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