def parse_named_character_class(self):
"""
Parse a named character class ([:...:]) subexpresion from the string at its current index.
@return: a CoverageSet object covering all the characters covered by the named class
"""
classes = {
u"alnum": [Parser.lowercase, Parser.uppercase, Parser.digits],
u"word": [Parser.lowercase, Parser.uppercase, Parser.digits, Parser.underscore],
u"alpha": [Parser.uppercase, Parser.lowercase],
u"blank": [Parser.space, Parser.tab],
u"cntrl": [(0, 31)],
u"digit": [Parser.digits],
u"graph": [(33, 127)],
u"lower": [Parser.lowercase],
u"print": [(32, 127)],
u"punct": [(ord(i), ord(i)) for i in "][!\"#$%&'()*+,./:;<=>?@\\^_`{|}~-"],
u"space": [(ord(i), ord(i)) for i in string.whitespace],
u"xdigit": [Parser.digits, (ord('a'), ord('f')), (ord('A'), ord('F'))],
u"upper": [Parser.uppercase]
}
unicode_classes = {
u"alnum": {False: [('alpha', None), ('digit', None)], True: []},
u"word": {False: [('alpha', None), ('gc', 'Mark'), ('digit', None), ('gc', 'Connector_Punctuation'), ('Join_Control', None)], True: []},
u"alpha": {False: [('alpha', None)], True: []},
u"blank": {False: [('gc', 'Space_Separator'), ('na1', 'CHARACTER TABULATION')], True: []},
u"cntrl": {False: [('gc', 'Control')], True: []},
u"digit": {False: [('gc', 'Decimal_Number')], True: []},
u"graph": {False: [], True: [('space', None), ('gc', 'Control'), ('gc', 'Surrogate'), ('gc', 'Unassigned')]},
u"lower": {False: [('Lowercase', None)], True: []},
u"print": {False: [('gc', 'Space_Separator'), ('na1', 'CHARACTER TABULATION')], True: [('space', None), ('gc', 'Control'), ('gc', 'Surrogate'), ('gc', 'Unassigned')]},
u"punct": {False: [('gc', 'Punctuation')], True: []},
u"space": {False: [('Whitespace', None)], True: []},
u"xdigit": {False: [('gc', 'Decimal_Number'), ('Hex_Digit', None)], True: []},
u"upper": {False: [('Uppercase', None)], True: []}
}
self.expect(u':')
class_name = ""
while self.next_is(string.ascii_letters):
class_name += self.get_next()
self.expect(":")
self.expect("]")
if self.is_unicode_defaults and class_name in unicode_classes:
class_queries = unicode_classes[class_name]
instance = UnicodeQuery.instance(self.unicode_db)
coverage = CoverageSet()
for k, v in class_queries[False]:
coverage.update(instance.query(k, v))
if len(class_queries[True]) > 0:
inverted_coverage = CoverageSet([(1, 0x10ffff)])
for k, v in class_queries[True]:
inverted_coverage.difference_update(instance.query(k, v))
coverage.update(inverted_coverage)
return coverage
elif not self.is_unicode_defaults and class_name in classes:
return CoverageSet(classes[class_name])
else:
raise RegexParserExceptionInternal("Character class '%s' not recognized" % class_name)
def parse_unicode_subexpression(self):
"""
Parse a single "NAME=VALUE" term for a unicode property.
@returns: A CoverageSet object containing the values covered by the term
"""
name = self.parse_unicode_word()
if self.get_next_if(u':='):
value = self.parse_unicode_word()
else:
value = None
coverage = UnicodeQuery.instance(self.unicode_db).query(name, value)
return coverage
def parse_unicode_name(self):
"""
Parse a unicode "{...}" expresion for querying unicode names.
@returns: A CoverageSet object containing all characters covered by
the specified unicode query
"""
self.expect("{")
name = self.parse_unicode_word()
coverage = UnicodeQuery.instance(self.unicode_db).query('na', name)
if coverage.empty():
raise ValueError("Name '{name}' not found".format(name=name))
return coverage
def parse_unicode_subexpression(self):
"""
Parse a single "NAME=VALUE" term for a unicode property.
@returns: A CoverageSet object containing the values covered by the term
"""
name = self.parse_unicode_word()
if self.get_next_if(u':='):
value = self.parse_unicode_word()
else:
value = None
coverage = UnicodeQuery.instance(self.unicode_db).query(name, value)
return coverage
def parse_unicode_name(self):
"""
Parse a unicode "{...}" expresion for querying unicode names.
@returns: A CoverageSet object containing all characters covered by
the specified unicode query
"""
self.expect("{")
name = self.parse_unicode_word()
coverage = UnicodeQuery.instance(self.unicode_db).query('na', name)
if coverage.empty():
raise ValueError("Name '{name}' not found".format(name=name))
return coverage
def parse_named_character_class(self):
"""
Parse a named character class ([:...:]) subexpresion from the string at its current index.
@return: a CoverageSet object covering all the characters covered by the named class
"""
classes = {
u"alnum": [Parser.lowercase, Parser.uppercase, Parser.digits],
u"word": [
Parser.lowercase, Parser.uppercase, Parser.digits,
Parser.underscore
],
u"alpha": [Parser.uppercase, Parser.lowercase],
u"blank": [Parser.space, Parser.tab],
u"cntrl": [(0, 31)],
u"digit": [Parser.digits],
u"graph": [(33, 127)],
u"lower": [Parser.lowercase],
u"print": [(32, 127)],
u"punct":
[(ord(i), ord(i)) for i in "][!\"#$%&'()*+,./:;<=>?@\\^_`{|}~-"],
u"space": [(ord(i), ord(i)) for i in string.whitespace],
u"xdigit":
[Parser.digits, (ord('a'), ord('f')), (ord('A'), ord('F'))],
u"upper": [Parser.uppercase]
}
unicode_classes = {
u"alnum": {
False: [('alpha', None), ('digit', None)],
True: []
},
u"word": {
False: [('alpha', None), ('gc', 'Mark'), ('digit', None),
('gc', 'Connector_Punctuation'),
('Join_Control', None)],
True: []
},
u"alpha": {
False: [('alpha', None)],
True: []
},
u"blank": {
False: [('gc', 'Space_Separator'),
('na1', 'CHARACTER TABULATION')],
True: []
},
u"cntrl": {
False: [('gc', 'Control')],
True: []
},
u"digit": {
False: [('gc', 'Decimal_Number')],
True: []
},
u"graph": {
False: [],
True: [('space', None), ('gc', 'Control'), ('gc', 'Surrogate'),
('gc', 'Unassigned')]
},
u"lower": {
False: [('Lowercase', None)],
True: []
},
u"print": {
False: [('gc', 'Space_Separator'),
('na1', 'CHARACTER TABULATION')],
True: [('space', None), ('gc', 'Control'), ('gc', 'Surrogate'),
('gc', 'Unassigned')]
},
u"punct": {
False: [('gc', 'Punctuation')],
True: []
},
u"space": {
False: [('Whitespace', None)],
True: []
},
u"xdigit": {
False: [('gc', 'Decimal_Number'), ('Hex_Digit', None)],
True: []
},
u"upper": {
False: [('Uppercase', None)],
True: []
}
}
self.expect(u':')
class_name = ""
while self.next_is(string.ascii_letters):
class_name += self.get_next()
self.expect(":")
self.expect("]")
if self.is_unicode_defaults and class_name in unicode_classes:
class_queries = unicode_classes[class_name]
instance = UnicodeQuery.instance(self.unicode_db)
coverage = CoverageSet()
for k, v in class_queries[False]:
coverage.update(instance.query(k, v))
if len(class_queries[True]) > 0:
inverted_coverage = CoverageSet([(1, 0x10ffff)])
for k, v in class_queries[True]:
inverted_coverage.difference_update(instance.query(k, v))
coverage.update(inverted_coverage)
return coverage
elif not self.is_unicode_defaults and class_name in classes:
return CoverageSet(classes[class_name])
else:
raise RegexParserExceptionInternal(
"Character class '%s' not recognized" % class_name)
