Lib/re.py

   1 #
   2 # Secret Labs' Regular Expression Engine
   3 #
   4 # re-compatible interface for the sre matching engine
   5 #
   6 # Copyright (c) 1998-2001 by Secret Labs AB.  All rights reserved.
   7 #
   8 # This version of the SRE library can be redistributed under CNRI's
   9 # Python 1.6 license.  For any other use, please contact Secret Labs
  10 # AB (info@pythonware.com).
  11 #
  12 # Portions of this engine have been developed in cooperation with
  13 # CNRI.  Hewlett-Packard provided funding for 1.6 integration and
  14 # other compatibility work.
  15 #
  16
  17 r"""Support for regular expressions (RE).
  18
  19 This module provides regular expression matching operations similar to
  20 those found in Perl.  It supports both 8-bit and Unicode strings; both
  21 the pattern and the strings being processed can contain null bytes and
  22 characters outside the US ASCII range.
  23
  24 Regular expressions can contain both special and ordinary characters.
  25 Most ordinary characters, like "A", "a", or "0", are the simplest
  26 regular expressions; they simply match themselves.  You can
  27 concatenate ordinary characters, so last matches the string 'last'.
  28
  29 The special characters are:
  30     "."      Matches any character except a newline.
  31     "^"      Matches the start of the string.
  32     "$"      Matches the end of the string or just before the newline at
  33              the end of the string.
  34     "*"      Matches 0 or more (greedy) repetitions of the preceding RE.
  35              Greedy means that it will match as many repetitions as possible.
  36     "+"      Matches 1 or more (greedy) repetitions of the preceding RE.
  37     "?"      Matches 0 or 1 (greedy) of the preceding RE.
  38     *?,+?,?? Non-greedy versions of the previous three special characters.
  39     {m,n}    Matches from m to n repetitions of the preceding RE.
  40     {m,n}?   Non-greedy version of the above.
  41     "\\"      Either escapes special characters or signals a special sequence.
  42     []       Indicates a set of characters.
  43              A "^" as the first character indicates a complementing set.
  44     "|"      A|B, creates an RE that will match either A or B.
  45     (...)    Matches the RE inside the parentheses.
  46              The contents can be retrieved or matched later in the string.
  47     (?iLmsux) Set the I, L, M, S, U, or X flag for the RE (see below).
  48     (?:...)  Non-grouping version of regular parentheses.
  49     (?P<name>...) The substring matched by the group is accessible by name.
  50     (?P=name)     Matches the text matched earlier by the group named name.
  51     (?#...)  A comment; ignored.
  52     (?=...)  Matches if ... matches next, but doesn't consume the string.
  53     (?!...)  Matches if ... doesn't match next.
  54
  55 The special sequences consist of "\\" and a character from the list
  56 below.  If the ordinary character is not on the list, then the
  57 resulting RE will match the second character.
  58     \number  Matches the contents of the group of the same number.
  59     \A       Matches only at the start of the string.
  60     \Z       Matches only at the end of the string.
  61     \b       Matches the empty string, but only at the start or end of a word.
  62     \B       Matches the empty string, but not at the start or end of a word.
  63     \d       Matches any decimal digit; equivalent to the set [0-9].
  64     \D       Matches any non-digit character; equivalent to the set [^0-9].
  65     \s       Matches any whitespace character; equivalent to [ \t\n\r\f\v].
  66     \S       Matches any non-whitespace character; equiv. to [^ \t\n\r\f\v].
  67     \w       Matches any alphanumeric character; equivalent to [a-zA-Z0-9_].
  68              With LOCALE, it will match the set [0-9_] plus characters defined
  69              as letters for the current locale.
  70     \W       Matches the complement of \w.
  71     \\       Matches a literal backslash.
  72
  73 This module exports the following functions:
  74     match    Match a regular expression pattern to the beginning of a string.
  75     search   Search a string for the presence of a pattern.
  76     sub      Substitute occurrences of a pattern found in a string.
  77     subn     Same as sub, but also return the number of substitutions made.
  78     split    Split a string by the occurrences of a pattern.
  79     findall  Find all occurrences of a pattern in a string.
  80     compile  Compile a pattern into a RegexObject.
  81     purge    Clear the regular expression cache.
  82     escape   Backslash all non-alphanumerics in a string.
  83
  84 Some of the functions in this module takes flags as optional parameters:
  85     I  IGNORECASE  Perform case-insensitive matching.
  86     L  LOCALE      Make \w, \W, \b, \B, dependent on the current locale.
  87     M  MULTILINE   "^" matches the beginning of lines (after a newline)
  88                    as well as the string.
  89                    "$" matches the end of lines (before a newline) as well
  90                    as the end of the string.
  91     S  DOTALL      "." matches any character at all, including the newline.
  92     X  VERBOSE     Ignore whitespace and comments for nicer looking RE's.
  93     U  UNICODE     Make \w, \W, \b, \B, dependent on the Unicode locale.
  94
  95 This module also defines an exception 'error'.
  96
  97 """
  98
  99 import sys
 100 import sre_compile
 101 import sre_parse
 102
 103 # public symbols
 104 __all__ = [ "match", "search", "sub", "subn", "split", "findall",
 105     "compile", "purge", "template", "escape", "I", "L", "M", "S", "X",
 106     "U", "IGNORECASE", "LOCALE", "MULTILINE", "DOTALL", "VERBOSE",
 107     "UNICODE", "error" ]
 108
 109 __version__ = "2.2.1"
 110
 111 # flags
 112 I = IGNORECASE = sre_compile.SRE_FLAG_IGNORECASE # ignore case
 113 L = LOCALE = sre_compile.SRE_FLAG_LOCALE # assume current 8-bit locale
 114 U = UNICODE = sre_compile.SRE_FLAG_UNICODE # assume unicode locale
 115 M = MULTILINE = sre_compile.SRE_FLAG_MULTILINE # make anchors look for newline
 116 S = DOTALL = sre_compile.SRE_FLAG_DOTALL # make dot match newline
 117 X = VERBOSE = sre_compile.SRE_FLAG_VERBOSE # ignore whitespace and comments
 118
 119 # sre extensions (experimental, don't rely on these)
 120 T = TEMPLATE = sre_compile.SRE_FLAG_TEMPLATE # disable backtracking
 121 DEBUG = sre_compile.SRE_FLAG_DEBUG # dump pattern after compilation
 122
 123 # sre exception
 124 error = sre_compile.error
 125
 126 # --------------------------------------------------------------------
 127 # public interface
 128
 129 def match(pattern, string, flags=0):
 130     """Try to apply the pattern at the start of the string, returning
 131     a match object, or None if no match was found."""
 132     return _compile(pattern, flags).match(string)
 133
 134 def search(pattern, string, flags=0):
 135     """Scan through string looking for a match to the pattern, returning
 136     a match object, or None if no match was found."""
 137     return _compile(pattern, flags).search(string)
 138
 139 def sub(pattern, repl, string, count=0):
 140     """Return the string obtained by replacing the leftmost
 141     non-overlapping occurrences of the pattern in string by the
 142     replacement repl.  repl can be either a string or a callable;
 143     if a callable, it's passed the match object and must return
 144     a replacement string to be used."""
 145     return _compile(pattern, 0).sub(repl, string, count)
 146
 147 def subn(pattern, repl, string, count=0):
 148     """Return a 2-tuple containing (new_string, number).
 149     new_string is the string obtained by replacing the leftmost
 150     non-overlapping occurrences of the pattern in the source
 151     string by the replacement repl.  number is the number of
 152     substitutions that were made. repl can be either a string or a
 153     callable; if a callable, it's passed the match object and must
 154     return a replacement string to be used."""
 155     return _compile(pattern, 0).subn(repl, string, count)
 156
 157 def split(pattern, string, maxsplit=0):
 158     """Split the source string by the occurrences of the pattern,
 159     returning a list containing the resulting substrings."""
 160     return _compile(pattern, 0).split(string, maxsplit)
 161
 162 def findall(pattern, string, flags=0):
 163     """Return a list of all non-overlapping matches in the string.
 164
 165     If one or more groups are present in the pattern, return a
 166     list of groups; this will be a list of tuples if the pattern
 167     has more than one group.
 168
 169     Empty matches are included in the result."""
 170     return _compile(pattern, flags).findall(string)
 171
 172 if sys.hexversion >= 0x02020000:
 173     __all__.append("finditer")
 174     def finditer(pattern, string, flags=0):
 175         """Return an iterator over all non-overlapping matches in the
 176         string.  For each match, the iterator returns a match object.
 177
 178         Empty matches are included in the result."""
 179         return _compile(pattern, flags).finditer(string)
 180
 181 def compile(pattern, flags=0):
 182     "Compile a regular expression pattern, returning a pattern object."
 183     return _compile(pattern, flags)
 184
 185 def purge():
 186     "Clear the regular expression cache"
 187     _cache.clear()
 188     _cache_repl.clear()
 189
 190 def template(pattern, flags=0):
 191     "Compile a template pattern, returning a pattern object"
 192     return _compile(pattern, flags|T)
 193
 194 _alphanum = {}
 195 for c in 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890':
 196     _alphanum[c] = 1
 197 del c
 198
 199 def escape(pattern):
 200     "Escape all non-alphanumeric characters in pattern."
 201     s = list(pattern)
 202     alphanum = _alphanum
 203     for i in range(len(pattern)):
 204         c = pattern[i]
 205         if c not in alphanum:
 206             if c == "\000":
 207                 s[i] = "\\000"
 208             else:
 209                 s[i] = "\\" + c
 210     return pattern[:0].join(s)
 211
 212 # --------------------------------------------------------------------
 213 # internals
 214
 215 _cache = {}
 216 _cache_repl = {}
 217
 218 _pattern_type = type(sre_compile.compile("", 0))
 219
 220 _MAXCACHE = 100
 221
 222 def _compile(*key):
 223     # internal: compile pattern
 224     cachekey = (type(key[0]),) + key
 225     p = _cache.get(cachekey)
 226     if p is not None:
 227         return p
 228     pattern, flags = key
 229     if isinstance(pattern, _pattern_type):
 230         if flags:
 231             raise ValueError('Cannot process flags argument with a compiled pattern')
 232         return pattern
 233     if not sre_compile.isstring(pattern):
 234         raise TypeError, "first argument must be string or compiled pattern"
 235     try:
 236         p = sre_compile.compile(pattern, flags)
 237     except error, v:
 238         raise error, v # invalid expression
 239     if len(_cache) >= _MAXCACHE:
 240         _cache.clear()
 241     _cache[cachekey] = p
 242     return p
 243
 244 def _compile_repl(*key):
 245     # internal: compile replacement pattern
 246     p = _cache_repl.get(key)
 247     if p is not None:
 248         return p
 249     repl, pattern = key
 250     try:
 251         p = sre_parse.parse_template(repl, pattern)
 252     except error, v:
 253         raise error, v # invalid expression
 254     if len(_cache_repl) >= _MAXCACHE:
 255         _cache_repl.clear()
 256     _cache_repl[key] = p
 257     return p
 258
 259 def _expand(pattern, match, template):
 260     # internal: match.expand implementation hook
 261     template = sre_parse.parse_template(template, pattern)
 262     return sre_parse.expand_template(template, match)
 263
 264 def _subx(pattern, template):
 265     # internal: pattern.sub/subn implementation helper
 266     template = _compile_repl(template, pattern)
 267     if not template[0] and len(template[1]) == 1:
 268         # literal replacement
 269         return template[1][0]
 270     def filter(match, template=template):
 271         return sre_parse.expand_template(template, match)
 272     return filter
 273
 274 # register myself for pickling
 275
 276 import copy_reg
 277
 278 def _pickle(p):
 279     return _compile, (p.pattern, p.flags)
 280
 281 copy_reg.pickle(_pattern_type, _pickle, _compile)
 282
 283 # --------------------------------------------------------------------
 284 # experimental stuff (see python-dev discussions for details)
 285
 286 class Scanner:
 287     def __init__(self, lexicon, flags=0):
 288         from sre_constants import BRANCH, SUBPATTERN
 289         self.lexicon = lexicon
 290         # combine phrases into a compound pattern
 291         p = []
 292         s = sre_parse.Pattern()
 293         s.flags = flags
 294         for phrase, action in lexicon:
 295             p.append(sre_parse.SubPattern(s, [
 296                 (SUBPATTERN, (len(p)+1, sre_parse.parse(phrase, flags))),
 297                 ]))
 298         p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
 299         s.groups = len(p)
 300         self.scanner = sre_compile.compile(p)
 301     def scan(self, string):
 302         result = []
 303         append = result.append
 304         match = self.scanner.scanner(string).match
 305         i = 0
 306         while 1:
 307             m = match()
 308             if not m:
 309                 break
 310             j = m.end()
 311             if i == j:
 312                 break
 313             action = self.lexicon[m.lastindex-1][1]
 314             if callable(action):
 315                 self.match = m
 316                 action = action(self, m.group())
 317             if action is not None:
 318                 append(action)
 319             i = j
 320         return result, string[i:]