mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-02-07 23:13:34 +00:00
311 lines
10 KiB
Python
311 lines
10 KiB
Python
#
|
|
# Secret Labs' Regular Expression Engine
|
|
#
|
|
# various symbols used by the regular expression engine.
|
|
# run this script to update the _sre include files!
|
|
#
|
|
# Copyright (c) 1998-2001 by Secret Labs AB. All rights reserved.
|
|
#
|
|
# See the sre.py file for information on usage and redistribution.
|
|
#
|
|
|
|
"""Internal support module for sre"""
|
|
|
|
# update when constants are added or removed
|
|
|
|
MAGIC = 20140917
|
|
|
|
from _sre import MAXREPEAT, MAXGROUPS
|
|
|
|
# SRE standard exception (access as sre.error)
|
|
# should this really be here?
|
|
|
|
class error(Exception):
|
|
"""Exception raised for invalid regular expressions.
|
|
|
|
Attributes:
|
|
|
|
msg: The unformatted error message
|
|
pattern: The regular expression pattern
|
|
pos: The index in the pattern where compilation failed (may be None)
|
|
lineno: The line corresponding to pos (may be None)
|
|
colno: The column corresponding to pos (may be None)
|
|
"""
|
|
|
|
def __init__(self, msg, pattern=None, pos=None):
|
|
self.msg = msg
|
|
self.pattern = pattern
|
|
self.pos = pos
|
|
if pattern is not None and pos is not None:
|
|
msg = '%s at position %d' % (msg, pos)
|
|
if isinstance(pattern, str):
|
|
newline = '\n'
|
|
else:
|
|
newline = b'\n'
|
|
self.lineno = pattern.count(newline, 0, pos) + 1
|
|
self.colno = pos - pattern.rfind(newline, 0, pos)
|
|
if newline in pattern:
|
|
msg = '%s (line %d, column %d)' % (msg, self.lineno, self.colno)
|
|
else:
|
|
self.lineno = self.colno = None
|
|
super().__init__(msg)
|
|
|
|
|
|
class _NamedIntConstant(int):
|
|
def __new__(cls, value, name):
|
|
self = super(_NamedIntConstant, cls).__new__(cls, value)
|
|
self.name = name
|
|
return self
|
|
|
|
def __str__(self):
|
|
return self.name
|
|
|
|
__repr__ = __str__
|
|
|
|
MAXREPEAT = _NamedIntConstant(MAXREPEAT, 'MAXREPEAT')
|
|
|
|
FAILURE = _NamedIntConstant(0, 'FAILURE')
|
|
SUCCESS = _NamedIntConstant(1, 'SUCCESS')
|
|
ANY = _NamedIntConstant(2, 'ANY')
|
|
ANY_ALL = _NamedIntConstant(3, 'ANY_ALL')
|
|
ASSERT = _NamedIntConstant(4, 'ASSERT')
|
|
ASSERT_NOT = _NamedIntConstant(5, 'ASSERT_NOT')
|
|
AT = _NamedIntConstant(6, 'AT')
|
|
BRANCH = _NamedIntConstant(7, 'BRANCH')
|
|
CALL = _NamedIntConstant(8, 'CALL')
|
|
CATEGORY = _NamedIntConstant(9, 'CATEGORY')
|
|
CHARSET = _NamedIntConstant(10, 'CHARSET')
|
|
BIGCHARSET = _NamedIntConstant(11, 'BIGCHARSET')
|
|
GROUPREF = _NamedIntConstant(12, 'GROUPREF')
|
|
GROUPREF_EXISTS = _NamedIntConstant(13, 'GROUPREF_EXISTS')
|
|
GROUPREF_IGNORE = _NamedIntConstant(14, 'GROUPREF_IGNORE')
|
|
IN = _NamedIntConstant(15, 'IN')
|
|
IN_IGNORE = _NamedIntConstant(16, 'IN_IGNORE')
|
|
INFO = _NamedIntConstant(17, 'INFO')
|
|
JUMP = _NamedIntConstant(18, 'JUMP')
|
|
LITERAL = _NamedIntConstant(19, 'LITERAL')
|
|
LITERAL_IGNORE = _NamedIntConstant(20, 'LITERAL_IGNORE')
|
|
MARK = _NamedIntConstant(21, 'MARK')
|
|
MAX_UNTIL = _NamedIntConstant(22, 'MAX_UNTIL')
|
|
MIN_UNTIL = _NamedIntConstant(23, 'MIN_UNTIL')
|
|
NOT_LITERAL = _NamedIntConstant(24, 'NOT_LITERAL')
|
|
NOT_LITERAL_IGNORE = _NamedIntConstant(25, 'NOT_LITERAL_IGNORE')
|
|
NEGATE = _NamedIntConstant(26, 'NEGATE')
|
|
RANGE = _NamedIntConstant(27, 'RANGE')
|
|
REPEAT = _NamedIntConstant(28, 'REPEAT')
|
|
REPEAT_ONE = _NamedIntConstant(29, 'REPEAT_ONE')
|
|
SUBPATTERN = _NamedIntConstant(30, 'SUBPATTERN')
|
|
MIN_REPEAT_ONE = _NamedIntConstant(31, 'MIN_REPEAT_ONE')
|
|
RANGE_IGNORE = _NamedIntConstant(32, 'RANGE_IGNORE')
|
|
MIN_REPEAT = _NamedIntConstant(33, 'MIN_REPEAT')
|
|
MAX_REPEAT = _NamedIntConstant(34, 'MAX_REPEAT')
|
|
|
|
OPCODES = [
|
|
FAILURE,
|
|
SUCCESS,
|
|
ANY,
|
|
ANY_ALL,
|
|
ASSERT,
|
|
ASSERT_NOT,
|
|
AT,
|
|
BRANCH,
|
|
CALL,
|
|
CATEGORY,
|
|
CHARSET,
|
|
BIGCHARSET,
|
|
GROUPREF,
|
|
GROUPREF_EXISTS,
|
|
GROUPREF_IGNORE,
|
|
IN,
|
|
IN_IGNORE,
|
|
INFO,
|
|
JUMP,
|
|
LITERAL,
|
|
LITERAL_IGNORE,
|
|
MARK,
|
|
MAX_UNTIL,
|
|
MIN_UNTIL,
|
|
NOT_LITERAL,
|
|
NOT_LITERAL_IGNORE,
|
|
NEGATE,
|
|
RANGE,
|
|
REPEAT,
|
|
REPEAT_ONE,
|
|
SUBPATTERN,
|
|
MIN_REPEAT_ONE,
|
|
RANGE_IGNORE,
|
|
]
|
|
|
|
AT_BEGINNING = _NamedIntConstant( 0, 'AT_BEGINNING')
|
|
AT_BEGINNING_LINE = _NamedIntConstant( 1, 'AT_BEGINNING_LINE')
|
|
AT_BEGINNING_STRING = _NamedIntConstant( 2, 'AT_BEGINNING_STRING')
|
|
AT_BOUNDARY = _NamedIntConstant( 3, 'AT_BOUNDARY')
|
|
AT_NON_BOUNDARY = _NamedIntConstant( 4, 'AT_NON_BOUNDARY')
|
|
AT_END = _NamedIntConstant( 5, 'AT_END')
|
|
AT_END_LINE = _NamedIntConstant( 6, 'AT_END_LINE')
|
|
AT_END_STRING = _NamedIntConstant( 7, 'AT_END_STRING')
|
|
AT_LOC_BOUNDARY = _NamedIntConstant( 8, 'AT_LOC_BOUNDARY')
|
|
AT_LOC_NON_BOUNDARY = _NamedIntConstant( 9, 'AT_LOC_NON_BOUNDARY')
|
|
AT_UNI_BOUNDARY = _NamedIntConstant(10, 'AT_UNI_BOUNDARY')
|
|
AT_UNI_NON_BOUNDARY = _NamedIntConstant(11, 'AT_UNI_NON_BOUNDARY')
|
|
|
|
ATCODES = [
|
|
AT_BEGINNING,
|
|
AT_BEGINNING_LINE,
|
|
AT_BEGINNING_STRING,
|
|
AT_BOUNDARY,
|
|
AT_NON_BOUNDARY,
|
|
AT_END,
|
|
AT_END_LINE,
|
|
AT_END_STRING,
|
|
AT_LOC_BOUNDARY,
|
|
AT_LOC_NON_BOUNDARY,
|
|
AT_UNI_BOUNDARY,
|
|
AT_UNI_NON_BOUNDARY,
|
|
]
|
|
|
|
CATEGORY_DIGIT = _NamedIntConstant( 1, 'CATEGORY_DIGIT')
|
|
CATEGORY_NOT_DIGIT = _NamedIntConstant( 2, 'CATEGORY_NOT_DIGIT')
|
|
CATEGORY_SPACE = _NamedIntConstant( 3, 'CATEGORY_SPACE')
|
|
CATEGORY_NOT_SPACE = _NamedIntConstant( 4, 'CATEGORY_NOT_SPACE')
|
|
CATEGORY_WORD = _NamedIntConstant( 5, 'CATEGORY_WORD')
|
|
CATEGORY_NOT_WORD = _NamedIntConstant( 6, 'CATEGORY_NOT_WORD')
|
|
CATEGORY_LINEBREAK = _NamedIntConstant( 7, 'CATEGORY_LINEBREAK')
|
|
CATEGORY_NOT_LINEBREAK = _NamedIntConstant( 8, 'CATEGORY_NOT_LINEBREAK')
|
|
CATEGORY_LOC_WORD = _NamedIntConstant( 9, 'CATEGORY_LOC_WORD')
|
|
CATEGORY_LOC_NOT_WORD = _NamedIntConstant(10, 'CATEGORY_LOC_NOT_WORD')
|
|
CATEGORY_UNI_DIGIT = _NamedIntConstant(11, 'CATEGORY_UNI_DIGIT')
|
|
CATEGORY_UNI_NOT_DIGIT = _NamedIntConstant(12, 'CATEGORY_UNI_NOT_DIGIT')
|
|
CATEGORY_UNI_SPACE = _NamedIntConstant(13, 'CATEGORY_UNI_SPACE')
|
|
CATEGORY_UNI_NOT_SPACE = _NamedIntConstant(14, 'CATEGORY_UNI_NOT_SPACE')
|
|
CATEGORY_UNI_WORD = _NamedIntConstant(15, 'CATEGORY_UNI_WORD')
|
|
CATEGORY_UNI_NOT_WORD = _NamedIntConstant(16, 'CATEGORY_UNI_NOT_WORD')
|
|
CATEGORY_UNI_LINEBREAK = _NamedIntConstant(17, 'CATEGORY_UNI_LINEBREAK')
|
|
CATEGORY_UNI_NOT_LINEBREAK = _NamedIntConstant(18, 'CATEGORY_UNI_NOT_LINEBREAK')
|
|
|
|
CHCODES = [
|
|
CATEGORY_DIGIT,
|
|
CATEGORY_NOT_DIGIT,
|
|
CATEGORY_SPACE,
|
|
CATEGORY_NOT_SPACE,
|
|
CATEGORY_WORD,
|
|
CATEGORY_NOT_WORD,
|
|
CATEGORY_LINEBREAK,
|
|
CATEGORY_NOT_LINEBREAK,
|
|
CATEGORY_LOC_WORD,
|
|
CATEGORY_LOC_NOT_WORD,
|
|
CATEGORY_UNI_DIGIT,
|
|
CATEGORY_UNI_NOT_DIGIT,
|
|
CATEGORY_UNI_SPACE,
|
|
CATEGORY_UNI_NOT_SPACE,
|
|
CATEGORY_UNI_WORD,
|
|
CATEGORY_UNI_NOT_WORD,
|
|
CATEGORY_UNI_LINEBREAK,
|
|
CATEGORY_UNI_NOT_LINEBREAK,
|
|
]
|
|
|
|
# replacement operations for "ignore case" mode
|
|
OP_IGNORE = {
|
|
GROUPREF: GROUPREF_IGNORE,
|
|
IN: IN_IGNORE,
|
|
LITERAL: LITERAL_IGNORE,
|
|
NOT_LITERAL: NOT_LITERAL_IGNORE,
|
|
RANGE: RANGE_IGNORE,
|
|
}
|
|
|
|
AT_MULTILINE = {
|
|
AT_BEGINNING: AT_BEGINNING_LINE,
|
|
AT_END: AT_END_LINE
|
|
}
|
|
|
|
AT_LOCALE = {
|
|
AT_BOUNDARY: AT_LOC_BOUNDARY,
|
|
AT_NON_BOUNDARY: AT_LOC_NON_BOUNDARY
|
|
}
|
|
|
|
AT_UNICODE = {
|
|
AT_BOUNDARY: AT_UNI_BOUNDARY,
|
|
AT_NON_BOUNDARY: AT_UNI_NON_BOUNDARY
|
|
}
|
|
|
|
CH_LOCALE = {
|
|
CATEGORY_DIGIT: CATEGORY_DIGIT,
|
|
CATEGORY_NOT_DIGIT: CATEGORY_NOT_DIGIT,
|
|
CATEGORY_SPACE: CATEGORY_SPACE,
|
|
CATEGORY_NOT_SPACE: CATEGORY_NOT_SPACE,
|
|
CATEGORY_WORD: CATEGORY_LOC_WORD,
|
|
CATEGORY_NOT_WORD: CATEGORY_LOC_NOT_WORD,
|
|
CATEGORY_LINEBREAK: CATEGORY_LINEBREAK,
|
|
CATEGORY_NOT_LINEBREAK: CATEGORY_NOT_LINEBREAK
|
|
}
|
|
|
|
CH_UNICODE = {
|
|
CATEGORY_DIGIT: CATEGORY_UNI_DIGIT,
|
|
CATEGORY_NOT_DIGIT: CATEGORY_UNI_NOT_DIGIT,
|
|
CATEGORY_SPACE: CATEGORY_UNI_SPACE,
|
|
CATEGORY_NOT_SPACE: CATEGORY_UNI_NOT_SPACE,
|
|
CATEGORY_WORD: CATEGORY_UNI_WORD,
|
|
CATEGORY_NOT_WORD: CATEGORY_UNI_NOT_WORD,
|
|
CATEGORY_LINEBREAK: CATEGORY_UNI_LINEBREAK,
|
|
CATEGORY_NOT_LINEBREAK: CATEGORY_UNI_NOT_LINEBREAK
|
|
}
|
|
|
|
# flags
|
|
SRE_FLAG_TEMPLATE = 1 # template mode (disable backtracking)
|
|
SRE_FLAG_IGNORECASE = 2 # case insensitive
|
|
SRE_FLAG_LOCALE = 4 # honour system locale
|
|
SRE_FLAG_MULTILINE = 8 # treat target as multiline string
|
|
SRE_FLAG_DOTALL = 16 # treat target as a single string
|
|
SRE_FLAG_UNICODE = 32 # use unicode "locale"
|
|
SRE_FLAG_VERBOSE = 64 # ignore whitespace and comments
|
|
SRE_FLAG_DEBUG = 128 # debugging
|
|
SRE_FLAG_ASCII = 256 # use ascii "locale"
|
|
|
|
# flags for INFO primitive
|
|
SRE_INFO_PREFIX = 1 # has prefix
|
|
SRE_INFO_LITERAL = 2 # entire pattern is literal (given by prefix)
|
|
SRE_INFO_CHARSET = 4 # pattern starts with character from given set
|
|
|
|
if __name__ == "__main__":
|
|
def dump(f, d, prefix):
|
|
items = sorted(d)
|
|
for item in items:
|
|
f.write("#define %s_%s %d\n" % (prefix, item, item))
|
|
with open("sre_constants.h", "w") as f:
|
|
f.write("""\
|
|
/*
|
|
* Secret Labs' Regular Expression Engine
|
|
*
|
|
* regular expression matching engine
|
|
*
|
|
* NOTE: This file is generated by sre_constants.py. If you need
|
|
* to change anything in here, edit sre_constants.py and run it.
|
|
*
|
|
* Copyright (c) 1997-2001 by Secret Labs AB. All rights reserved.
|
|
*
|
|
* See the _sre.c file for information on usage and redistribution.
|
|
*/
|
|
|
|
""")
|
|
|
|
f.write("#define SRE_MAGIC %d\n" % MAGIC)
|
|
|
|
dump(f, OPCODES, "SRE_OP")
|
|
dump(f, ATCODES, "SRE")
|
|
dump(f, CHCODES, "SRE")
|
|
|
|
f.write("#define SRE_FLAG_TEMPLATE %d\n" % SRE_FLAG_TEMPLATE)
|
|
f.write("#define SRE_FLAG_IGNORECASE %d\n" % SRE_FLAG_IGNORECASE)
|
|
f.write("#define SRE_FLAG_LOCALE %d\n" % SRE_FLAG_LOCALE)
|
|
f.write("#define SRE_FLAG_MULTILINE %d\n" % SRE_FLAG_MULTILINE)
|
|
f.write("#define SRE_FLAG_DOTALL %d\n" % SRE_FLAG_DOTALL)
|
|
f.write("#define SRE_FLAG_UNICODE %d\n" % SRE_FLAG_UNICODE)
|
|
f.write("#define SRE_FLAG_VERBOSE %d\n" % SRE_FLAG_VERBOSE)
|
|
f.write("#define SRE_FLAG_DEBUG %d\n" % SRE_FLAG_DEBUG)
|
|
f.write("#define SRE_FLAG_ASCII %d\n" % SRE_FLAG_ASCII)
|
|
|
|
f.write("#define SRE_INFO_PREFIX %d\n" % SRE_INFO_PREFIX)
|
|
f.write("#define SRE_INFO_LITERAL %d\n" % SRE_INFO_LITERAL)
|
|
f.write("#define SRE_INFO_CHARSET %d\n" % SRE_INFO_CHARSET)
|
|
|
|
print("done")
|