This patch looks large, but it just deletes the ^M characters and
untabifies the files. No actual code changes were made.
This commit is contained in:
parent
e188d52a7e
commit
e3ba931aa4
|
@ -26,7 +26,7 @@ from sre_constants import *
|
||||||
# find an array type code that matches the engine's code size
|
# find an array type code that matches the engine's code size
|
||||||
for WORDSIZE in "BHil":
|
for WORDSIZE in "BHil":
|
||||||
if len(array.array(WORDSIZE, [0]).tostring()) == _sre.getcodesize():
|
if len(array.array(WORDSIZE, [0]).tostring()) == _sre.getcodesize():
|
||||||
break
|
break
|
||||||
else:
|
else:
|
||||||
raise RuntimeError, "cannot find a useable array type"
|
raise RuntimeError, "cannot find a useable array type"
|
||||||
|
|
||||||
|
@ -34,18 +34,18 @@ else:
|
||||||
|
|
||||||
class Code:
|
class Code:
|
||||||
def __init__(self):
|
def __init__(self):
|
||||||
self.data = []
|
self.data = []
|
||||||
def __len__(self):
|
def __len__(self):
|
||||||
return len(self.data)
|
return len(self.data)
|
||||||
def __getitem__(self, index):
|
def __getitem__(self, index):
|
||||||
return self.data[index]
|
return self.data[index]
|
||||||
def __setitem__(self, index, code):
|
def __setitem__(self, index, code):
|
||||||
self.data[index] = code
|
self.data[index] = code
|
||||||
def append(self, code):
|
def append(self, code):
|
||||||
self.data.append(code)
|
self.data.append(code)
|
||||||
def todata(self):
|
def todata(self):
|
||||||
# print self.data
|
# print self.data
|
||||||
return array.array(WORDSIZE, self.data).tostring()
|
return array.array(WORDSIZE, self.data).tostring()
|
||||||
|
|
||||||
def _lower(literal):
|
def _lower(literal):
|
||||||
# return _sre._lower(literal) # FIXME
|
# return _sre._lower(literal) # FIXME
|
||||||
|
@ -54,122 +54,122 @@ def _lower(literal):
|
||||||
def _compile(code, pattern, flags):
|
def _compile(code, pattern, flags):
|
||||||
append = code.append
|
append = code.append
|
||||||
for op, av in pattern:
|
for op, av in pattern:
|
||||||
if op is ANY:
|
if op is ANY:
|
||||||
if "s" in flags:
|
if "s" in flags:
|
||||||
append(CODES[op]) # any character at all!
|
append(CODES[op]) # any character at all!
|
||||||
else:
|
else:
|
||||||
append(CODES[NOT_LITERAL])
|
append(CODES[NOT_LITERAL])
|
||||||
append(10)
|
append(10)
|
||||||
elif op in (SUCCESS, FAILURE):
|
elif op in (SUCCESS, FAILURE):
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
elif op is AT:
|
elif op is AT:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
append(POSITIONS[av])
|
append(POSITIONS[av])
|
||||||
elif op is BRANCH:
|
elif op is BRANCH:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
tail = []
|
tail = []
|
||||||
for av in av[1]:
|
for av in av[1]:
|
||||||
skip = len(code); append(0)
|
skip = len(code); append(0)
|
||||||
_compile(code, av, flags)
|
_compile(code, av, flags)
|
||||||
append(CODES[JUMP])
|
append(CODES[JUMP])
|
||||||
tail.append(len(code)); append(0)
|
tail.append(len(code)); append(0)
|
||||||
code[skip] = len(code) - skip
|
code[skip] = len(code) - skip
|
||||||
append(0) # end of branch
|
append(0) # end of branch
|
||||||
for tail in tail:
|
for tail in tail:
|
||||||
code[tail] = len(code) - tail
|
code[tail] = len(code) - tail
|
||||||
elif op is CALL:
|
elif op is CALL:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
skip = len(code); append(0)
|
skip = len(code); append(0)
|
||||||
_compile(code, av, flags)
|
_compile(code, av, flags)
|
||||||
append(CODES[SUCCESS])
|
append(CODES[SUCCESS])
|
||||||
code[skip] = len(code) - skip
|
code[skip] = len(code) - skip
|
||||||
elif op is CATEGORY: # not used by current parser
|
elif op is CATEGORY: # not used by current parser
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
append(CATEGORIES[av])
|
append(CATEGORIES[av])
|
||||||
elif op is GROUP:
|
elif op is GROUP:
|
||||||
if "i" in flags:
|
if "i" in flags:
|
||||||
append(CODES[MAP_IGNORE[op]])
|
append(CODES[MAP_IGNORE[op]])
|
||||||
else:
|
else:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
append(av)
|
append(av)
|
||||||
elif op is IN:
|
elif op is IN:
|
||||||
if "i" in flags:
|
if "i" in flags:
|
||||||
append(CODES[MAP_IGNORE[op]])
|
append(CODES[MAP_IGNORE[op]])
|
||||||
def fixup(literal):
|
def fixup(literal):
|
||||||
return ord(_lower(literal))
|
return ord(_lower(literal))
|
||||||
else:
|
else:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
fixup = ord
|
fixup = ord
|
||||||
skip = len(code); append(0)
|
skip = len(code); append(0)
|
||||||
for op, av in av:
|
for op, av in av:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
if op is NEGATE:
|
if op is NEGATE:
|
||||||
pass
|
pass
|
||||||
elif op is LITERAL:
|
elif op is LITERAL:
|
||||||
append(fixup(av))
|
append(fixup(av))
|
||||||
elif op is RANGE:
|
elif op is RANGE:
|
||||||
append(fixup(av[0]))
|
append(fixup(av[0]))
|
||||||
append(fixup(av[1]))
|
append(fixup(av[1]))
|
||||||
elif op is CATEGORY:
|
elif op is CATEGORY:
|
||||||
append(CATEGORIES[av])
|
append(CATEGORIES[av])
|
||||||
else:
|
else:
|
||||||
raise ValueError, "unsupported set operator"
|
raise ValueError, "unsupported set operator"
|
||||||
append(CODES[FAILURE])
|
append(CODES[FAILURE])
|
||||||
code[skip] = len(code) - skip
|
code[skip] = len(code) - skip
|
||||||
elif op in (LITERAL, NOT_LITERAL):
|
elif op in (LITERAL, NOT_LITERAL):
|
||||||
if "i" in flags:
|
if "i" in flags:
|
||||||
append(CODES[MAP_IGNORE[op]])
|
append(CODES[MAP_IGNORE[op]])
|
||||||
append(ord(_lower(av)))
|
append(ord(_lower(av)))
|
||||||
else:
|
else:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
append(ord(av))
|
append(ord(av))
|
||||||
elif op is MARK:
|
elif op is MARK:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
append(av)
|
append(av)
|
||||||
elif op in (REPEAT, MIN_REPEAT, MAX_REPEAT):
|
elif op in (REPEAT, MIN_REPEAT, MAX_REPEAT):
|
||||||
lo, hi = av[2].getwidth()
|
lo, hi = av[2].getwidth()
|
||||||
if lo == 0:
|
if lo == 0:
|
||||||
raise SyntaxError, "cannot repeat zero-width items"
|
raise SyntaxError, "cannot repeat zero-width items"
|
||||||
if lo == hi == 1 and op is MAX_REPEAT:
|
if lo == hi == 1 and op is MAX_REPEAT:
|
||||||
append(CODES[MAX_REPEAT_ONE])
|
append(CODES[MAX_REPEAT_ONE])
|
||||||
skip = len(code); append(0)
|
skip = len(code); append(0)
|
||||||
append(av[0])
|
append(av[0])
|
||||||
append(av[1])
|
append(av[1])
|
||||||
_compile(code, av[2], flags)
|
_compile(code, av[2], flags)
|
||||||
append(CODES[SUCCESS])
|
append(CODES[SUCCESS])
|
||||||
code[skip] = len(code) - skip
|
code[skip] = len(code) - skip
|
||||||
else:
|
else:
|
||||||
append(CODES[op])
|
append(CODES[op])
|
||||||
skip = len(code); append(0)
|
skip = len(code); append(0)
|
||||||
append(av[0])
|
append(av[0])
|
||||||
append(av[1])
|
append(av[1])
|
||||||
_compile(code, av[2], flags)
|
_compile(code, av[2], flags)
|
||||||
if op is MIN_REPEAT:
|
if op is MIN_REPEAT:
|
||||||
append(CODES[MIN_UNTIL])
|
append(CODES[MIN_UNTIL])
|
||||||
else:
|
else:
|
||||||
# FIXME: MAX_REPEAT PROBABLY DOESN'T WORK (?)
|
# FIXME: MAX_REPEAT PROBABLY DOESN'T WORK (?)
|
||||||
append(CODES[MAX_UNTIL])
|
append(CODES[MAX_UNTIL])
|
||||||
code[skip] = len(code) - skip
|
code[skip] = len(code) - skip
|
||||||
elif op is SUBPATTERN:
|
elif op is SUBPATTERN:
|
||||||
## group = av[0]
|
## group = av[0]
|
||||||
## if group:
|
## if group:
|
||||||
## append(CODES[MARK])
|
## append(CODES[MARK])
|
||||||
## append((group-1)*2)
|
## append((group-1)*2)
|
||||||
_compile(code, av[1], flags)
|
_compile(code, av[1], flags)
|
||||||
## if group:
|
## if group:
|
||||||
## append(CODES[MARK])
|
## append(CODES[MARK])
|
||||||
## append((group-1)*2+1)
|
## append((group-1)*2+1)
|
||||||
else:
|
else:
|
||||||
raise ValueError, ("unsupported operand type", op)
|
raise ValueError, ("unsupported operand type", op)
|
||||||
|
|
||||||
def compile(p, flags=()):
|
def compile(p, flags=()):
|
||||||
# convert pattern list to internal format
|
# convert pattern list to internal format
|
||||||
if type(p) is type(""):
|
if type(p) is type(""):
|
||||||
import sre_parse
|
import sre_parse
|
||||||
pattern = p
|
pattern = p
|
||||||
p = sre_parse.parse(p)
|
p = sre_parse.parse(p)
|
||||||
else:
|
else:
|
||||||
pattern = None
|
pattern = None
|
||||||
# print p.getwidth()
|
# print p.getwidth()
|
||||||
# print p
|
# print p
|
||||||
code = Code()
|
code = Code()
|
||||||
|
@ -178,10 +178,10 @@ def compile(p, flags=()):
|
||||||
# print list(code.data)
|
# print list(code.data)
|
||||||
data = code.todata()
|
data = code.todata()
|
||||||
if 0: # debugging
|
if 0: # debugging
|
||||||
print
|
print
|
||||||
print "-" * 68
|
print "-" * 68
|
||||||
import sre_disasm
|
import sre_disasm
|
||||||
sre_disasm.disasm(data)
|
sre_disasm.disasm(data)
|
||||||
print "-" * 68
|
print "-" * 68
|
||||||
# print len(data), p.pattern.groups, len(p.pattern.groupdict)
|
# print len(data), p.pattern.groups, len(p.pattern.groupdict)
|
||||||
return _sre.compile(pattern, data, p.pattern.groups-1, p.pattern.groupdict)
|
return _sre.compile(pattern, data, p.pattern.groups-1, p.pattern.groupdict)
|
||||||
|
|
|
@ -126,6 +126,6 @@ if __name__ == "__main__":
|
||||||
f = open("sre_constants.h", "w")
|
f = open("sre_constants.h", "w")
|
||||||
f.write("/* generated by sre_constants.py */\n")
|
f.write("/* generated by sre_constants.py */\n")
|
||||||
for k, v in items:
|
for k, v in items:
|
||||||
f.write("#define SRE_OP_" + string.upper(k) + " " + str(v) + "\n")
|
f.write("#define SRE_OP_" + string.upper(k) + " " + str(v) + "\n")
|
||||||
f.close()
|
f.close()
|
||||||
print "done"
|
print "done"
|
||||||
|
|
698
Lib/sre_parse.py
698
Lib/sre_parse.py
|
@ -55,168 +55,168 @@ CATEGORIES = {
|
||||||
class Pattern:
|
class Pattern:
|
||||||
# FIXME: <fl> rename class, and store flags in here too!
|
# FIXME: <fl> rename class, and store flags in here too!
|
||||||
def __init__(self):
|
def __init__(self):
|
||||||
self.flags = []
|
self.flags = []
|
||||||
self.groups = 1
|
self.groups = 1
|
||||||
self.groupdict = {}
|
self.groupdict = {}
|
||||||
def getgroup(self, name=None):
|
def getgroup(self, name=None):
|
||||||
gid = self.groups
|
gid = self.groups
|
||||||
self.groups = gid + 1
|
self.groups = gid + 1
|
||||||
if name:
|
if name:
|
||||||
self.groupdict[name] = gid
|
self.groupdict[name] = gid
|
||||||
return gid
|
return gid
|
||||||
def setflag(self, flag):
|
def setflag(self, flag):
|
||||||
if flag not in self.flags:
|
if flag not in self.flags:
|
||||||
self.flags.append(flag)
|
self.flags.append(flag)
|
||||||
|
|
||||||
class SubPattern:
|
class SubPattern:
|
||||||
# a subpattern, in intermediate form
|
# a subpattern, in intermediate form
|
||||||
def __init__(self, pattern, data=None):
|
def __init__(self, pattern, data=None):
|
||||||
self.pattern = pattern
|
self.pattern = pattern
|
||||||
if not data:
|
if not data:
|
||||||
data = []
|
data = []
|
||||||
self.data = data
|
self.data = data
|
||||||
self.flags = []
|
self.flags = []
|
||||||
self.width = None
|
self.width = None
|
||||||
def __repr__(self):
|
def __repr__(self):
|
||||||
return repr(self.data)
|
return repr(self.data)
|
||||||
def __len__(self):
|
def __len__(self):
|
||||||
return len(self.data)
|
return len(self.data)
|
||||||
def __delitem__(self, index):
|
def __delitem__(self, index):
|
||||||
del self.data[index]
|
del self.data[index]
|
||||||
def __getitem__(self, index):
|
def __getitem__(self, index):
|
||||||
return self.data[index]
|
return self.data[index]
|
||||||
def __setitem__(self, index, code):
|
def __setitem__(self, index, code):
|
||||||
self.data[index] = code
|
self.data[index] = code
|
||||||
def __getslice__(self, start, stop):
|
def __getslice__(self, start, stop):
|
||||||
return SubPattern(self.pattern, self.data[start:stop])
|
return SubPattern(self.pattern, self.data[start:stop])
|
||||||
def insert(self, index, code):
|
def insert(self, index, code):
|
||||||
self.data.insert(index, code)
|
self.data.insert(index, code)
|
||||||
def append(self, code):
|
def append(self, code):
|
||||||
self.data.append(code)
|
self.data.append(code)
|
||||||
def getwidth(self):
|
def getwidth(self):
|
||||||
# determine the width (min, max) for this subpattern
|
# determine the width (min, max) for this subpattern
|
||||||
if self.width:
|
if self.width:
|
||||||
return self.width
|
return self.width
|
||||||
lo = hi = 0L
|
lo = hi = 0L
|
||||||
for op, av in self.data:
|
for op, av in self.data:
|
||||||
if op is BRANCH:
|
if op is BRANCH:
|
||||||
l = sys.maxint
|
l = sys.maxint
|
||||||
h = 0
|
h = 0
|
||||||
for av in av[1]:
|
for av in av[1]:
|
||||||
i, j = av.getwidth()
|
i, j = av.getwidth()
|
||||||
l = min(l, i)
|
l = min(l, i)
|
||||||
h = min(h, j)
|
h = min(h, j)
|
||||||
lo = lo + i
|
lo = lo + i
|
||||||
hi = hi + j
|
hi = hi + j
|
||||||
elif op is CALL:
|
elif op is CALL:
|
||||||
i, j = av.getwidth()
|
i, j = av.getwidth()
|
||||||
lo = lo + i
|
lo = lo + i
|
||||||
hi = hi + j
|
hi = hi + j
|
||||||
elif op is SUBPATTERN:
|
elif op is SUBPATTERN:
|
||||||
i, j = av[1].getwidth()
|
i, j = av[1].getwidth()
|
||||||
lo = lo + i
|
lo = lo + i
|
||||||
hi = hi + j
|
hi = hi + j
|
||||||
elif op in (MIN_REPEAT, MAX_REPEAT):
|
elif op in (MIN_REPEAT, MAX_REPEAT):
|
||||||
i, j = av[2].getwidth()
|
i, j = av[2].getwidth()
|
||||||
lo = lo + i * av[0]
|
lo = lo + i * av[0]
|
||||||
hi = hi + j * av[1]
|
hi = hi + j * av[1]
|
||||||
elif op in (ANY, RANGE, IN, LITERAL, NOT_LITERAL, CATEGORY):
|
elif op in (ANY, RANGE, IN, LITERAL, NOT_LITERAL, CATEGORY):
|
||||||
lo = lo + 1
|
lo = lo + 1
|
||||||
hi = hi + 1
|
hi = hi + 1
|
||||||
elif op == SUCCESS:
|
elif op == SUCCESS:
|
||||||
break
|
break
|
||||||
self.width = int(min(lo, sys.maxint)), int(min(hi, sys.maxint))
|
self.width = int(min(lo, sys.maxint)), int(min(hi, sys.maxint))
|
||||||
return self.width
|
return self.width
|
||||||
def set(self, flag):
|
def set(self, flag):
|
||||||
if not flag in self.flags:
|
if not flag in self.flags:
|
||||||
self.flags.append(flag)
|
self.flags.append(flag)
|
||||||
def reset(self, flag):
|
def reset(self, flag):
|
||||||
if flag in self.flags:
|
if flag in self.flags:
|
||||||
self.flags.remove(flag)
|
self.flags.remove(flag)
|
||||||
|
|
||||||
class Tokenizer:
|
class Tokenizer:
|
||||||
def __init__(self, string):
|
def __init__(self, string):
|
||||||
self.string = list(string)
|
self.string = list(string)
|
||||||
self.next = self.__next()
|
self.next = self.__next()
|
||||||
def __next(self):
|
def __next(self):
|
||||||
if not self.string:
|
if not self.string:
|
||||||
return None
|
return None
|
||||||
char = self.string[0]
|
char = self.string[0]
|
||||||
if char[0] == "\\":
|
if char[0] == "\\":
|
||||||
try:
|
try:
|
||||||
c = self.string[1]
|
c = self.string[1]
|
||||||
except IndexError:
|
except IndexError:
|
||||||
raise SyntaxError, "bogus escape"
|
raise SyntaxError, "bogus escape"
|
||||||
char = char + c
|
char = char + c
|
||||||
try:
|
try:
|
||||||
if c == "x":
|
if c == "x":
|
||||||
# hexadecimal constant
|
# hexadecimal constant
|
||||||
for i in xrange(2, sys.maxint):
|
for i in xrange(2, sys.maxint):
|
||||||
c = self.string[i]
|
c = self.string[i]
|
||||||
if c not in HEXDIGITS:
|
if c not in HEXDIGITS:
|
||||||
break
|
break
|
||||||
char = char + c
|
char = char + c
|
||||||
elif c in string.digits:
|
elif c in string.digits:
|
||||||
# decimal (or octal) number
|
# decimal (or octal) number
|
||||||
for i in xrange(2, sys.maxint):
|
for i in xrange(2, sys.maxint):
|
||||||
c = self.string[i]
|
c = self.string[i]
|
||||||
# FIXME: if larger than current number of
|
# FIXME: if larger than current number of
|
||||||
# groups, interpret as an octal number
|
# groups, interpret as an octal number
|
||||||
if c not in string.digits:
|
if c not in string.digits:
|
||||||
break
|
break
|
||||||
char = char + c
|
char = char + c
|
||||||
except IndexError:
|
except IndexError:
|
||||||
pass # use what we've got this far
|
pass # use what we've got this far
|
||||||
del self.string[0:len(char)]
|
del self.string[0:len(char)]
|
||||||
return char
|
return char
|
||||||
def match(self, char):
|
def match(self, char):
|
||||||
if char == self.next:
|
if char == self.next:
|
||||||
self.next = self.__next()
|
self.next = self.__next()
|
||||||
return 1
|
return 1
|
||||||
return 0
|
return 0
|
||||||
def match_set(self, set):
|
def match_set(self, set):
|
||||||
if self.next in set:
|
if self.next in set:
|
||||||
self.next = self.__next()
|
self.next = self.__next()
|
||||||
return 1
|
return 1
|
||||||
return 0
|
return 0
|
||||||
def get(self):
|
def get(self):
|
||||||
this = self.next
|
this = self.next
|
||||||
self.next = self.__next()
|
self.next = self.__next()
|
||||||
return this
|
return this
|
||||||
|
|
||||||
def _fixescape(escape, character_class=0):
|
def _fixescape(escape, character_class=0):
|
||||||
# convert escape to (type, value)
|
# convert escape to (type, value)
|
||||||
if character_class:
|
if character_class:
|
||||||
# inside a character class, we'll look in the character
|
# inside a character class, we'll look in the character
|
||||||
# escapes dictionary first
|
# escapes dictionary first
|
||||||
code = ESCAPES.get(escape)
|
code = ESCAPES.get(escape)
|
||||||
if code:
|
if code:
|
||||||
return code
|
return code
|
||||||
code = CATEGORIES.get(escape)
|
code = CATEGORIES.get(escape)
|
||||||
else:
|
else:
|
||||||
code = CATEGORIES.get(escape)
|
code = CATEGORIES.get(escape)
|
||||||
if code:
|
if code:
|
||||||
return code
|
return code
|
||||||
code = ESCAPES.get(escape)
|
code = ESCAPES.get(escape)
|
||||||
if code:
|
if code:
|
||||||
return code
|
return code
|
||||||
if not character_class:
|
if not character_class:
|
||||||
try:
|
try:
|
||||||
group = int(escape[1:])
|
group = int(escape[1:])
|
||||||
# FIXME: only valid if group <= current number of groups
|
# FIXME: only valid if group <= current number of groups
|
||||||
return GROUP, group
|
return GROUP, group
|
||||||
except ValueError:
|
except ValueError:
|
||||||
pass
|
pass
|
||||||
try:
|
try:
|
||||||
if escape[1:2] == "x":
|
if escape[1:2] == "x":
|
||||||
escape = escape[2:]
|
escape = escape[2:]
|
||||||
return LITERAL, chr(string.atoi(escape[-2:], 16) & 0xff)
|
return LITERAL, chr(string.atoi(escape[-2:], 16) & 0xff)
|
||||||
elif escape[1:2] in string.digits:
|
elif escape[1:2] in string.digits:
|
||||||
return LITERAL, chr(string.atoi(escape[1:], 8) & 0xff)
|
return LITERAL, chr(string.atoi(escape[1:], 8) & 0xff)
|
||||||
elif len(escape) == 2:
|
elif len(escape) == 2:
|
||||||
return LITERAL, escape[1]
|
return LITERAL, escape[1]
|
||||||
except ValueError:
|
except ValueError:
|
||||||
pass
|
pass
|
||||||
raise SyntaxError, "bogus escape: %s" % repr(escape)
|
raise SyntaxError, "bogus escape: %s" % repr(escape)
|
||||||
|
|
||||||
def _branch(subpattern, items):
|
def _branch(subpattern, items):
|
||||||
|
@ -226,35 +226,35 @@ def _branch(subpattern, items):
|
||||||
|
|
||||||
# check if all items share a common prefix
|
# check if all items share a common prefix
|
||||||
while 1:
|
while 1:
|
||||||
prefix = None
|
prefix = None
|
||||||
for item in items:
|
for item in items:
|
||||||
if not item:
|
if not item:
|
||||||
break
|
break
|
||||||
if prefix is None:
|
if prefix is None:
|
||||||
prefix = item[0]
|
prefix = item[0]
|
||||||
elif item[0] != prefix:
|
elif item[0] != prefix:
|
||||||
break
|
break
|
||||||
else:
|
else:
|
||||||
# all subitems start with a common "prefix".
|
# all subitems start with a common "prefix".
|
||||||
# move it out of the branch
|
# move it out of the branch
|
||||||
for item in items:
|
for item in items:
|
||||||
del item[0]
|
del item[0]
|
||||||
subpattern.append(prefix)
|
subpattern.append(prefix)
|
||||||
continue # check next one
|
continue # check next one
|
||||||
break
|
break
|
||||||
|
|
||||||
# check if the branch can be replaced by a character set
|
# check if the branch can be replaced by a character set
|
||||||
for item in items:
|
for item in items:
|
||||||
if len(item) != 1 or item[0][0] != LITERAL:
|
if len(item) != 1 or item[0][0] != LITERAL:
|
||||||
break
|
break
|
||||||
else:
|
else:
|
||||||
# we can store this as a character set instead of a
|
# we can store this as a character set instead of a
|
||||||
# branch (FIXME: use a range if possible)
|
# branch (FIXME: use a range if possible)
|
||||||
set = []
|
set = []
|
||||||
for item in items:
|
for item in items:
|
||||||
set.append(item[0])
|
set.append(item[0])
|
||||||
subpattern.append((IN, set))
|
subpattern.append((IN, set))
|
||||||
return
|
return
|
||||||
|
|
||||||
subpattern.append((BRANCH, (None, items)))
|
subpattern.append((BRANCH, (None, items)))
|
||||||
|
|
||||||
|
@ -268,178 +268,178 @@ def _parse(source, pattern, flags=()):
|
||||||
|
|
||||||
while 1:
|
while 1:
|
||||||
|
|
||||||
if source.next in ("|", ")"):
|
if source.next in ("|", ")"):
|
||||||
break # end of subpattern
|
break # end of subpattern
|
||||||
this = source.get()
|
this = source.get()
|
||||||
if this is None:
|
if this is None:
|
||||||
break # end of pattern
|
break # end of pattern
|
||||||
|
|
||||||
if this and this[0] not in SPECIAL_CHARS:
|
if this and this[0] not in SPECIAL_CHARS:
|
||||||
subpattern.append((LITERAL, this))
|
subpattern.append((LITERAL, this))
|
||||||
|
|
||||||
elif this == "[":
|
elif this == "[":
|
||||||
# character set
|
# character set
|
||||||
set = []
|
set = []
|
||||||
## if source.match(":"):
|
## if source.match(":"):
|
||||||
## pass # handle character classes
|
## pass # handle character classes
|
||||||
if source.match("^"):
|
if source.match("^"):
|
||||||
set.append((NEGATE, None))
|
set.append((NEGATE, None))
|
||||||
# check remaining characters
|
# check remaining characters
|
||||||
start = set[:]
|
start = set[:]
|
||||||
while 1:
|
while 1:
|
||||||
this = source.get()
|
this = source.get()
|
||||||
if this == "]" and set != start:
|
if this == "]" and set != start:
|
||||||
break
|
break
|
||||||
elif this and this[0] == "\\":
|
elif this and this[0] == "\\":
|
||||||
code1 = _fixescape(this, 1)
|
code1 = _fixescape(this, 1)
|
||||||
elif this:
|
elif this:
|
||||||
code1 = LITERAL, this
|
code1 = LITERAL, this
|
||||||
else:
|
else:
|
||||||
raise SyntaxError, "unexpected end of regular expression"
|
raise SyntaxError, "unexpected end of regular expression"
|
||||||
if source.match("-"):
|
if source.match("-"):
|
||||||
# potential range
|
# potential range
|
||||||
this = source.get()
|
this = source.get()
|
||||||
if this == "]":
|
if this == "]":
|
||||||
set.append(code1)
|
set.append(code1)
|
||||||
set.append((LITERAL, "-"))
|
set.append((LITERAL, "-"))
|
||||||
break
|
break
|
||||||
else:
|
else:
|
||||||
if this[0] == "\\":
|
if this[0] == "\\":
|
||||||
code2 = _fixescape(this, 1)
|
code2 = _fixescape(this, 1)
|
||||||
else:
|
else:
|
||||||
code2 = LITERAL, this
|
code2 = LITERAL, this
|
||||||
if code1[0] != LITERAL or code2[0] != LITERAL:
|
if code1[0] != LITERAL or code2[0] != LITERAL:
|
||||||
raise SyntaxError, "illegal range"
|
raise SyntaxError, "illegal range"
|
||||||
if len(code1[1]) != 1 or len(code2[1]) != 1:
|
if len(code1[1]) != 1 or len(code2[1]) != 1:
|
||||||
raise SyntaxError, "illegal range"
|
raise SyntaxError, "illegal range"
|
||||||
set.append((RANGE, (code1[1], code2[1])))
|
set.append((RANGE, (code1[1], code2[1])))
|
||||||
else:
|
else:
|
||||||
if code1[0] is IN:
|
if code1[0] is IN:
|
||||||
code1 = code1[1][0]
|
code1 = code1[1][0]
|
||||||
set.append(code1)
|
set.append(code1)
|
||||||
|
|
||||||
# FIXME: <fl> move set optimization to support function
|
# FIXME: <fl> move set optimization to support function
|
||||||
if len(set)==1 and set[0][0] is LITERAL:
|
if len(set)==1 and set[0][0] is LITERAL:
|
||||||
subpattern.append(set[0]) # optimization
|
subpattern.append(set[0]) # optimization
|
||||||
elif len(set)==2 and set[0][0] is NEGATE and set[1][0] is LITERAL:
|
elif len(set)==2 and set[0][0] is NEGATE and set[1][0] is LITERAL:
|
||||||
subpattern.append((NOT_LITERAL, set[1][1])) # optimization
|
subpattern.append((NOT_LITERAL, set[1][1])) # optimization
|
||||||
else:
|
else:
|
||||||
# FIXME: <fl> add charmap optimization
|
# FIXME: <fl> add charmap optimization
|
||||||
subpattern.append((IN, set))
|
subpattern.append((IN, set))
|
||||||
|
|
||||||
elif this and this[0] in REPEAT_CHARS:
|
elif this and this[0] in REPEAT_CHARS:
|
||||||
# repeat previous item
|
# repeat previous item
|
||||||
if this == "?":
|
if this == "?":
|
||||||
min, max = 0, 1
|
min, max = 0, 1
|
||||||
elif this == "*":
|
elif this == "*":
|
||||||
min, max = 0, sys.maxint
|
min, max = 0, sys.maxint
|
||||||
elif this == "+":
|
elif this == "+":
|
||||||
min, max = 1, sys.maxint
|
min, max = 1, sys.maxint
|
||||||
elif this == "{":
|
elif this == "{":
|
||||||
min, max = 0, sys.maxint
|
min, max = 0, sys.maxint
|
||||||
lo = hi = ""
|
lo = hi = ""
|
||||||
while source.next in string.digits:
|
while source.next in string.digits:
|
||||||
lo = lo + source.get()
|
lo = lo + source.get()
|
||||||
if source.match(","):
|
if source.match(","):
|
||||||
while source.next in string.digits:
|
while source.next in string.digits:
|
||||||
hi = hi + source.get()
|
hi = hi + source.get()
|
||||||
else:
|
else:
|
||||||
hi = lo
|
hi = lo
|
||||||
if not source.match("}"):
|
if not source.match("}"):
|
||||||
raise SyntaxError, "bogus range"
|
raise SyntaxError, "bogus range"
|
||||||
if lo:
|
if lo:
|
||||||
min = int(lo)
|
min = int(lo)
|
||||||
if hi:
|
if hi:
|
||||||
max = int(hi)
|
max = int(hi)
|
||||||
# FIXME: <fl> check that hi >= lo!
|
# FIXME: <fl> check that hi >= lo!
|
||||||
else:
|
else:
|
||||||
raise SyntaxError, "not supported"
|
raise SyntaxError, "not supported"
|
||||||
# figure out which item to repeat
|
# figure out which item to repeat
|
||||||
# FIXME: should back up to the right mark, right?
|
# FIXME: should back up to the right mark, right?
|
||||||
if subpattern:
|
if subpattern:
|
||||||
index = len(subpattern)-1
|
index = len(subpattern)-1
|
||||||
while subpattern[index][0] is MARK:
|
while subpattern[index][0] is MARK:
|
||||||
index = index - 1
|
index = index - 1
|
||||||
item = subpattern[index:index+1]
|
item = subpattern[index:index+1]
|
||||||
else:
|
else:
|
||||||
raise SyntaxError, "nothing to repeat"
|
raise SyntaxError, "nothing to repeat"
|
||||||
if source.match("?"):
|
if source.match("?"):
|
||||||
subpattern[index] = (MIN_REPEAT, (min, max, item))
|
subpattern[index] = (MIN_REPEAT, (min, max, item))
|
||||||
else:
|
else:
|
||||||
subpattern[index] = (MAX_REPEAT, (min, max, item))
|
subpattern[index] = (MAX_REPEAT, (min, max, item))
|
||||||
elif this == ".":
|
elif this == ".":
|
||||||
subpattern.append((ANY, None))
|
subpattern.append((ANY, None))
|
||||||
elif this == "(":
|
elif this == "(":
|
||||||
group = 1
|
group = 1
|
||||||
name = None
|
name = None
|
||||||
if source.match("?"):
|
if source.match("?"):
|
||||||
group = 0
|
group = 0
|
||||||
# options
|
# options
|
||||||
if source.match("P"):
|
if source.match("P"):
|
||||||
# named group: skip forward to end of name
|
# named group: skip forward to end of name
|
||||||
if source.match("<"):
|
if source.match("<"):
|
||||||
name = ""
|
name = ""
|
||||||
while 1:
|
while 1:
|
||||||
char = source.get()
|
char = source.get()
|
||||||
if char in (">", None):
|
if char in (">", None):
|
||||||
break
|
break
|
||||||
name = name + char
|
name = name + char
|
||||||
group = 1
|
group = 1
|
||||||
elif source.match(":"):
|
elif source.match(":"):
|
||||||
# non-capturing group
|
# non-capturing group
|
||||||
group = 2
|
group = 2
|
||||||
elif source.match_set("iI"):
|
elif source.match_set("iI"):
|
||||||
pattern.setflag("i")
|
pattern.setflag("i")
|
||||||
elif source.match_set("lL"):
|
elif source.match_set("lL"):
|
||||||
pattern.setflag("l")
|
pattern.setflag("l")
|
||||||
elif source.match_set("mM"):
|
elif source.match_set("mM"):
|
||||||
pattern.setflag("m")
|
pattern.setflag("m")
|
||||||
elif source.match_set("sS"):
|
elif source.match_set("sS"):
|
||||||
pattern.setflag("s")
|
pattern.setflag("s")
|
||||||
elif source.match_set("xX"):
|
elif source.match_set("xX"):
|
||||||
pattern.setflag("x")
|
pattern.setflag("x")
|
||||||
if group:
|
if group:
|
||||||
# parse group contents
|
# parse group contents
|
||||||
b = []
|
b = []
|
||||||
if group == 2:
|
if group == 2:
|
||||||
# anonymous group
|
# anonymous group
|
||||||
group = None
|
group = None
|
||||||
else:
|
else:
|
||||||
group = pattern.getgroup(name)
|
group = pattern.getgroup(name)
|
||||||
if group:
|
if group:
|
||||||
subpattern.append((MARK, (group-1)*2))
|
subpattern.append((MARK, (group-1)*2))
|
||||||
while 1:
|
while 1:
|
||||||
p = _parse(source, pattern, flags)
|
p = _parse(source, pattern, flags)
|
||||||
if source.match(")"):
|
if source.match(")"):
|
||||||
if b:
|
if b:
|
||||||
b.append(p)
|
b.append(p)
|
||||||
_branch(subpattern, b)
|
_branch(subpattern, b)
|
||||||
else:
|
else:
|
||||||
subpattern.append((SUBPATTERN, (group, p)))
|
subpattern.append((SUBPATTERN, (group, p)))
|
||||||
break
|
break
|
||||||
elif source.match("|"):
|
elif source.match("|"):
|
||||||
b.append(p)
|
b.append(p)
|
||||||
else:
|
else:
|
||||||
raise SyntaxError, "group not properly closed"
|
raise SyntaxError, "group not properly closed"
|
||||||
if group:
|
if group:
|
||||||
subpattern.append((MARK, (group-1)*2+1))
|
subpattern.append((MARK, (group-1)*2+1))
|
||||||
else:
|
else:
|
||||||
# FIXME: should this really be a while loop?
|
# FIXME: should this really be a while loop?
|
||||||
while source.get() not in (")", None):
|
while source.get() not in (")", None):
|
||||||
pass
|
pass
|
||||||
|
|
||||||
elif this == "^":
|
elif this == "^":
|
||||||
subpattern.append((AT, AT_BEGINNING))
|
subpattern.append((AT, AT_BEGINNING))
|
||||||
|
|
||||||
elif this == "$":
|
elif this == "$":
|
||||||
subpattern.append((AT, AT_END))
|
subpattern.append((AT, AT_END))
|
||||||
|
|
||||||
elif this and this[0] == "\\":
|
elif this and this[0] == "\\":
|
||||||
code =_fixescape(this)
|
code =_fixescape(this)
|
||||||
subpattern.append(code)
|
subpattern.append(code)
|
||||||
|
|
||||||
else:
|
else:
|
||||||
raise SyntaxError, "parser error"
|
raise SyntaxError, "parser error"
|
||||||
|
|
||||||
return subpattern
|
return subpattern
|
||||||
|
|
||||||
|
@ -448,20 +448,20 @@ def parse(source, flags=()):
|
||||||
g = Pattern()
|
g = Pattern()
|
||||||
b = []
|
b = []
|
||||||
while 1:
|
while 1:
|
||||||
p = _parse(s, g, flags)
|
p = _parse(s, g, flags)
|
||||||
tail = s.get()
|
tail = s.get()
|
||||||
if tail == "|":
|
if tail == "|":
|
||||||
b.append(p)
|
b.append(p)
|
||||||
elif tail == ")":
|
elif tail == ")":
|
||||||
raise SyntaxError, "unbalanced parenthesis"
|
raise SyntaxError, "unbalanced parenthesis"
|
||||||
elif tail is None:
|
elif tail is None:
|
||||||
if b:
|
if b:
|
||||||
b.append(p)
|
b.append(p)
|
||||||
p = SubPattern(g)
|
p = SubPattern(g)
|
||||||
_branch(p, b)
|
_branch(p, b)
|
||||||
break
|
break
|
||||||
else:
|
else:
|
||||||
raise SyntaxError, "bogus characters at end of regular expression"
|
raise SyntaxError, "bogus characters at end of regular expression"
|
||||||
return p
|
return p
|
||||||
|
|
||||||
if __name__ == "__main__":
|
if __name__ == "__main__":
|
||||||
|
@ -469,23 +469,23 @@ if __name__ == "__main__":
|
||||||
from testpatterns import PATTERNS
|
from testpatterns import PATTERNS
|
||||||
a = b = c = 0
|
a = b = c = 0
|
||||||
for pattern, flags in PATTERNS:
|
for pattern, flags in PATTERNS:
|
||||||
if flags:
|
if flags:
|
||||||
continue
|
continue
|
||||||
print "-"*68
|
print "-"*68
|
||||||
try:
|
try:
|
||||||
p = parse(pattern)
|
p = parse(pattern)
|
||||||
print repr(pattern), "->"
|
print repr(pattern), "->"
|
||||||
pprint(p.data)
|
pprint(p.data)
|
||||||
import sre_compile
|
import sre_compile
|
||||||
try:
|
try:
|
||||||
code = sre_compile.compile(p)
|
code = sre_compile.compile(p)
|
||||||
c = c + 1
|
c = c + 1
|
||||||
except:
|
except:
|
||||||
pass
|
pass
|
||||||
a = a + 1
|
a = a + 1
|
||||||
except SyntaxError, v:
|
except SyntaxError, v:
|
||||||
print "**", repr(pattern), v
|
print "**", repr(pattern), v
|
||||||
b = b + 1
|
b = b + 1
|
||||||
print "-"*68
|
print "-"*68
|
||||||
print a, "of", b, "patterns successfully parsed"
|
print a, "of", b, "patterns successfully parsed"
|
||||||
print c, "of", b, "patterns successfully compiled"
|
print c, "of", b, "patterns successfully compiled"
|
||||||
|
|
Loading…
Reference in New Issue