def search_for_release_group(self):
token_end = None
while True:
# Find the first enclosed unknown token
if token_end:
token_begin = Tokens.find_next(
token_end, TokenFlags.ENCLOSED | TokenFlags.UNKNOWN)
else:
token_begin = Tokens.find(
TokenFlags.ENCLOSED | TokenFlags.UNKNOWN)
if token_begin is None:
return
# Continue until a bracket or identifier is found
token_end = Tokens.find_next(
token_begin, TokenFlags.BRACKET | TokenFlags.IDENTIFIER)
if token_end is None:
return
if token_end.category != TokenCategory.BRACKET:
continue
# Ignore if it's not the first non-delimiter token in group
previous_token = Tokens.find_previous(
token_begin, TokenFlags.NOT_DELIMITER)
if previous_token is not None and \
previous_token.category != TokenCategory.BRACKET:
continue
# Build release group, token end is a bracket, so we get the
# previous token to be included in the element
token_end = Tokens.find_previous(token_end, TokenFlags.VALID)
parser_helper.build_element(
ElementCategory.RELEASE_GROUP, token_begin, token_end,
keep_delimiters=True)
return
def search_for_episode_title(self):
token_end = None
while True:
# Find the first non-enclosed unknown token
if token_end:
token_begin = Tokens.find_next(
token_end, TokenFlags.NOT_ENCLOSED | TokenFlags.UNKNOWN)
else:
token_begin = Tokens.find(
TokenFlags.NOT_ENCLOSED | TokenFlags.UNKNOWN)
if token_begin is None:
return
# Continue until a bracket or identifier is found
token_end = Tokens.find_next(
token_begin, TokenFlags.BRACKET | TokenFlags.IDENTIFIER)
# Ignore if it's only a dash
if Tokens.distance(token_begin, token_end) <= 2 and \
parser_helper.is_dash_character(token_begin.content):
continue
# If token end is a bracket, then we get the previous token to be
# included in the element
if token_end and token_end.category == TokenCategory.BRACKET:
token_end = Tokens.find_previous(token_end, TokenFlags.VALID)
# Build episode title
parser_helper.build_element(
ElementCategory.EPISODE_TITLE, token_begin, token_end,
keep_delimiters=False)
return
def search_for_equivalent_numbers(tokens):
for token in tokens:
if parser_helper.is_token_isolated(token) or \
not is_valid_episode_number(token.content):
continue
# Find the first enclosed, non-delimiter token
next_token = Tokens.find_next(token, TokenFlags.NOT_DELIMITER)
if next_token is None or next_token.category != TokenCategory.BRACKET:
continue
next_token = Tokens.find_next(
next_token, TokenFlags.ENCLOSED | TokenFlags.NOT_DELIMITER)
if next_token.category != TokenCategory.UNKNOWN:
continue
# Check if it's an isolated number
if not parser_helper.is_token_isolated(next_token) or \
not next_token.content.isdigit() or \
not is_valid_episode_number(next_token.content):
continue
episode = min(token, next_token, key=lambda t: int(t.content))
alt_episode = max(token, next_token, key=lambda t: int(t.content))
set_episode_number(episode.content, episode, validate=False)
set_alternative_episode_number(alt_episode.content, alt_episode)
return True
return False
def number_comes_before_another_number(token):
separator_token = Tokens.find_next(token, TokenFlags.NOT_DELIMITER)
if separator_token:
separator = separator_token.content
if separator == '&' or separator == 'of':
other_token = Tokens.find_next(separator_token,
TokenFlags.NOT_DELIMITER)
if other_token and other_token.content.isdigit():
set_episode_number(token.content, token, validate=False)
if separator == '&':
set_episode_number(other_token.content,
token,
validate=False)
separator_token.category = TokenCategory.IDENTIFIER
other_token.category = TokenCategory.IDENTIFIER
return True
return False
def is_token_isolated(token):
previous_token = Tokens.find_previous(token, TokenFlags.NOT_DELIMITER)
if previous_token.category != TokenCategory.BRACKET:
return False
next_token = Tokens.find_next(token, TokenFlags.NOT_DELIMITER)
if next_token.category != TokenCategory.BRACKET:
return False
return True
def check_anime_season_keyword(token):
def set_anime_season(first, second, content):
Elements.insert(ElementCategory.ANIME_SEASON, content)
first.category = TokenCategory.IDENTIFIER
second.category = TokenCategory.IDENTIFIER
previous_token = Tokens.find_previous(token, TokenFlags.NOT_DELIMITER)
if previous_token:
number = get_number_from_ordinal(previous_token.content)
if number:
set_anime_season(previous_token, token, number)
return True
next_token = Tokens.find_next(token, TokenFlags.NOT_DELIMITER)
if next_token and next_token.content.isdigit():
set_anime_season(token, next_token, next_token.content)
return True
return False
def check_extent_keyword(category, token):
next_token = Tokens.find_next(token, TokenFlags.NOT_DELIMITER)
if next_token.category == TokenCategory.UNKNOWN:
if next_token and \
parser_helper.find_number_in_string(next_token.content) \
is not None:
if category == ElementCategory.EPISODE_NUMBER:
if not match_episode_patterns(next_token.content, next_token):
set_episode_number(next_token.content,
next_token,
validate=False)
elif category == ElementCategory.VOLUME_NUMBER:
if not match_volume_patterns(next_token.content, next_token):
set_volume_number(next_token.content,
next_token,
validate=False)
else:
return False
token.category = TokenCategory.IDENTIFIER
return True
return False
def search_for_anime_title(self):
enclosed_title = False
# Find the first non-enclosed unknown token
token_begin = Tokens.find(TokenFlags.NOT_ENCLOSED | TokenFlags.UNKNOWN)
# If that doesn't work, find the first unknown token in the second
# enclosed group, assuming that the first one is the release group
if token_begin is None:
enclosed_title = True
token_begin = Tokens.get(0)
skipped_previous_group = False
while token_begin is not None:
token_begin = Tokens.find_next(token_begin, TokenFlags.UNKNOWN)
if token_begin is None:
break
# Ignore groups that are composed of non-Latin characters
if parser_helper.is_mostly_latin_string(token_begin.content):
if skipped_previous_group:
break # Found it
# Get the first unknown token of the next group
token_begin = Tokens.find_next(token_begin, TokenFlags.BRACKET)
skipped_previous_group = True
if token_begin is None:
return
# Continue until an identifier (or a bracket, if the title is enclosed)
# is found
token_end = Tokens.find_next(
token_begin, TokenFlags.IDENTIFIER | (
TokenFlags.BRACKET if enclosed_title else TokenFlags.NONE
))
# If within the interval there's an open bracket without its matching
# pair, move the upper endpoint back to the bracket
if not enclosed_title:
last_bracket = token_end
bracket_open = False
for token in Tokens.get_list(TokenFlags.BRACKET, begin=token_begin,
end=token_end):
last_bracket = token
bracket_open = not bracket_open
if bracket_open:
token_end = last_bracket
# If the interval ends with an enclosed group (e.g. "Anime Title
# [Fansub]"), move the upper endpoint back to the beginning of the
# group. We ignore parentheses in order to keep certain groups (e.g.
# "(TV)") intact.
if not enclosed_title:
token = Tokens.find_previous(token_end, TokenFlags.NOT_DELIMITER)
while token.category == TokenCategory.BRACKET and \
token.content != ')':
token = Tokens.find_previous(token, TokenFlags.BRACKET)
if token is not None:
token_end = token
token = Tokens.find_previous(
token_end, TokenFlags.NOT_DELIMITER)
# Token end is a bracket, so we get the previous token to be included
# in the element
token_end = Tokens.find_previous(token_end, TokenFlags.VALID)
parser_helper.build_element(ElementCategory.ANIME_TITLE, token_begin,
token_end, keep_delimiters=False)
def find_next_valid_token(token):
return Tokens.find_next(token, TokenFlags.VALID)