def __get_remaining_set(self):
ignored = (E_CharacterCountType.BAD,
E_CharacterCountType.BEGIN_NEWLINE_SUPPRESSOR,
E_CharacterCountType.BEGIN_NEWLINE,
E_CharacterCountType.END_NEWLINE)
result = NumberSet()
for character_set, info in self.__map:
if info.cc_type in ignored: continue
result.unite_with(character_set)
return result.get_complement(Setup.buffer_codec.source_set)
def indentation_count_character_set(self):
"""Returns the superset of all characters that are involved in
indentation counting. That is the set of character that can
appear between newline and the first non whitespace character.
"""
result = NumberSet()
for character_set in self.space_db.values():
result.unite_with(character_set.get())
for character_set in self.grid_db.values():
result.unite_with(character_set.get())
return result
def indentation_count_character_set(self):
"""Returns the superset of all characters that are involved in
indentation counting. That is the set of character that can
appear between newline and the first non whitespace character.
"""
result = NumberSet()
for character_set in self.space_db.values():
result.unite_with(character_set.get())
for character_set in self.grid_db.values():
result.unite_with(character_set.get())
return result
def __wildcard_value_match(self, WildCardValue):
result = NumberSet()
value_list = self.get_wildcard_value_matches(WildCardValue)
if len(value_list) == 0:
return None
for value in value_list:
result.unite_with(NumberSet(self.code_point_db[value]))
# No decoupling, since result is computed each fresh and new
return result
def __wildcard_value_match(self, WildCardValue):
result = NumberSet()
value_list = self.get_wildcard_value_matches(WildCardValue)
if len(value_list) == 0:
return None
for value in value_list:
result.unite_with(NumberSet(self.code_point_db[value]))
# No decoupling, since result is computed each fresh and new
return result
def __whitespace_default(self):
"""Try to define default whitespace ' ' or '\t' if their positions
are not yet occupied in the count_command_map.
"""
cs0 = NumberSet(ord(" "))
cs1 = NumberSet(ord("\t"))
result = NumberSet()
if not self.count_command_map.find_occupier(cs0, set()):
result.unite_with(cs0)
if not self.count_command_map.find_occupier(cs1, set()):
result.unite_with(cs1)
if result.is_empty():
error_msg("Trying to implement default whitespace ' ' or '\\t' failed.\n"
"Characters are occupied by other elements.", self.sr)
return result
def is_DFA_compliant(self):
"""Checks if the current state transitions are DFA compliant, i.e. it
investigates if trigger sets pointing to different targets intersect.
RETURNS: True => OK
False => Same triggers point to different target. This cannot
be part of a deterministic finite automaton (DFA).
"""
# DFA's do not have epsilon transitions
if len(self.__epsilon_target_index_list) != 0: return False
# check whether trigger sets intersect
all_trigger_sets = NumberSet()
for trigger_set in self.__db.itervalues():
if all_trigger_sets.has_intersection(trigger_set):
return False
else:
all_trigger_sets.unite_with(trigger_set)
return True
def is_DFA_compliant(self):
"""Checks if the current state transitions are DFA compliant, i.e. it
investigates if trigger sets pointing to different targets intersect.
RETURN: True => OK
False => Same triggers point to different target. This cannot
be part of a deterministic finite automaton (DFA).
"""
# DFA's do not have epsilon transitions
if len(self.__epsilon_target_index_list) != 0: return False
# check whether trigger sets intersect
all_trigger_sets = NumberSet()
for trigger_set in self.__db.values():
if all_trigger_sets.has_intersection(trigger_set):
return False
else:
all_trigger_sets.unite_with(trigger_set)
return True
def get_character_set(self, Value=None):
"""Returns the character set that corresponds to 'Property==Value'.
'Value' can be a property value or a property value alias.
For binary properties 'Value' must be None.
"""
assert self.type != "Binary" or Value is None
def get_value_combination(CmbAlias):
result = []
for alias in self.alias_to_alias_combination_db[CmbAlias]:
name = self.alias_to_name_map.get(alias)
if name is None:
return "Unicode database error: no name related to alias '%s'" % alias
result.append(name)
return result
if self.type != "Binary" and Value is None:
return "Property '%s' requires a value setting.\n" % self.name + \
"Possible Values: " + \
self.get_value_list_help()
if self.code_point_db is None:
self.init_code_point_db()
if self.type == "Binary":
# Decouple, since we refer to an internal database
return deepcopy(self.code_point_db)
adapted_value = Value.replace(" ", "_")
if self.code_point_db.has_key(adapted_value):
# 'value' is present as name in the code point database
value = adapted_value
elif Value in self.alias_to_name_map.keys():
# 'value' is present as alias in code pointer database
value = self.alias_to_name_map[adapted_value]
elif Value in self.alias_to_alias_combination_db.keys():
# 'value' is present as a combination of aliases
value = get_value_combination(adapted_value)
elif self.name_to_alias_map.has_key(adapted_value):
# The value was a combination of values
value = get_value_combination(self.name_to_alias_map[adapted_value])
else:
# -- WILDCARD MATCH: Results in a list of property values
character_set = self.__wildcard_value_match(adapted_value)
if character_set is None:
return "Property '%s' cannot have a value or value alias '%s'.\n" % (self.name, Value) + \
"Possible Values: " + \
self.get_value_list_help()
# No need to decouple, since character is not a reference to
# internal database (for safety, do it)
return deepcopy(character_set)
if type(value) == list:
result = NumberSet()
for element in value:
if element == "Unassigned": continue
entry = self.code_point_db.get(element)
if entry is None:
return "%s/%s is not supported by Unicode database." % (self.name, repr(element))
result.unite_with(entry)
else:
result = self.code_point_db.get(value)
if result is None:
return "%s/%s is not supported by Unicode database." % (self.name, repr(value))
# Reference to internal database --> decouple with 'deepcopy'
return deepcopy(result)
def do(DB):
combined = NumberSet()
for number_set in DB.itervalues():
assert not number_set.has_intersection(combined)
combined.unite_with(number_set)
def get_trigger_set_union(self):
result = NumberSet()
for trigger_set in self.__db.itervalues():
result.unite_with(trigger_set)
return result
def do(DB):
combined = NumberSet()
for number_set in DB.itervalues():
assert not number_set.has_intersection(combined)
combined.unite_with(number_set)
def get_character_set(self, Value=None):
"""Returns the character set that corresponds to 'Property==Value'.
'Value' can be a property value or a property value alias.
For binary properties 'Value' must be None.
"""
assert self.type != "Binary" or Value is None
def get_value_combination(CmbAlias):
result = []
for alias in self.alias_to_alias_combination_db[CmbAlias]:
name = self.alias_to_name_map.get(alias)
if name is None:
return "Unicode database error: no name related to alias '%s'" % alias
result.append(name)
return result
if self.type != "Binary" and Value is None:
return "Property '%s' requires a value setting.\n" % self.name + \
"Possible Values: " + \
self.get_value_list_help()
if self.code_point_db is None:
self.init_code_point_db()
if self.type == "Binary":
# Decouple, since we refer to an internal database
return deepcopy(self.code_point_db)
adapted_value = Value.replace(" ", "_")
if self.code_point_db.has_key(adapted_value):
# 'value' is present as name in the code point database
value = adapted_value
elif Value in self.alias_to_name_map.keys():
# 'value' is present as alias in code pointer database
value = self.alias_to_name_map[adapted_value]
elif Value in self.alias_to_alias_combination_db.keys():
# 'value' is present as a combination of aliases
value = get_value_combination(adapted_value)
elif self.name_to_alias_map.has_key(adapted_value):
# The value was a combination of values
value = get_value_combination(
self.name_to_alias_map[adapted_value])
else:
# -- WILDCARD MATCH: Results in a list of property values
character_set = self.__wildcard_value_match(adapted_value)
if character_set is None:
return "Property '%s' cannot have a value or value alias '%s'.\n" % (self.name, Value) + \
"Possible Values: " + \
self.get_value_list_help()
# No need to decouple, since character is not a reference to
# internal database (for safety, do it)
return deepcopy(character_set)
if type(value) == list:
result = NumberSet()
for element in value:
if element == "Unassigned": continue
entry = self.code_point_db.get(element)
if entry is None:
return "%s/%s is not supported by Unicode database." % (
self.name, repr(element))
result.unite_with(entry)
else:
result = self.code_point_db.get(value)
if result is None:
return "%s/%s is not supported by Unicode database." % (
self.name, repr(value))
# Reference to internal database --> decouple with 'deepcopy'
return deepcopy(result)
def get_trigger_set_union(self):
result = NumberSet()
for trigger_set in self.__db.values():
result.unite_with(trigger_set)
return result
def _get_all_character_set(*DbList):
result = NumberSet()
for db in DbList:
for character_set in db.itervalues():
result.unite_with(character_set)
return result
def covers(self, Min, Max):
result = NumberSet()
for info in self.__map:
result.unite_with(info.character_set)
return result.covers_range(Min, Max)