def parse(self, data=None):
"""Parse movie name. Populates name, year, quality and proper_count attributes"""
# Reset before parsing, so the parser can be reused.
self.reset()
if not data:
data = self.data
for char in '[]()_,.':
data = data.replace(char, ' ')
# if there are no spaces
if data.find(' ') == -1:
data = data.replace('-', ' ')
# remove unwanted words (imax, ..)
self.remove_words(data, self.remove)
data = self.strip_spaces(data)
# split to parts
parts = data.split(' ')
year = None
cut_part = 256
self.quality = 'unknown'
for part in parts:
cut = False
# check for year
if part.isdigit():
num = int(part)
if num > 1930 and num < 2050:
year = part
cut = True
# if length > 3 and whole word in uppers, consider as cut word (most likely a group name)
if len(part) > 3 and part.isupper() and part.isalpha():
cut = True
# check for cutoff words
if part.lower() in self.cutoffs:
cut = True
# check for propers
if part.lower() in self.propers:
self.proper_count += 1
cut = True
# update cut position
if cut and parts.index(part) < cut_part:
cut_part = parts.index(part)
if cut_part != 256:
log.debug('parts: %s, cut is: %s' % (parts, parts[cut_part]))
# calculate cut positon from cut_part
abs_cut = len(' '.join(parts[:cut_part]))
log.debug('after parts check, cut data would be: `%s` abs_cut: %i' % (data[:abs_cut], abs_cut))
# parse quality
quality, remaining = qualities.quality_match(data)
if quality:
self.quality = quality
# remaining string is same as data but quality information removed
# find out position where there is first difference, this is earlies
# quality bit, anything after that has no relevance to the movie name
#dp = diff_pos(data, remaining)
dp = None
if dp is not None:
log.debug('quality start: %s' % dp)
if dp < abs_cut:
log.debug('quality cut is even shorter')
abs_cut = dp
# make cut
data = data[:abs_cut].strip()
log.debug('data cut to `%s` - this will be the name' % data)
# save results
self.name = data
if year:
if year.isdigit():
self.year = int(year)
def parse(self, data=None, field=None, quality=qualities.UNKNOWN):
# Clear the output variables before parsing
self._reset()
self.field = field
self.quality = quality
if data:
self.data = data
if not self.name or not self.data:
raise Exception('SeriesParser initialization error, name: %s data: %s' % \
(repr(self.name), repr(self.data)))
if self.expect_ep and self.expect_id:
raise Exception('Flags expect_ep and expect_id are mutually exclusive')
name = self.remove_dirt(self.name)
# check if data appears to be unwanted (abort)
if self.parse_unwanted(self.remove_dirt(self.data)):
return
def name_to_re(name):
"""Convert 'foo bar' to '^[^...]*foo[^...]*bar[^...]+"""
# TODO: Still doesn't handle the case where the user wants
# "Schmost" and the feed contains "Schmost at Sea".
blank = r'[\W_]'
ignore = '(?:' + '|'.join(self.ignore_prefixes) + ')?'
# accept either '&' or 'and'
name = name.replace('&', '(?:and|&)')
res = re.sub(blank + '+', ' ', name)
res = res.strip()
# check for 'and' surrounded by spaces so it is not replaced within a word or from above replacement
res = res.replace(' and ', ' (?:and|&) ')
res = re.sub(' +', blank + '*', res)
res = '^' + ignore + blank + '*' + '(' + res + ')' + blank + '+'
return res
log.debug('name: %s data: %s' % (name, self.data))
# name end position
name_start = 0
name_end = 0
# regexp name matching
if not self.name_regexps:
# if we don't have name_regexps, generate one from the name
self.name_regexps = [name_to_re(name)]
self.re_from_name = True
# try all specified regexps on this data
for name_re in self.name_regexps:
match = re.search(name_re, self.data)
if match:
if self.re_from_name:
name_start, name_end = match.span(1)
else:
name_start, name_end = match.span()
log.debug('NAME SUCCESS: %s matched to %s' % (name_re.pattern, self.data))
break
else:
# leave this invalid
log.debug('FAIL: name regexps %s do not match %s' % ([regexp.pattern for regexp in self.name_regexps],
self.data))
return
# remove series name from raw data, move any prefix to end of string
data_stripped = self.data[name_end:] + ' ' + self.data[:name_start]
data_stripped = data_stripped.lower()
log.debug('data stripped: %s' % data_stripped)
# allow group(s)
if self.allow_groups:
for group in self.allow_groups:
group = group.lower()
for fmt in ['[%s]', '-%s']:
if fmt % group in data_stripped:
log.debug('%s is from group %s' % (self.data, group))
self.group = group
data_stripped = data_stripped.replace(fmt % group, '')
break
if self.group:
break
else:
log.debug('%s is not from groups %s' % (self.data, self.allow_groups))
return # leave invalid
# search tags and quality if one was not provided to parse method
if not quality or quality == qualities.UNKNOWN:
log.debug('parsing quality ->')
quality, remaining = qualities.quality_match(data_stripped)
self.quality = quality
if remaining:
# Remove quality string from data
log.debug('quality detected, using remaining data `%s`' % remaining)
data_stripped = remaining
# Remove unwanted words (qualities and such) from data for ep / id parsing
data_stripped = self.remove_words(data_stripped, self.remove + qualities.registry.keys() +
self.codecs + self.sounds, not_in_word=True)
data_parts = re.split('[\W_]+', data_stripped)
for part in data_parts[:]:
if part in self.propers:
self.proper_count += 1
data_parts.remove(part)
elif part in self.specials:
self.special = True
data_parts.remove(part)
data_stripped = ' '.join(data_parts).strip()
log.debug("data for id/ep parsing '%s'" % data_stripped)
ep_match = self.parse_episode(data_stripped)
if ep_match:
# strict_name
if self.strict_name:
if ep_match['match'].start() > 1:
return
if self.expect_id:
log.debug('found episode number, but expecting id, aborting!')
return
if ep_match['end_episode'] > ep_match['episode'] + 2:
# This is a pack of too many episodes, ignore it.
log.debug('Series pack contains too many episodes (%d). Rejecting' %
(ep_match['end_episode'] - ep_match['episode']))
return
self.season = ep_match['season']
self.episode = ep_match['episode']
self.end_episode = ep_match['end_episode']
self.valid = True
return
log.debug('-> no luck with ep_regexps')
# search for ids later as last since they contain somewhat broad matches
if self.expect_ep:
# we should be getting season, ep !
# try to look up idiotic numbering scheme 101,102,103,201,202
# ressu: Added matching for 0101, 0102... It will fail on
# season 11 though
log.debug('expect_ep enabled')
match = re.search(self.re_not_in_word(r'(0?\d)(\d\d)'), data_stripped, re.IGNORECASE | re.UNICODE)
if match:
# strict_name
if self.strict_name:
if match.start() > 1:
return
self.season = int(match.group(1))
self.episode = int(match.group(2))
log.debug(self)
self.valid = True
return
log.debug('-> no luck with the expect_ep')
else:
if self.parse_unwanted_id(data_stripped):
return
for id_re in self.id_regexps:
match = re.search(id_re, data_stripped)
if match:
# strict_name
if self.strict_name:
if match.start() - name_end >= 2:
return
if 'version' in match.groupdict():
if match.group('version'):
self.proper_count = int(match.group('version')) - 1
self.id = match.group(1)
else:
self.id = '-'.join(match.groups())
self.id_groups = match.groups()
if self.special:
self.id += '-SPECIAL'
self.valid = True
log.debug('found id \'%s\' with regexp \'%s\'' % (self.id, id_re.pattern))
return
log.debug('-> no luck with id_regexps')
# No id found, check if this is a special
if self.special:
# Attempt to set id as the title of the special
self.id = data_stripped
self.valid = True
log.debug('found special, setting id to \'%s\'' % self.id)
return
raise ParseWarning('Title \'%s\' looks like series \'%s\' but I cannot find any episode or id numbering' % (self.data, self.name))
def parse(self, data=None):
"""Parse movie name. Populates name, year, quality and proper_count attributes"""
# Reset before parsing, so the parser can be reused.
self.reset()
if not data:
data = self.data
for char in '[]()_,.':
data = data.replace(char, ' ')
# if there are no spaces
if data.find(' ') == -1:
data = data.replace('-', ' ')
# remove unwanted words (imax, ..)
self.remove_words(data, self.remove)
data = self.strip_spaces(data)
# split to parts
parts = data.split(' ')
year = None
cut_part = 256
self.quality = 'unknown'
for part in parts:
cut = False
# check for year
if part.isdigit():
num = int(part)
if num > 1930 and num < 2050:
year = part
cut = True
# if length > 3 and whole word in uppers, consider as cut word (most likely a group name)
if len(part) > 3 and part.isupper() and part.isalpha():
cut = True
# check for cutoff words
if part.lower() in self.cutoffs:
cut = True
# check for propers
if part.lower() in self.propers:
self.proper_count += 1
cut = True
# update cut position
if cut and parts.index(part) < cut_part:
cut_part = parts.index(part)
if cut_part != 256:
log.debug('parts: %s, cut is: %s' % (parts, parts[cut_part]))
# calculate cut positon from cut_part
abs_cut = len(' '.join(parts[:cut_part]))
log.debug('after parts check, cut data would be: `%s` abs_cut: %i' % (data[:abs_cut], abs_cut))
# parse quality
quality, remaining = qualities.quality_match(data)
if quality:
self.quality = quality
# remaining string is same as data but quality information removed
# find out position where there is first difference, this is earlies
# quality bit, anything after that has no relevance to the movie name
#dp = diff_pos(data, remaining)
dp = None
if dp is not None:
log.debug('quality start: %s' % dp)
if dp < abs_cut:
log.debug('quality cut is even shorter')
abs_cut = dp
# make cut
data = data[:abs_cut].strip()
log.debug('data cut to `%s` - this will be the name' % data)
# save results
self.name = data
if year:
if year.isdigit():
self.year = int(year)
def parse(self, data=None, field=None, quality=qualities.UNKNOWN):
# Clear the output variables before parsing
self._reset()
self.field = field
self.quality = quality
if data:
self.data = data
if not self.name or not self.data:
raise Exception('SeriesParser initialization error, name: %s data: %s' % \
(repr(self.name), repr(self.data)))
if self.expect_ep and self.expect_id:
raise Exception(
'Flags expect_ep and expect_id are mutually exclusive')
name = self.remove_dirt(self.name)
# check if data appears to be unwanted (abort)
if self.parse_unwanted(self.remove_dirt(self.data)):
return
def name_to_re(name):
"""Convert 'foo bar' to '^[^...]*foo[^...]*bar[^...]+"""
# TODO: Still doesn't handle the case where the user wants
# "Schmost" and the feed contains "Schmost at Sea".
blank = r'[\W_]'
ignore = '(?:' + '|'.join(self.ignore_prefixes) + ')?'
# accept either '&' or 'and'
name = name.replace('&', '(?:and|&)')
res = re.sub(blank + '+', ' ', name)
res = res.strip()
# check for 'and' surrounded by spaces so it is not replaced within a word or from above replacement
res = res.replace(' and ', ' (?:and|&) ')
res = re.sub(' +', blank + '*', res)
res = '^' + ignore + blank + '*' + '(' + res + ')' + blank + '+'
return res
log.debug('name: %s data: %s' % (name, self.data))
# name end position
name_start = 0
name_end = 0
# regexp name matching
if not self.name_regexps:
# if we don't have name_regexps, generate one from the name
self.name_regexps = [name_to_re(name)]
self.re_from_name = True
# try all specified regexps on this data
for name_re in self.name_regexps:
match = re.search(name_re, self.data)
if match:
if self.re_from_name:
name_start, name_end = match.span(1)
else:
name_start, name_end = match.span()
log.debug('NAME SUCCESS: %s matched to %s' %
(name_re.pattern, self.data))
break
else:
# leave this invalid
log.debug('FAIL: name regexps %s do not match %s' %
([regexp.pattern
for regexp in self.name_regexps], self.data))
return
# remove series name from raw data, move any prefix to end of string
data_stripped = self.data[name_end:] + ' ' + self.data[:name_start]
data_stripped = data_stripped.lower()
log.debug('data stripped: %s' % data_stripped)
# allow group(s)
if self.allow_groups:
for group in self.allow_groups:
group = group.lower()
for fmt in ['[%s]', '-%s']:
if fmt % group in data_stripped:
log.debug('%s is from group %s' % (self.data, group))
self.group = group
data_stripped = data_stripped.replace(fmt % group, '')
break
if self.group:
break
else:
log.debug('%s is not from groups %s' %
(self.data, self.allow_groups))
return # leave invalid
# search tags and quality if one was not provided to parse method
if not quality or quality == qualities.UNKNOWN:
log.debug('parsing quality ->')
quality, remaining = qualities.quality_match(data_stripped)
self.quality = quality
if remaining:
# Remove quality string from data
log.debug('quality detected, using remaining data `%s`' %
remaining)
data_stripped = remaining
# Remove unwanted words (qualities and such) from data for ep / id parsing
data_stripped = self.remove_words(
data_stripped,
self.remove + qualities.registry.keys() + self.codecs +
self.sounds,
not_in_word=True)
data_parts = re.split('[\W_]+', data_stripped)
for part in data_parts[:]:
if part in self.propers:
self.proper_count += 1
data_parts.remove(part)
elif part in self.specials:
self.special = True
data_parts.remove(part)
data_stripped = ' '.join(data_parts).strip()
log.debug("data for id/ep parsing '%s'" % data_stripped)
ep_match = self.parse_episode(data_stripped)
if ep_match:
# strict_name
if self.strict_name:
if ep_match['match'].start() > 1:
return
if self.expect_id:
log.debug('found episode number, but expecting id, aborting!')
return
if ep_match['end_episode'] > ep_match['episode'] + 2:
# This is a pack of too many episodes, ignore it.
log.debug(
'Series pack contains too many episodes (%d). Rejecting' %
(ep_match['end_episode'] - ep_match['episode']))
return
self.season = ep_match['season']
self.episode = ep_match['episode']
self.end_episode = ep_match['end_episode']
self.valid = True
return
log.debug('-> no luck with ep_regexps')
# search for ids later as last since they contain somewhat broad matches
if self.expect_ep:
# we should be getting season, ep !
# try to look up idiotic numbering scheme 101,102,103,201,202
# ressu: Added matching for 0101, 0102... It will fail on
# season 11 though
log.debug('expect_ep enabled')
match = re.search(self.re_not_in_word(r'(0?\d)(\d\d)'),
data_stripped, re.IGNORECASE | re.UNICODE)
if match:
# strict_name
if self.strict_name:
if match.start() > 1:
return
self.season = int(match.group(1))
self.episode = int(match.group(2))
log.debug(self)
self.valid = True
return
log.debug('-> no luck with the expect_ep')
else:
if self.parse_unwanted_id(data_stripped):
return
for id_re in self.id_regexps:
match = re.search(id_re, data_stripped)
if match:
# strict_name
if self.strict_name:
if match.start() - name_end >= 2:
return
if 'version' in match.groupdict():
if match.group('version'):
self.proper_count = int(match.group('version')) - 1
self.id = match.group(1)
else:
self.id = '-'.join(match.groups())
self.id_groups = match.groups()
if self.special:
self.id += '-SPECIAL'
self.valid = True
log.debug('found id \'%s\' with regexp \'%s\'' %
(self.id, id_re.pattern))
return
log.debug('-> no luck with id_regexps')
# No id found, check if this is a special
if self.special:
# Attempt to set id as the title of the special
self.id = data_stripped
self.valid = True
log.debug('found special, setting id to \'%s\'' % self.id)
return
raise ParseWarning(
'Title \'%s\' looks like series \'%s\' but I cannot find any episode or id numbering'
% (self.data, self.name))
