cablegates/pandora/item/utils.py

186 lines
6.1 KiB
Python
Raw Normal View History

2009-06-08 16:08:59 +00:00
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# vi:si:et:sw=4:sts=4:ts=4
#
import errno
import os
import sys
import re
import hashlib
import unicodedata
2009-06-08 16:08:59 +00:00
2010-07-07 22:46:41 +00:00
import ox
import ox.iso
2010-09-03 13:28:44 +00:00
from ox.normalize import normalizeName, normalizeTitle, canonicalTitle
2009-06-08 16:08:59 +00:00
2010-08-24 17:16:33 +00:00
def plural_key(term):
return {
'country': 'countries',
}.get(term, term + 's')
def oxid(title, directors, year='', seriesTitle='', episodeTitle='', season=0, episode=0):
director = ', '.join(directors)
2009-06-08 16:08:59 +00:00
oxid_value = u"\n".join([title, director, year])
oxid = hashlib.sha1(oxid_value.encode('utf-8')).hexdigest()
if seriesTitle:
oxid_value = u"\n".join([seriesTitle, "%02d" % season])
oxid = hashlib.sha1(oxid_value.encode('utf-8')).hexdigest()[:20]
oxid_value = u"\n".join(["%02d" % episode, episodeTitle, director, year])
oxid += hashlib.sha1(oxid_value.encode('utf-8')).hexdigest()[:20]
return u"0x" + oxid
2010-11-28 16:03:23 +00:00
def oxdb_id(title, directors=[], year='', season='', episode='', episode_title='', episode_directors=[], episode_year=''):
2010-09-17 14:11:37 +00:00
# new id function, will replace oxid()
def get_hash(string):
return hashlib.sha1(string.encode('utf-8')).hexdigest().upper()
director = ', '.join(directors)
2010-11-28 16:03:23 +00:00
episode_director = ', '.join(episode_directors)
2010-09-17 14:11:37 +00:00
if not episode:
2010-12-07 18:35:55 +00:00
oxdb_id = get_hash(director)[:8] + get_hash('\n'.join([title, str(year)]))[:8]
2010-09-17 14:11:37 +00:00
else:
2010-12-07 18:35:55 +00:00
oxdb_id = get_hash('\n'.join([director, title, str(year), str(season)]))[:8] + \
get_hash('\n'.join([str(episode), episode_director, episode_title, str(episode_year)]))[:8]
2010-09-17 14:11:37 +00:00
return u'0x' + oxdb_id
def oxdb_directors(director):
2010-12-08 00:30:45 +00:00
director = os.path.basename(os.path.dirname(director))
if director.endswith('_'):
director = "%s." % director[:-1]
directors = [normalizeName(d) for d in director.split('; ')]
def cleanup(director):
director = director.strip()
director = director.replace('Series', '')
director = director.replace('Unknown Director', '')
director = director.replace('Various Directors', '')
return director
directors = filter(None, [cleanup(d) for d in directors])
return directors
def oxdb_title(_title, searchTitle = False):
'''
2010-09-23 16:01:48 +00:00
normalize filename to get item title
'''
_title = os.path.basename(_title)
_title = _title.replace('... ', '_dot_dot_dot_')
_title = _title.replace('. ', '_dot__space_')
_title = _title.replace(' .', '_space__dot_')
title = _title.split('.')[0]
title = re.sub('([a-z0-9])_ ', '\\1: ', title)
se = re.compile('Season (\d+).Episode (\d+)').findall(_title)
if se:
se = "S%02dE%02d" % (int(se[0][0]), int(se[0][1]))
if 'Part' in _title.split('.')[-2] and 'Episode' not in _title.split('.')[-3]:
stitle = _title.split('.')[-3]
else:
stitle = _title.split('.')[-2]
if stitle.startswith('Episode '):
stitle = ''
if searchTitle:
title = '"%s" %s' % (title, stitle)
else:
title = '%s (%s) %s' % (title, se, stitle)
title = title.strip()
title = title.replace('_dot_dot_dot_', '... ')
title = title.replace('_dot__space_', '. ')
title = title.replace('_space__dot_', ' .')
2010-08-07 14:31:20 +00:00
year = ox.findRe(title, '(\(\d{4}\))')
if year and title.endswith(year):
2010-08-07 14:31:20 +00:00
title = title[:-len(year)].strip()
title = normalizeTitle(title)
return title
def oxdb_year(data):
2010-07-07 22:46:41 +00:00
return ox.findRe(data, '\.(\d{4})\.')
def oxdb_series_title(path):
seriesTitle = u''
if path.startswith('Series'):
2010-12-08 00:30:45 +00:00
seriesTitle = os.path.basename(path)
else:
t = oxdb_title(path)
if " (S" in t:
seriesTitle = t.split(" (S")[0]
return seriesTitle
def oxdb_episode_title(path):
episodeTitle = u''
ep = re.compile('.Episode \d+?\.(.*?)\.[a-zA-Z]').findall(path)
if ep:
2010-01-22 23:57:06 +00:00
episodeTitle = ep[0]
return episodeTitle
def oxdb_season_episode(path):
season = 0
episode = 0
path = os.path.basename(path)
se = re.compile('Season (\d+).Episode (\d+)').findall(path)
if se:
season = int(se[0][0])
episode = int(se[0][1])
else:
ep = re.compile('.Episode (\d+?)').findall(path)
if ep:
episode = int(ep[0][0])
if season == 0 and episode == 0:
se = re.compile('S(\d\d)E(\d\d)').findall(path)
if se:
season = int(se[0][0])
episode = int(se[0][1])
return (season, episode)
def oxdb_part(path):
part = 1
path = path.lower()
p = re.compile('part\s*?(\d+)\.').findall(path)
if p:
part = p[0]
else:
p = re.compile('cd\s*?(\d+)\.').findall(path)
if p:
part = p[0]
return part
2010-08-07 14:31:20 +00:00
def parse_path(path):
2010-12-08 00:30:45 +00:00
'''
expects path in the form
L/Last, First/Title (YYYY)
M/McCarthy, Thomas/The Visitor (2007)
G/Godard, Jean-Luc/Histoire(s) du cinema_ Toutes les histoires (1988)
'''
2010-07-07 22:46:41 +00:00
import ox.web.imdb
search_title = oxdb_title(path, True)
r = {}
r['title'] = oxdb_title(path)
r['directors'] = oxdb_directors(path)
2010-08-07 14:31:20 +00:00
year = ox.findRe(path, '\((\d{4})\)')
if year:
r['year'] = year
#FIXME: only include it its actually a series
r['episode_title'] = oxdb_episode_title(path)
r['season'], r['episode'] = oxdb_season_episode(path)
r['series_title'] = oxdb_series_title(path)
2010-08-07 14:31:20 +00:00
r['imdbId'] = ox.web.imdb.guess(search_title, ', '.join(r['directors']), timeout=-1)
r['oxdbId'] = oxdb_id(r['title'], r['directors'], r.get('year', ''),
r.get('season', ''), r.get('episode', ''),
episode_title=r['episode_title'],
episode_directors=[],
episode_year='')
return r
2010-09-03 13:28:44 +00:00
def sort_title(title):
#title
title = re.sub(u'[\'!¿¡,\.;\-"\:\*\[\]]', '', title)
#title = title.replace(u'Æ', 'Ae')
2010-11-26 17:16:57 +00:00
if isinstance(title, str):
title = unicode(title)
title = unicodedata.normalize('NFKD',title)
2010-09-03 13:28:44 +00:00
#pad numbered titles
title = re.sub('(\d+)', lambda x: '%010d' % int(x.group(0)), title)
return title.strip()