def ParseComplEsc(self):
"""Returns a CharClass, parsing the complement of a category escape."""
if self.Match("\\P{"):
self.SetPos(self.pos + 3)
cClass = CharClass(self.ParseCharProp())
if self.theChar == '}':
self.NextChar()
cClass.Negate()
return cClass
raise RegularExpressionError("Expected \\P{...} at [%i]" % self.pos)
def ParseComplEsc(self):
"""Returns a CharClass, parsing the complement of a category escape."""
if self.match("\\P{"):
self.setpos(self.pos + 3)
cClass = CharClass(self.ParseCharProp())
if self.the_char == '}':
self.next_char()
cClass.negate()
return cClass
raise RegularExpressionError("Expected \\P{...} at [%i]" % self.pos)
def ParsePosCharGroup(self):
"""Returns a CharClass representing a positive range"""
cClass = CharClass()
nRanges = 0
while True:
savePos = self.pos
if self.theChar == u"-" and nRanges:
# This had better be the last hyphen in the posCharGroup
if self.Match(u"-["):
# a subtraction
break
elif self.Match(u"-]") or self.Match(u"--["):
cClass.AddChar(u"-")
self.NextChar()
break
else:
# this is all wrong
raise RegularExpressionError(
"hyphen must be first or last character in posCharGroup [%i]"
% self.pos)
try:
cClass.AddClass(self.ParseCharRange())
nRanges += 1
continue
except RegularExpressionError:
self.SetPos(savePos)
pass
try:
cClass.AddClass(self.ParseCharClassEsc())
nRanges += 1
continue
except RegularExpressionError:
if nRanges:
self.SetPos(savePos)
break
else:
# We expected either a charRange or a charClassEsc
raise RegularExpressionError(
"Expected charRange or charClassEsc at [%i]" %
self.pos)
return cClass
def ParseIsCategory(self):
"""Returns a CharClass corresponding to one of the character categories
or raises an error."""
if self.the_char in self.Categories:
cat = self.the_char
self.next_char()
if self.the_char is not None and (cat + self.the_char) in self.Categories:
cat = cat + self.the_char
self.next_char()
return CharClass.ucd_category(cat)
else:
raise RegularExpressionError(
"Expected category name [%i]" % self.pos)
def ParseCharRange(self):
"""Returns a CharClass representing this range."""
savePos = self.pos
try:
cClass = self.ParseSERange()
except RegularExpressionError:
self.SetPos(savePos)
if self.IsXmlCharIncDash():
cClass = CharClass(self.theChar)
self.NextChar()
else:
raise
return cClass
def ParseIsCategory(self):
"""Returns a CharClass corresponding to one of the character categories
or raises an error."""
if self.theChar in self.Categories:
cat = self.theChar
self.NextChar()
if self.theChar is not None and (cat + self.theChar) in self.Categories:
cat = cat + self.theChar
self.NextChar()
return CharClass.UCDCategory(cat)
else:
raise RegularExpressionError(
"Expected category name [%i]" % self.pos)
def ParseSERange(self):
"""Returns a CharClass representing this range."""
s = self.ParseCharOrEsc()
if self.theChar == u"-":
self.NextChar()
else:
raise RegularExpressionError(
"Expected '-' in seRange [%i]" % self.pos)
e = self.ParseCharOrEsc()
if ord(s) > ord(e):
raise RegularExpressionError(
"Empty SERange: %s-%s" % (repr(s), repr(e)))
return CharClass((s, e))
def ParseIsBlock(self):
"""Returns a CharClass corresponding to one of the Unicode blocks."""
block = []
while self.IsBlockClass.test(self.the_char):
block.append(self.the_char)
self.next_char()
block = string.join(block, '')
if block.startswith("Is"):
try:
return CharClass.ucd_block(block[2:])
except KeyError:
raise RegularExpressionError(
"Invalid IsBlock name: %s" % block[2:])
else:
raise RegularExpressionError("Expected IsBlock [%i]" % self.pos)
def ParseIsBlock(self):
"""Returns a CharClass corresponding to one of the Unicode blocks."""
block = []
while self.IsBlockClass.Test(self.theChar):
block.append(self.theChar)
self.NextChar()
block = string.join(block, '')
if block.startswith("Is"):
try:
return CharClass.UCDBlock(block[2:])
except KeyError:
raise RegularExpressionError(
"Invalid IsBlock name: %s" % block[2:])
else:
raise RegularExpressionError("Expected IsBlock [%i]" % self.pos)
def ParseCharClassEsc(self):
"""Returns a CharClass instance representing one of the escape sequences."""
if self.Match(u"\\p"):
cClass = self.ParseCatEsc()
elif self.Match(u"\\P"):
cClass = self.ParseComplEsc()
elif self.theChar == u"\\":
try:
savePos = self.pos
cClass = self.ParseMultiCharEsc()
except RegularExpressionError:
self.SetPos(savePos)
cClass = CharClass(self.ParseSingleCharEsc())
else:
raise RegularExpressionError(
"Expected charClassEsc at [%i]" % self.pos)
return cClass
def ParsePosCharGroup(self):
"""Returns a CharClass representing a positive range"""
cClass = CharClass()
nRanges = 0
while True:
savepos = self.pos
if self.the_char == u"-" and nRanges:
# This had better be the last hyphen in the posCharGroup
if self.match(u"-["):
# a subtraction
break
elif self.match(u"-]") or self.match(u"--["):
cClass.add_char(u"-")
self.next_char()
break
else:
# this is all wrong
raise RegularExpressionError(
"hyphen must be first or last character in posCharGroup [%i]" % self.pos)
try:
cClass.add_class(self.ParseCharRange())
nRanges += 1
continue
except RegularExpressionError:
self.setpos(savepos)
pass
try:
cClass.add_class(self.ParseCharClassEsc())
nRanges += 1
continue
except RegularExpressionError:
if nRanges:
self.setpos(savepos)
break
else:
# We expected either a charRange or a charClassEsc
raise RegularExpressionError(
"Expected charRange or charClassEsc at [%i]" % self.pos)
return cClass
"""A convenience function, returns True if the expression matches *target*."""
m = self.p.match(target)
if m is None or m.end(0) < len(target):
# print "No match"
return 0
else:
# print "match"
return 1
class RegularExpressionError(Exception):
pass
sClass = CharClass(u'\x09', u'\x0A', u'\x0D', u' ')
SClass = CharClass(sClass)
SClass.negate()
iClass = CharClass(LetterCharClass, u'_', u':')
IClass = CharClass(iClass)
IClass.negate()
CClass = CharClass(NameCharClass)
CClass.negate()
dClass = CharClass.ucd_category('Nd')
DClass = CharClass(dClass)
DClass.negate()
WClass = CharClass(CharClass.ucd_category(
'P'), CharClass.ucd_category('Z'), CharClass.ucd_category('C'))
wClass = CharClass(WClass)
wClass.negate()
class RegularExpressionParser(BasicParser):
"""A custom parser for XML schema regular expressions.
The parser is initialised from a source string, the string to be parsed."""
def _REEscape(self, c=None):
if c is None:
c == self.theChar
if c in u".^$*+?{}\\[]|()":
return "\\" + c
else:
return c
def ParseRegularExpression(self):
return self.src
def ParseRegExp(self):
"""Returns a unicode string representing the regular expression."""
result = []
while True:
# expression ends at the end of the string or at a closing bracket
result.append(self.ParseBranch())
if self.theChar == u"|":
self.NextChar()
result.append(u"|")
continue
else:
break
return string.join(result, '')
def ParseBranch(self):
"""Returns a unicode string representing this piece as a python regular expression."""
result = []
while self.IsChar() or self.MatchOne(u".\\[("):
result.append(self.ParsePiece())
return string.join(result, '')
def ParsePiece(self):
result = self.ParseAtom()
if self.MatchOne("?*+{"):
n, m = self.ParseQuantifier()
if n == 0:
if m is None:
return result + u"*"
elif m == 0:
return u""
elif m == 1:
return result + u"?"
else:
return "%s{,%i}" % (result, m)
elif n == 1:
if m is None:
return result + u"+"
elif m == 1:
return result
else:
return "%s{1,%i}" % (result, m)
elif m is None:
return "%s{%i,}" % (result, n)
elif n == m:
return "%s{%i}" % (result, n)
else:
return "%s{%i,%i}" % (result, n, m)
else:
return result
def ParseQuantifier(self):
"""Returns a tuple of n,m.
Symbolic values are expanded to the appropriate pair. The second
value may be None indicating unbounded."""
if self.theChar == u"?":
self.NextChar()
return 0, 1
elif self.theChar == u"*":
self.NextChar()
return 0, None
elif self.theChar == u"+":
self.NextChar()
return 1, None
elif self.theChar == u"{":
self.NextChar()
result = self.ParseQuantity()
if self.theChar == u"}":
self.NextChar()
return result
else:
raise RegularExpressionError("Expected } at [%i]" % self.pos)
else:
raise RegularExpressionError(
"Expected quantifier at [%i]" % self.pos)
def ParseQuantity(self):
"""Returns a tuple of n,m even if an exact quantity is given.
In other words, the exact quantity 'n' returns n,n. The second
value may be None indicated unbounded."""
n = self.ParseQuantExact()
m = None
if self.theChar == u",":
self.NextChar()
if self.MatchOne(u"0123456789"):
m = self.ParseQuantExact()
if n > m:
raise RegularExpressionError(
"Illegal quantity: {%i,%i}" % (n, m))
else:
m = n
return n, m
def ParseQuantExact(self):
"""Returns an integer."""
result = 0
nDigits = 0
while self.MatchOne(u"0123456789"):
result = result * 10 + ord(self.theChar) - 0x30
self.NextChar()
nDigits += 1
if nDigits == 0:
raise RegularExpressionError("Expected digit at [%i]" % self.pos)
return result
def ParseAtom(self):
"""Returns a unicode string representing this atom as a python regular expression."""
if self.IsChar():
result = self._REEscape(self.theChar)
self.NextChar()
elif self.theChar == "(":
# a regular expression
self.NextChar()
result = "(%s)" % self.ParseRegExp()
if self.theChar != ")":
raise RegularExpressionError("Expected ) at [%i]" % self.pos)
self.NextChar()
else:
cClass = self.ParseCharClass()
result = unicode(cClass)
return result
def IsChar(self, c=None):
"""The definition of this function is designed to be conservative with
respect to the specification, which is clearly in error around
production [10] as the prose and the BNF do not match. It appears that
| was intended to be excluded in the prose but has been omitted, the
reverse being true for the curly-brackets."""
if c is None:
c = self.theChar
if c is None or c in ".\\?*+{}()[]|":
return False
else:
return True
def ParseCharClass(self):
"""Returns a CharClass instance representing this class."""
if self.theChar == u"[":
return self.ParseCharClassExpr()
elif self.theChar == u"\\":
return self.ParseCharClassEsc()
elif self.theChar == u".":
return self.ParseWildcardEsc()
else:
raise RegularExpressionError(
"Expected [, \\ or . at [%i]" % self.pos)
def ParseCharClassExpr(self):
"""Returns a CharClass instance representing this class expression."""
if self.theChar == "[":
self.NextChar()
cClass = self.ParseCharGroup()
if self.theChar == "]":
self.NextChar()
return cClass
else:
raise RegularExpressionError("Expected ] at [%i]" % self.pos)
else:
raise RegularExpressionError("Expected [ at [%i]" % self.pos)
def ParseCharGroup(self):
"""Returns a CharClass representing this group. This method also
handles the case of a class subtraction directly to reduce the need for
look-ahead. If you specifically want to parse a subtraction you can do
this with :py:meth:`ParseCharClassSub`."""
if self.theChar == u"^":
cClass = self.ParseNegCharGroup()
else:
cClass = self.ParsePosCharGroup()
if self.theChar == u"-":
self.NextChar()
subClass = self.ParseCharClassExpr()
cClass.SubtractClass(subClass)
return cClass
def ParsePosCharGroup(self):
"""Returns a CharClass representing a positive range"""
cClass = CharClass()
nRanges = 0
while True:
savePos = self.pos
if self.theChar == u"-" and nRanges:
# This had better be the last hyphen in the posCharGroup
if self.Match(u"-["):
# a subtraction
break
elif self.Match(u"-]") or self.Match(u"--["):
cClass.AddChar(u"-")
self.NextChar()
break
else:
# this is all wrong
raise RegularExpressionError(
"hyphen must be first or last character in posCharGroup [%i]" % self.pos)
try:
cClass.AddClass(self.ParseCharRange())
nRanges += 1
continue
except RegularExpressionError:
self.SetPos(savePos)
pass
try:
cClass.AddClass(self.ParseCharClassEsc())
nRanges += 1
continue
except RegularExpressionError:
if nRanges:
self.SetPos(savePos)
break
else:
# We expected either a charRange or a charClassEsc
raise RegularExpressionError(
"Expected charRange or charClassEsc at [%i]" % self.pos)
return cClass
def ParseNegCharGroup(self):
"""Returns a CharClass representing this range."""
if self.theChar == u"^":
# we have a negative range
self.NextChar()
cClass = self.ParsePosCharGroup()
cClass.Negate()
return cClass
else:
raise RegularExpressionError(
"Expected negCharGroup at [%i]" % self.pos)
def ParseCharClassSub(self):
"""Returns a CharClass representing this range - this method is not
normally used by the parser as in present for completeness. See
:py:meth:`ParseCharGroup`."""
if self.theChar == u"^":
cClass = self.ParseNegCharGroup()
else:
cClass = self.ParsePosCharGroup()
if self.theChar == u"-":
self.NextChar()
subClass = self.ParseCharClassExpr()
cClass.SubtractClass(subClass)
return cClass
else:
raise RegularExpressionError("Expected - at [%i]" % self.pos)
def ParseCharRange(self):
"""Returns a CharClass representing this range."""
savePos = self.pos
try:
cClass = self.ParseSERange()
except RegularExpressionError:
self.SetPos(savePos)
if self.IsXmlCharIncDash():
cClass = CharClass(self.theChar)
self.NextChar()
else:
raise
return cClass
def ParseSERange(self):
"""Returns a CharClass representing this range."""
s = self.ParseCharOrEsc()
if self.theChar == u"-":
self.NextChar()
else:
raise RegularExpressionError(
"Expected '-' in seRange [%i]" % self.pos)
e = self.ParseCharOrEsc()
if ord(s) > ord(e):
raise RegularExpressionError(
"Empty SERange: %s-%s" % (repr(s), repr(e)))
return CharClass((s, e))
def ParseCharOrEsc(self):
"""Returns a single unicode character."""
if self.IsXmlChar():
result = self.theChar
self.NextChar()
return result
else:
return self.ParseSingleCharEsc()
def IsXmlChar(self, c=None):
if c is None:
c = self.theChar
return c is not None and c not in "\\-[]"
def IsXmlCharIncDash(self, c=None):
if c is None:
c = self.theChar
return c is not None and c not in "\\[]"
def ParseCharClassEsc(self):
"""Returns a CharClass instance representing one of the escape sequences."""
if self.Match(u"\\p"):
cClass = self.ParseCatEsc()
elif self.Match(u"\\P"):
cClass = self.ParseComplEsc()
elif self.theChar == u"\\":
try:
savePos = self.pos
cClass = self.ParseMultiCharEsc()
except RegularExpressionError:
self.SetPos(savePos)
cClass = CharClass(self.ParseSingleCharEsc())
else:
raise RegularExpressionError(
"Expected charClassEsc at [%i]" % self.pos)
return cClass
SingleCharEscapes = {
u'n': unichr(0x0A),
u'r': unichr(0x0D),
u't': unichr(0x09),
u'\\': u'\\',
u'|': u'|',
u'.': u'.',
u'-': u'-',
u'^': u'^',
u'?': u'?',
u'*': u'*',
u'+': u'+',
u'{': u'{',
u'}': u'}',
u'(': u'(',
u')': u')',
u'[': u'[',
u']': u']'
}
def ParseSingleCharEsc(self):
"""Returns a single unicode character parsed from a single char escape."""
if self.theChar == u"\\":
self.NextChar()
if self.theChar in self.SingleCharEscapes:
result = self.SingleCharEscapes[self.theChar]
self.NextChar()
return result
raise RegularExpressionError(
"Expected single character escape at [%i]" % self.pos)
def ParseCatEsc(self):
"""Returns a CharClass, parsing a category escape."""
if self.Match("\\p{"):
self.SetPos(self.pos + 3)
cClass = self.ParseCharProp()
if self.theChar == '}':
self.NextChar()
return cClass
raise RegularExpressionError("Expected \\p{...} at [%i]" % self.pos)
def ParseComplEsc(self):
"""Returns a CharClass, parsing the complement of a category escape."""
if self.Match("\\P{"):
self.SetPos(self.pos + 3)
cClass = CharClass(self.ParseCharProp())
if self.theChar == '}':
self.NextChar()
cClass.Negate()
return cClass
raise RegularExpressionError("Expected \\P{...} at [%i]" % self.pos)
def ParseCharProp(self):
"""Returns a CharClass, parsing an IsCategory or IsBlock."""
savePos = self.pos
try:
cClass = self.ParseIsCategory()
except RegularExpressionError:
self.SetPos(savePos)
cClass = self.ParseIsBlock()
return cClass
Categories = {
u"L": True,
u"Lu": True,
u"Ll": True,
u"Lt": True,
u"Lm": True,
u"Lo": True,
u"M": True,
u"Mn": True,
u"Mc": True,
u"Me": True,
u"N": True,
u"Nd": True,
u"Nl": True,
u"No": True,
u"P": True,
u"Pc": True,
u"Pd": True,
u"Ps": True,
u"Pe": True,
u"Pi": True,
u"Pf": True,
u"Po": True,
u"Z": True,
u"Zs": True,
u"Zl": True,
u"Zp": True,
u"S": True,
u"Sm": True,
u"Sc": True,
u"Sk": True,
u"So": True,
u"C": True,
u"Cc": True,
u"Cf": True,
u"Co": True,
u"Cn": True
}
def ParseIsCategory(self):
"""Returns a CharClass corresponding to one of the character categories
or raises an error."""
if self.theChar in self.Categories:
cat = self.theChar
self.NextChar()
if self.theChar is not None and (cat + self.theChar) in self.Categories:
cat = cat + self.theChar
self.NextChar()
return CharClass.UCDCategory(cat)
else:
raise RegularExpressionError(
"Expected category name [%i]" % self.pos)
IsBlockClass = CharClass(
(u'a', u'z'),
(u'A', u'Z'),
(u'0', u'9'),
u'-')
def ParseIsBlock(self):
"""Returns a CharClass corresponding to one of the Unicode blocks."""
block = []
while self.IsBlockClass.Test(self.theChar):
block.append(self.theChar)
self.NextChar()
block = string.join(block, '')
if block.startswith("Is"):
try:
return CharClass.UCDBlock(block[2:])
except KeyError:
raise RegularExpressionError(
"Invalid IsBlock name: %s" % block[2:])
else:
raise RegularExpressionError("Expected IsBlock [%i]" % self.pos)
MultiCharEscapes = {
's': sClass,
'S': SClass,
'i': iClass,
'I': IClass,
'c': NameCharClass,
'C': CClass,
'd': dClass,
'D': DClass,
'w': wClass,
'W': WClass
}
def ParseMultiCharEsc(self):
"""Returns a CharClass corresponding to one of the multichar escapes, if parsed."""
if self.theChar == u"\\":
self.NextChar()
try:
result = self.MultiCharEscapes[self.theChar]
self.NextChar()
return result
except KeyError:
# unknown escape
raise RegularExpressionError(
"Unknown multichar escape at [%i], \\%s" % (self.pos, repr(self.theChar)))
else:
raise RegularExpressionError("Expected '\\' at [%i]" % self.pos)
DotClass = CharClass(
(unichr(0), unichr(9)),
(unichr(11), unichr(12)),
(unichr(14), unichr(maxunicode)))
def ParseWildcardEsc(self):
"""Returns a CharClass corresponding to the wildcard '.' character if parsed."""
if self.theChar == u".":
self.NextChar()
return self.DotClass
else:
raise RegularExpressionError("Expected '.' at [%i]" % self.pos)
def Match(self, target):
"""A convenience function, returns True if the expression matches *target*."""
m = self.p.match(target)
if m is None or m.end(0) < len(target):
# print "No Match"
return 0
else:
# print "Match"
return 1
class RegularExpressionError(Exception):
pass
sClass = CharClass(u'\x09', u'\x0A', u'\x0D', u' ')
SClass = CharClass(sClass)
SClass.Negate()
iClass = CharClass(LetterCharClass, u'_', u':')
IClass = CharClass(iClass)
IClass.Negate()
CClass = CharClass(NameCharClass)
CClass.Negate()
dClass = CharClass.UCDCategory('Nd')
DClass = CharClass(dClass)
DClass.Negate()
WClass = CharClass(CharClass.UCDCategory(
'P'), CharClass.UCDCategory('Z'), CharClass.UCDCategory('C'))
wClass = CharClass(WClass)
wClass.Negate()
