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     (?aiLmsux) Set the A, 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     (?<=...) Matches if preceded by ... (must be fixed length).
  55     (?<!...) Matches if not preceded by ... (must be fixed length).
  56     (?(id/name)yes|no) Matches yes pattern if the group with id/name matched,
  57                        the (optional) no pattern otherwise.
  58 
  59 The special sequences consist of "\\" and a character from the list
  60 below.  If the ordinary character is not on the list, then the
  61 resulting RE will match the second character.
  62     \number  Matches the contents of the group of the same number.
  63     \A       Matches only at the start of the string.
  64     \Z       Matches only at the end of the string.
  65     \b       Matches the empty string, but only at the start or end of a word.
  66     \B       Matches the empty string, but not at the start or end of a word.
  67     \d       Matches any decimal digit; equivalent to the set [0-9] in
  68              bytes patterns or string patterns with the ASCII flag.
  69              In string patterns without the ASCII flag, it will match the whole
  70              range of Unicode digits.
  71     \D       Matches any non-digit character; equivalent to [^\d].
  72     \s       Matches any whitespace character; equivalent to [ \t\n\r\f\v] in
  73              bytes patterns or string patterns with the ASCII flag.
  74              In string patterns without the ASCII flag, it will match the whole
  75              range of Unicode whitespace characters.
  76     \S       Matches any non-whitespace character; equivalent to [^\s].
  77     \w       Matches any alphanumeric character; equivalent to [a-zA-Z0-9_]
  78              in bytes patterns or string patterns with the ASCII flag.
  79              In string patterns without the ASCII flag, it will match the
  80              range of Unicode alphanumeric characters (letters plus digits
  81              plus underscore).
  82              With LOCALE, it will match the set [0-9_] plus characters defined
  83              as letters for the current locale.
  84     \W       Matches the complement of \w.
  85     \\       Matches a literal backslash.
  86 
  87 This module exports the following functions:
  88     match    Match a regular expression pattern to the beginning of a string.
  89     search   Search a string for the presence of a pattern.
  90     sub      Substitute occurrences of a pattern found in a string.
  91     subn     Same as sub, but also return the number of substitutions made.
  92     split    Split a string by the occurrences of a pattern.
  93     findall  Find all occurrences of a pattern in a string.
  94     finditer Return an iterator yielding a match object for each match.
  95     compile  Compile a pattern into a RegexObject.
  96     purge    Clear the regular expression cache.
  97     escape   Backslash all non-alphanumerics in a string.
  98 
  99 Some of the functions in this module takes flags as optional parameters:
 100     A  ASCII       For string patterns, make \w, \W, \b, \B, \d, \D
 101                    match the corresponding ASCII character categories
 102                    (rather than the whole Unicode categories, which is the
 103                    default).
 104                    For bytes patterns, this flag is the only available
 105                    behaviour and needn't be specified.
 106     I  IGNORECASE  Perform case-insensitive matching.
 107     L  LOCALE      Make \w, \W, \b, \B, dependent on the current locale.
 108     M  MULTILINE   "^" matches the beginning of lines (after a newline)
 109                    as well as the string.
 110                    "$" matches the end of lines (before a newline) as well
 111                    as the end of the string.
 112     S  DOTALL      "." matches any character at all, including the newline.
 113     X  VERBOSE     Ignore whitespace and comments for nicer looking RE's.
 114     U  UNICODE     For compatibility only. Ignored for string patterns (it
 115                    is the default), and forbidden for bytes patterns.
 116 
 117 This module also defines an exception 'error'.
 118 
 119 """
 120 
 121 import sys
 122 import sre_compile
 123 import sre_parse
 124 
 125 # public symbols
 126 __all__ = [ "match", "search", "sub", "subn", "split", "findall",
 127     "compile", "purge", "template", "escape", "A", "I", "L", "M", "S", "X",
 128     "U", "ASCII", "IGNORECASE", "LOCALE", "MULTILINE", "DOTALL", "VERBOSE",
 129     "UNICODE", "error" ]
 130 
 131 __version__ = "2.2.1"
 132 
 133 # flags
 134 A = ASCII = sre_compile.SRE_FLAG_ASCII # assume ascii "locale"
 135 I = IGNORECASE = sre_compile.SRE_FLAG_IGNORECASE # ignore case
 136 L = LOCALE = sre_compile.SRE_FLAG_LOCALE # assume current 8-bit locale
 137 U = UNICODE = sre_compile.SRE_FLAG_UNICODE # assume unicode "locale"
 138 M = MULTILINE = sre_compile.SRE_FLAG_MULTILINE # make anchors look for newline
 139 S = DOTALL = sre_compile.SRE_FLAG_DOTALL # make dot match newline
 140 X = VERBOSE = sre_compile.SRE_FLAG_VERBOSE # ignore whitespace and comments
 141 
 142 # sre extensions (experimental, don't rely on these)
 143 T = TEMPLATE = sre_compile.SRE_FLAG_TEMPLATE # disable backtracking
 144 DEBUG = sre_compile.SRE_FLAG_DEBUG # dump pattern after compilation
 145 
 146 # sre exception
 147 error = sre_compile.error
 148 
 149 # --------------------------------------------------------------------
 150 # public interface
 151 
 152 def match(pattern, string, flags=0):
 153     """Try to apply the pattern at the start of the string, returning
 154     a match object, or None if no match was found."""
 155     return _compile(pattern, flags).match(string)
 156 
 157 def search(pattern, string, flags=0):
 158     """Scan through string looking for a match to the pattern, returning
 159     a match object, or None if no match was found."""
 160     return _compile(pattern, flags).search(string)
 161 
 162 def sub(pattern, repl, string, count=0, flags=0):
 163     """Return the string obtained by replacing the leftmost
 164     non-overlapping occurrences of the pattern in string by the
 165     replacement repl.  repl can be either a string or a callable;
 166     if a string, backslash escapes in it are processed.  If it is
 167     a callable, it's passed the match object and must return
 168     a replacement string to be used."""
 169     return _compile(pattern, flags).sub(repl, string, count)
 170 
 171 def subn(pattern, repl, string, count=0, flags=0):
 172     """Return a 2-tuple containing (new_string, number).
 173     new_string is the string obtained by replacing the leftmost
 174     non-overlapping occurrences of the pattern in the source
 175     string by the replacement repl.  number is the number of
 176     substitutions that were made. repl can be either a string or a
 177     callable; if a string, backslash escapes in it are processed.
 178     If it is a callable, it's passed the match object and must
 179     return a replacement string to be used."""
 180     return _compile(pattern, flags).subn(repl, string, count)
 181 
 182 def split(pattern, string, maxsplit=0, flags=0):
 183     """Split the source string by the occurrences of the pattern,
 184     returning a list containing the resulting substrings.  If
 185     capturing parentheses are used in pattern, then the text of all
 186     groups in the pattern are also returned as part of the resulting
 187     list.  If maxsplit is nonzero, at most maxsplit splits occur,
 188     and the remainder of the string is returned as the final element
 189     of the list."""
 190     return _compile(pattern, flags).split(string, maxsplit)
 191 
 192 def findall(pattern, string, flags=0):
 193     """Return a list of all non-overlapping matches in the string.
 194 
 195     If one or more capturing groups are present in the pattern, return
 196     a list of groups; this will be a list of tuples if the pattern
 197     has more than one group.
 198 
 199     Empty matches are included in the result."""
 200     return _compile(pattern, flags).findall(string)
 201 
 202 if sys.hexversion >= 0x02020000:
 203     __all__.append("finditer")
 204     def finditer(pattern, string, flags=0):
 205         """Return an iterator over all non-overlapping matches in the
 206         string.  For each match, the iterator returns a match object.
 207 
 208         Empty matches are included in the result."""
 209         return _compile(pattern, flags).finditer(string)
 210 
 211 def compile(pattern, flags=0):
 212     "Compile a regular expression pattern, returning a pattern object."
 213     return _compile(pattern, flags)
 214 
 215 def purge():
 216     "Clear the regular expression caches"
 217     _cache.clear()
 218     _cache_repl.clear()
 219 
 220 def template(pattern, flags=0):
 221     "Compile a template pattern, returning a pattern object"
 222     return _compile(pattern, flags|T)
 223 
 224 _alphanum_str = frozenset(
 225     "_abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890")
 226 _alphanum_bytes = frozenset(
 227     b"_abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890")
 228 
 229 def escape(pattern):
 230     """
 231     Escape all the characters in pattern except ASCII letters, numbers and '_'.
 232     """
 233     if isinstance(pattern, str):
 234         alphanum = _alphanum_str
 235         s = list(pattern)
 236         for i, c in enumerate(pattern):
 237             if c not in alphanum:
 238                 if c == "\000":
 239                     s[i] = "\\000"
 240                 else:
 241                     s[i] = "\\" + c
 242         return "".join(s)
 243     else:
 244         alphanum = _alphanum_bytes
 245         s = []
 246         esc = ord(b"\\")
 247         for c in pattern:
 248             if c in alphanum:
 249                 s.append(c)
 250             else:
 251                 if c == 0:
 252                     s.extend(b"\\000")
 253                 else:
 254                     s.append(esc)
 255                     s.append(c)
 256         return bytes(s)
 257 
 258 # --------------------------------------------------------------------
 259 # internals
 260 
 261 _cache = {}
 262 _cache_repl = {}
 263 
 264 _pattern_type = type(sre_compile.compile("", 0))
 265 
 266 _MAXCACHE = 512
 267 def _compile(pattern, flags):
 268     # internal: compile pattern
 269     try:
 270         return _cache[type(pattern), pattern, flags]
 271     except KeyError:
 272         pass
 273     if isinstance(pattern, _pattern_type):
 274         if flags:
 275             raise ValueError(
 276                 "Cannot process flags argument with a compiled pattern")
 277         return pattern
 278     if not sre_compile.isstring(pattern):
 279         raise TypeError("first argument must be string or compiled pattern")
 280     p = sre_compile.compile(pattern, flags)
 281     if len(_cache) >= _MAXCACHE:
 282         _cache.clear()
 283     _cache[type(pattern), pattern, flags] = p
 284     return p
 285 
 286 def _compile_repl(repl, pattern):
 287     # internal: compile replacement pattern
 288     try:
 289         return _cache_repl[repl, pattern]
 290     except KeyError:
 291         pass
 292     p = sre_parse.parse_template(repl, pattern)
 293     if len(_cache_repl) >= _MAXCACHE:
 294         _cache_repl.clear()
 295     _cache_repl[repl, pattern] = p
 296     return p
 297 
 298 def _expand(pattern, match, template):
 299     # internal: match.expand implementation hook
 300     template = sre_parse.parse_template(template, pattern)
 301     return sre_parse.expand_template(template, match)
 302 
 303 def _subx(pattern, template):
 304     # internal: pattern.sub/subn implementation helper
 305     template = _compile_repl(template, pattern)
 306     if not template[0] and len(template[1]) == 1:
 307         # literal replacement
 308         return template[1][0]
 309     def filter(match, template=template):
 310         return sre_parse.expand_template(template, match)
 311     return filter
 312 
 313 # register myself for pickling
 314 
 315 import copyreg
 316 
 317 def _pickle(p):
 318     return _compile, (p.pattern, p.flags)
 319 
 320 copyreg.pickle(_pattern_type, _pickle, _compile)
 321 
 322 # --------------------------------------------------------------------
 323 # experimental stuff (see python-dev discussions for details)
 324 
 325 class Scanner:
 326     def __init__(self, lexicon, flags=0):
 327         from sre_constants import BRANCH, SUBPATTERN
 328         self.lexicon = lexicon
 329         # combine phrases into a compound pattern
 330         p = []
 331         s = sre_parse.Pattern()
 332         s.flags = flags
 333         for phrase, action in lexicon:
 334             p.append(sre_parse.SubPattern(s, [
 335                 (SUBPATTERN, (len(p)+1, sre_parse.parse(phrase, flags))),
 336                 ]))
 337         s.groups = len(p)+1
 338         p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
 339         self.scanner = sre_compile.compile(p)
 340     def scan(self, string):
 341         result = []
 342         append = result.append
 343         match = self.scanner.scanner(string).match
 344         i = 0
 345         while 1:
 346             m = match()
 347             if not m:
 348                 break
 349             j = m.end()
 350             if i == j:
 351                 break
 352             action = self.lexicon[m.lastindex-1][1]
 353             if callable(action):
 354                 self.match = m
 355                 action = action(self, m.group())
 356             if action is not None:
 357                 append(action)
 358             i = j
 359         return result, string[i:]