; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
# 
# See the file COPYING for a complete copy of the LGPL.
#
# 
# This module automatically constructs a lexical analysis module from regular
# expression rules defined in a user-defined module.  The idea is essentially the same
# as that used in John Aycock's Spark framework, but the implementation works
# at the module level rather than requiring the use of classes.
#
# This module tries to provide an interface that is closely modeled after
# the traditional lex interface in Unix.  It also differs from Spark
# in that:
#
#   -  It provides more extensive error checking and reporting if
#      the user supplies a set of regular expressions that can't
#      be compiled or if there is any other kind of a problem in
#      the specification.
#
#   -  The interface is geared towards LALR(1) and LR(1) parser
#      generators.  That is tokens are generated one at a time
#      rather than being generated in advanced all in one step.
#-----------------------------------------------------------------------------

r"""
lex.py

This module builds lex-like scanners based on regular expression rules.
To use the module, simply write a collection of regular expression rules
and actions like this:

# lexer.py
import lex

# Define a list of valid tokens
tokens = (
    'IDENTIFIER', 'NUMBER', 'PLUS', 'MINUS'
    )

# Define tokens as functions
def t_IDENTIFIER(t):
    r' ([a-zA-Z_](\w|_)* '
    return t

def t_NUMBER(t):
    r' \d+ '
    return t

# Some simple tokens with no actions
t_PLUS = r'\+'
t_MINUS = r'-'

# Initialize the lexer
lex.lex()

The tokens list is required and contains a complete list of all valid
token types that the lexer is allowed to produce.  Token types are
restricted to be valid identifiers.  This means that 'MINUS' is a valid
token type whereas '-' is not.

Rules are defined by writing a function with a name of the form
t_rulename.  Each rule must accept a single argument which is
a token object generated by the lexer. This token has the following
attributes:

    t.type   = type string of the token.  This is initially set to the
               name of the rule without the leading t_
    t.value  = The value of the lexeme.
    t.lineno = The value of the line number where the token was encountered
    
For example, the t_NUMBER() rule above might be called with the following:
    
    t.type  = 'NUMBER'
    t.value = '42'
    t.lineno = 3

Each rule returns the token object it would like to supply to the
parser.  In most cases, the token t is returned with few, if any
modifications.  To discard a token for things like whitespace or
comments, simply return nothing.  For instance:

def t_whitespace(t):
    r' \s+ '
    pass

For faster lexing, you can also define this in terms of the ignore set like this:

t_ignore = ' \t'

The characters in this string are ignored by the lexer. Use of this feature can speed
up parsing significantly since scanning will immediately proceed to the next token.

lex requires that the token returned by each rule has an attribute
t.type.  Other than this, rules are free to return any kind of token
object that they wish and may construct a new type of token object
from the attributes of t (provided the new object has the required
type attribute).

If illegal characters are encountered, the scanner executes the
function t_error(t) where t is a token representing the rest of the
string that hasn't been matched.  If this function isn't defined, a
LexError exception is raised.  The .text attribute of this exception
object contains the part of the string that wasn't matched.

The t.skip(n) method can be used to skip ahead n characters in the
input stream.  This is usually only used in the error handling rule.
For instance, the following rule would print an error message and
continue:

def t_error(t):
    print "Illegal character in input %s" % t.value[0]
    t.skip(1)

Of course, a nice scanner might wish to skip more than one character
if the input looks very corrupted.

The lex module defines a t.lineno attribute on each token that can be used
to track the current line number in the input.  The value of this
variable is not modified by lex so it is up to your lexer module
to correctly update its value depending on the lexical properties
of the input language.  To do this, you might write rules such as
the following:

def t_newline(t):
    r' \n+ '
    t.lineno += t.value.count("\n")

To initialize your lexer so that it can be used, simply call the lex.lex()
function in your rule file.  If there are any errors in your
specification, warning messages or an exception will be generated to
alert you to the problem.

To use the newly constructed lexer from another module, simply do
this:

    import lex
    import lexer
    lex.input("position = initial + rate*60")

    while 1:
        token = lex.token()       # Get a token
        if not token: break        # No more tokens
        ... do whatever ...

Assuming that the module 'lexer' has initialized lex as shown
above, parsing modules can safely import 'lex' without having
to import the rule file or any additional imformation about the
scanner you have defined.
"""    

# -----------------------------------------------------------------------------


__version__ = "2.1"

import re, types, sys, copy

# Available instance types.  This is used when lexers are defined by a class.
# It's a little funky because I want to preserve backwards compatibility
# with Python 2.0 where types.ObjectType is undefined.

try:
   _INSTANCETYPE = (types.InstanceType, types.ObjectType)
except AttributeError:
   _INSTANCETYPE = types.InstanceType
   class object: pass       # Note: needed if no new-style classes present

# Exception thrown when invalid token encountered and no default
class LexError(Exception):
    def __init__(self,message,s):
         self.args = (message,)
         self.text = s

# Token class
class LexToken(object):
    def __str__(self):
        return "LexToken(%s,%r,%d,%d)" % (self.type,self.value,self.lineno,self.lexpos)
    def __repr__(self):
        return str(self)
    def skip(self,n):
        try:
            self._skipn += n
        except AttributeError:
            self._skipn = n

# -----------------------------------------------------------------------------
# Lexer class
#
# This class encapsulates all of the methods and data associated with a lexer.
#
#    input()          -  Store a new string in the lexer
#    token()          -  Get the next token
# -----------------------------------------------------------------------------

class Lexer:
    def __init__(self):
        self.lexre = None             # Master regular expression. This is a list of 
                                      # tuples (re,findex) where re is a compiled
                                      # regular expression and findex is a list
                                      # mapping regex group numbers to rules

        self.lexreflags = 0           # Optional re compile flags
        self.lexdata = None           # Actual input data (as a string)
        self.lexpos = 0               # Current position in input text
        self.lexlen = 0               # Length of the input text
        self.lexerrorf = None         # Error rule (if any)
        self.lextokens = None         # List of valid tokens
        self.lexignore = None         # Ignored characters
        self.lexliterals = ""         # Literal characters that can be passed through
        self.lexmodule = None         # Module
        self.lineno = 1               # Current line number
        self.lexdebug = 0             # Debugging mode
        self.lexoptimize = 0          # Optimized mode

    def clone(self,object=None):
        c = Lexer()
        c.lexre = self.lexre
        c.lexreflags = self.lexreflags
        c.lexdata = self.lexdata
        c.lexpos = self.lexpos
        c.lexlen = self.lexlen
        c.lexerrorf = self.lexerrorf
        c.lextokens = self.lextokens
        c.lexignore = self.lexignore
        c.lexdebug = self.lexdebug
        c.lineno = self.lineno
        c.lexoptimize = self.lexoptimize
        c.lexliterals = self.lexliterals
        c.lexmodule   = self.lexmodule

        # If the object parameter has been supplied, it means we are attaching the
        # lexer to a new object.  In this case, we have to rebind the methods

        if object:
            newre = []
            for cre, findex in c.lexre:
                 # Loop over findex and adjust methods
                 newfindex = []
                 for f in findex:
                     if not f or not f[0]: 
                          newfindex.append(f)
                          continue
                     newfindex.append((getattr(object,f[0].__name__),f[1]))
                 newre.append((cre,newfindex))
            c.lexre = newre 
            c.lexmodule = object

        return c

    # ------------------------------------------------------------
    # input() - Push a new string into the lexer
    # ------------------------------------------------------------
    def input(self,s):
        if not (isinstance(s,types.StringType) or isinstance(s,types.UnicodeType)):
            raise ValueError, "Expected a string"
        self.lexdata = s
        self.lexpos = 0
        self.lexlen = len(s)

    # ------------------------------------------------------------
    # token() - Return the next token from the Lexer
    #
    # Note: This function has been carefully implemented to be as fast
    # as possible.  Don't make changes unless you really know what
    # you are doing
    # ------------------------------------------------------------
    def token(self):
        # Make local copies of frequently referenced attributes
        lexpos    = self.lexpos
        lexlen    = self.lexlen
        lexignore = self.lexignore
        lexdata   = self.lexdata

        while lexpos < lexlen:
            # This code provides some short-circuit code for whitespace, tabs, and other ignored characters
            if lexdata[lexpos] in lexignore:
                lexpos += 1
                continue

            # Look for a regular expression match
            for lexre,lexindexfunc in self.lexre:
                m = lexre.match(lexdata,lexpos)
                if not m: continue

                tok = LexToken()
                tok.value = m.group()
                tok.lineno = self.lineno
                tok.lexpos = lexpos
                tok.lexer = self

                lexpos = m.end()
                i = m.lastindex
                func,tok.type = lexindexfunc[i]
                self.lexpos = lexpos

                if not func:
                    return tok
                
                # If token is processed by a function, call it
                newtok = func(tok)
                
                # Every function must return a token, if nothing, we just move to next token
                if not newtok: 
                    lexpos = self.lexpos        # This is here in case user has updated lexpos.
                    break
                
                # Verify type of the token.  If not in the token map, raise an error
                if not self.lexoptimize:
                    if not self.lextokens.has_key(newtok.type):
                        raise LexError, ("%s:%d: Rule '%s' returned an unknown token type '%s'" % (
                            func.func_code.co_filename, func.func_code.co_firstlineno,
                            func.__name__, newtok.type),lexdata[lexpos:])

                return newtok
            else:
                # No match, see if in literals
                if lexdata[lexpos] in self.lexliterals:
                    tok = LexToken()
                    tok.value = lexdata[lexpos]
                    tok.lineno = self.lineno
                    tok.lexer = self
                    tok.type = tok.value
                    tok.lexpos = lexpos
                    self.lexpos = lexpos + 1
                    return tok
        
                # No match. Call t_error() if defined.
                if self.lexerrorf:
                    tok = LexToken()
                    tok.value = self.lexdata[lexpos:]
                    tok.lineno = self.lineno
                    tok.type = "error"
                    tok.lexer = self
                    tok.lexpos = lexpos
                    oldpos = lexpos
                    newtok = self.lexerrorf(tok)
                    lexpos += getattr(tok,"_skipn",0)
                    if oldpos == lexpos:
                        # Error method didn't change text position at all. This is an error.
                        self.lexpos = lexpos
                        raise LexError, ("Scanning error. Illegal character '%s'" % (lexdata[lexpos]), lexdata[lexpos:])
                    if not newtok: continue
                    self.lexpos = lexpos
                    return newtok

                self.lexpos = lexpos
                raise LexError, ("No match found", lexdata[lexpos:])

        self.lexpos = lexpos + 1
        if self.lexdata is None:
             raise RuntimeError, "No input string given with input()"
        return None
        
# -----------------------------------------------------------------------------
# validate_file()
#
# This checks to see if there are duplicated t_rulename() functions or strings
# in the parser input file.  This is done using a simple regular expression
# match on each line in the filename.
# -----------------------------------------------------------------------------

def validate_file(filename):
    import os.path
    base,ext = os.path.splitext(filename)
    if ext != '.py': return 1        # No idea what the file is. Return OK

    try:
        f = open(filename)
        lines = f.readlines()
        f.close()
    except IOError:
        return 1                       # Oh well

    fre = re.compile(r'\s*def\s+(t_[a-zA-Z_0-9]*)\(')
    sre = re.compile(r'\s*(t_[a-zA-Z_0-9]*)\s*=')
    counthash = { }
    linen = 1
    noerror = 1
    for l in lines:
        m = fre.match(l)
        if not m:
            m = sre.match(l)
        if m:
            name = m.group(1)
            prev = counthash.get(name)
            if not prev:
                counthash[name] = linen
            else:
                print "%s:%d: Rule %s redefined. Previously defined on line %d" % (filename,linen,name,prev)
                noerror = 0
        linen += 1
    return noerror

# -----------------------------------------------------------------------------
# _read_lextab(module)
#
# Reads lexer table from a lextab file instead of using introspection.
# -----------------------------------------------------------------------------

def _read_lextab(lexer, fdict, module):
    exec "import %s as lextab" % module
    lexre = []
    for regex,ltab in lextab._lexre:
        ftab = []
        for t in ltab:
            if t and t[0]:
                 ftab.append((fdict[t[0]], t[1]))
            else:
                 ftab.append(t)
        lexre.append((re.compile(regex, re.VERBOSE | lextab._lexreflags), ftab))
    lexer.lexre = lexre
    lexer.lexreflags = lextab._lexreflags
    lexer.lextokens = lextab._lextokens
    lexer.lexignore = lextab._lexignore
    if lextab._lexerrorf:
        lexer.lexerrorf = fdict[lextab._lexerrorf]
    lexer.lexliterals = lextab._lexliterals

# -----------------------------------------------------------------------------
# _form_master_re()
#
# This function takes a list of all of the regex components and attempts to
# form the master regular expression.  Given limitations in the Python re
# module, it may be necessary to break the master regex into separate expressions.
# -----------------------------------------------------------------------------

def _form_master_re(relist,reflags,ldict):
    if not relist: return []
    regex = "|".join(relist)
    try:
        lexre = re.compile(regex,re.VERBOSE | reflags)

        # Build the index to function map for the matching engine
        lexindexfunc = [ None ] * (max(lexre.groupindex.values())+1)
        for f,i in lexre.groupindex.items():
            handle = ldict.get(f,None)
            if type(handle) in (types.FunctionType, types.MethodType):
                lexindexfunc[i] = (handle,handle.__name__[2:])
            elif handle is not None:
                # If rule was specified as a string, we build an anonymous
                # callback function to carry out the action
                lexindexfunc[i] = (None,f[2:])
         
        return [(lexre,lexindexfunc)],[regex]
    except Exception,e:
        m = int(len(relist)/2)
        if m == 0: m = 1
        llist, lre = _form_master_re(relist[:m],reflags,ldict)
        rlist, rre = _form_master_re(relist[m:],reflags,ldict)
        return llist+rlist, lre+rre
 
# -----------------------------------------------------------------------------
# lex(module)
#
# Build all of the regular expression rules from definitions in the supplied module
# -----------------------------------------------------------------------------
def lex(module=None,object=None,debug=0,optimize=0,lextab="lextab",reflags=0):
    global lexer
    ldict = None
    regex_list = []
    error = 0
    files = { }
    lexobj = Lexer()
    lexobj.lexdebug = debug
    lexobj.lexoptimize = optimize
    global token,input
    
    if object: module = object

    if module:
        # User supplied a module object.
        if isinstance(module, types.ModuleType):
            ldict = module.__dict__
        elif isinstance(module, _INSTANCETYPE):
            _items = [(k,getattr(module,k)) for k in dir(module)]
            ldict = { }
            for (i,v) in _items:
                ldict[i] = v
        else:
            raise ValueError,"Expected a module or instance"
        lexobj.lexmodule = module
        
    else:
        # No module given.  We might be able to get information from the caller.
        try:
            raise RuntimeError
        except RuntimeError:
            e,b,t = sys.exc_info()
            f = t.tb_frame
            f = f.f_back           # Walk out to our calling function
            ldict = f.f_globals    # Grab its globals dictionary

    if optimize and lextab:
        try:
            _read_lextab(lexobj,ldict, lextab)
            if not lexobj.lexignore: lexobj.lexignore = ""            
            token = lexobj.token
            input = lexobj.input
            lexer = lexobj
            return lexobj
        
        except ImportError:
            pass
        
    # Get the tokens map
    if (module and isinstance(module,_INSTANCETYPE)):
        tokens = getattr(module,"tokens",None)
    else:
        try:
            tokens = ldict["tokens"]
        except KeyError:
            tokens = None
        
    if not tokens:
        raise SyntaxError,"lex: module does not define 'tokens'"
    if not (isinstance(tokens,types.ListType) or isinstance(tokens,types.TupleType)):
        raise SyntaxError,"lex: tokens must be a list or tuple."

    # Build a dictionary of valid token names
    lexobj.lextokens = { }
    if not optimize:

        # Utility function for verifying tokens
        def is_identifier(s):
            for c in s:
                if not (c.isalnum() or c == '_'): return 0
            return 1
        
        for n in tokens:
            if not is_identifier(n):
                print "lex: Bad token name '%s'" % n
                error = 1
            if lexobj.lextokens.has_key(n):
                print "lex: Warning. Token '%s' multiply defined." % n
            lexobj.lextokens[n] = None
    else:
        for n in tokens: lexobj.lextokens[n] = None
        

    if debug:
        print "lex: tokens = '%s'" % lexobj.lextokens.keys()

    # Get literals
    if (module and isinstance(module,_INSTANCETYPE)):
        literals = getattr(module,"literals","")
    else:
        try:
            literals = ldict["literals"]
        except KeyError:
            literals = ""

    lexobj.lexliterals = literals

    # Get a list of symbols with the t_ prefix
    tsymbols = [f for f in ldict.keys() if f[:2] == 't_']
    
    # Now build up a list of functions and a list of strings
    fsymbols = [ ]
    ssymbols = [ ]
    for f in tsymbols:
        if callable(ldict[f]):
            fsymbols.append(ldict[f])
        elif (isinstance(ldict[f], types.StringType) or isinstance(ldict[f],types.UnicodeType)):
            ssymbols.append((f,ldict[f]))
        else:
            print "lex: %s not defined as a function or string" % f
            error = 1
            
    # Sort the functions by line number
    fsymbols.sort(lambda x,y: cmp(x.func_code.co_firstlineno,y.func_code.co_firstlineno))

    # Sort the strings by regular expression length
    ssymbols.sort(lambda x,y: (len(x[1]) < len(y[1])) - (len(x[1]) > len(y[1])))
    
    # Check for non-empty symbols
    if len(fsymbols) == 0 and len(ssymbols) == 0:
        raise SyntaxError,"lex: no rules of the form t_rulename are defined."

    # Add all of the rules defined with actions first
    for f in fsymbols:
        
        line = f.func_code.co_firstlineno
        file = f.func_code.co_filename
        files[file] = None

        ismethod = isinstance(f, types.MethodType)

        if not optimize:
            nargs = f.func_code.co_argcount
            if ismethod:
                reqargs = 2
            else:
                reqargs = 1
            if nargs > reqargs:
                print "%s:%d: Rule '%s' has too many arguments." % (file,line,f.__name__)
                error = 1
                continue

            if nargs < reqargs:
                print "%s:%d: Rule '%s' requires an argument." % (file,line,f.__name__)
                error = 1
                continue

            if f.__name__ == 't_ignore':
                print "%s:%d: Rule '%s' must be defined as a string." % (file,line,f.__name__)
                error = 1
                continue
        
        if f.__name__ == 't_error':
            lexobj.lexerrorf = f
            continue

        if f.__doc__:
            if not optimize:
                try:
                    c = re.compile("(?P<%s>%s)" % (f.__name__,f.__doc__), re.VERBOSE | reflags)
                    if c.match(""):
                         print "%s:%d: Regular expression for rule '%s' matches empty string." % (file,line,f.__name__)
                         error = 1
                         continue
                except re.error,e:
                    print "%s:%d: Invalid regular expression for rule '%s'. %s" % (file,line,f.__name__,e)
                    if '#' in f.__doc__:
                         print "%s:%d. Make sure '#' in rule '%s' is escaped with '\\#'." % (file,line, f.__name__)                 
                    error = 1
                    continue

                if debug:
                    print "lex: Adding rule %s -> '%s'" % (f.__name__,f.__doc__)

            # Okay. The regular expression seemed okay.  Let's append it to the master regular
            # expression we're building
  
            regex_list.append("(?P<%s>%s)" % (f.__name__,f.__doc__))
        else:
            print "%s:%d: No regular expression defined for rule '%s'" % (file,line,f.__name__)

    # Now add all of the simple rules
    for name,r in ssymbols:

        if name == 't_ignore':
            lexobj.lexignore = r
            continue
        
        if not optimize:
            if name == 't_error':
                raise SyntaxError,"lex: Rule 't_error' must be defined as a function"
                error = 1
                continue
        
            if not lexobj.lextokens.has_key(name[2:]):
                print "lex: Rule '%s' defined for an unspecified token %s." % (name,name[2:])
                error = 1
                continue
            try:
                c = re.compile("(?P<%s>%s)" % (name,r),re.VERBOSE | reflags)
                if (c.match("")):
                     print "lex: Regular expression for rule '%s' matches empty string." % name
                     error = 1
                     continue
            except re.error,e:
                print "lex: Invalid regular expression for rule '%s'. %s" % (name,e)
                if '#' in r:
                     print "lex: Make sure '#' in rule '%s' is escaped with '\\#'." % name

                error = 1
                continue
            if debug:
                print "lex: Adding rule %s -> '%s'" % (name,r)
                
        regex_list.append("(?P<%s>%s)" % (name,r))

    if not optimize:
        for f in files.keys():
            if not validate_file(f):
                error = 1
    try:

        lexobj.lexre, re_groups = _form_master_re(regex_list,reflags,ldict)
        if debug:
            for i in range(len(re_groups)):
                 print "lex: regex[%d] = '%s'" % (i, re_groups[i])

        # If a lextab was specified, we create a file containing the precomputed
        # regular expression and index table
        
        if lextab and optimize:
            lt = open(lextab+".py","w")
            lt.write("# %s.py.  This file automatically created by PLY. Don't edit.\n" % lextab)

            tabre = []
            for i in range(len(re_groups)):
                 indexf = []
                 for t in lexobj.lexre[i][1]:
                     if t:
                         if t[0]:
                             indexf.append((t[0].__name__,t[1]))
                         else:
                             indexf.append((None,t[1]))
                     else:
                             indexf.append(None)
                 tabre.append((re_groups[i],indexf))
            
            lt.write("_lexre = %s\n" % repr(tabre))

            # Create an alternative lexre representation

            lt.write("_lexreflags = %d\n" % reflags)
            lt.write("_lextokens = %s\n" % repr(lexobj.lextokens))
            lt.write("_lexignore = %s\n" % repr(lexobj.lexignore))
            if (lexobj.lexerrorf):
                lt.write("_lexerrorf = %s\n" % repr(lexobj.lexerrorf.__name__))
            else:
                lt.write("_lexerrorf = None\n")
            lt.write("_lexliterals = %s\n" % repr(lexobj.lexliterals))
            lt.close()
        
    except re.error,e:
        print "lex: Fatal error. Unable to compile regular expression rules. %s" % e
        error = 1
    if error:
        raise SyntaxError,"lex: Unable to build lexer."
    if not lexobj.lexerrorf:
        print "lex: Warning. no t_error rule is defined."

    if not lexobj.lexignore: lexobj.lexignore = ""
    
    # Create global versions of the token() and input() functions
    token = lexobj.token
    input = lexobj.input
    lexer = lexobj
    
    return lexobj

# -----------------------------------------------------------------------------
# run()
#
# This runs the lexer as a main program
# -----------------------------------------------------------------------------

def runmain(lexer=None,data=None):
    if not data:
        try:
            filename = sys.argv[1]
            f = open(filename)
            data = f.read()
            f.close()
        except IndexError:
            print "Reading from standard input (type EOF to end):"
            data = sys.stdin.read()

    if lexer:
        _input = lexer.input
    else:
        _input = input
    _input(data)
    if lexer:
        _token = lexer.token
    else:
        _token = token
        
    while 1:
        tok = _token()
        if not tok: break
        print "(%s,%r,%d,%d)" % (tok.type, tok.value, tok.lineno,tok.lexpos)
        

# -----------------------------------------------------------------------------
# @TOKEN(regex)
#
# This decorator function can be used to set the regex expression on a function
# when its docstring might need to be set in an alternative way
# -----------------------------------------------------------------------------

def TOKEN(r):
    def set_doc(f):
        f.__doc__ = r
        return f
    return set_doc