def getFilmography(dataF,
indexF,
keyF,
attrIF,
attrKF,
offset,
charNF=None,
doCast=0,
doWriters=0):
"""Gather information from the given files about the
person entry found at offset; return a list of Movie objects,
with the relevant attributes."""
name, res = getRawData(dataF, offset, doCast, doWriters)
resList = []
for movie in res:
title = getLabel(movie['movieID'], indexF, keyF)
if not title: continue
curRole = movie.get('currentRole', u'')
roleID = None
if curRole and charNF:
curRole, roleID = getCharactersIDs(curRole, charNF)
m = Movie(title=title,
movieID=movie['movieID'],
currentRole=curRole,
roleID=roleID,
accessSystem='local')
if movie.has_key('attributeID'):
attr = getLabel(movie['attributeID'], attrIF, attrKF)
if attr: m.notes = attr
resList.append(m)
return resList
def _buildEpisodes(self, eps_list, parentID):
episodes = {}
parentTitle = getLabel(parentID, '%stitles.index' % self.__db,
'%stitles.key' % self.__db)
parentSeries = Movie(title=parentTitle,
movieID=parentID,
accessSystem='local')
for episodeID, episodeTitle in eps_list:
episodeTitle = unicode(episodeTitle, 'latin_1', 'replace')
data = analyze_title(episodeTitle, canonical=1)
m = Movie(data=data, movieID=episodeID, accessSystem='local')
m['episode of'] = parentSeries
if data.get('year') is None:
year = getFullIndex('%smovies.data' % self.__db,
key=episodeID,
kind='moviedata',
rindex=1)
if year: m['year'] = year
season = data.get('season', 'UNKNOWN')
if not episodes.has_key(season): episodes[season] = {}
ep_number = data.get('episode')
if ep_number is None:
ep_number = max((episodes[season].keys() or [0])) + 1
episodes[season][ep_number] = m
return episodes
def _buildGuests(gl):
"""Return a list of Movie objects from a list of GA lines."""
rl = []
rlapp = rl.append
for g in gl:
# When used by the imdbpy2sql.py script, latin_1 strings are passed.
if not isinstance(g, UnicodeType):
g = unicode(g, 'latin_1', 'replace')
titl = re_titleRef.findall(g)
if len(titl) != 1: continue
note = u''
if g[-1] == ')':
opi = g.rfind('(episode')
if opi == -1: opi = g.rfind('(')
if opi != -1:
note = g[opi:].replace('_', '"').strip()
g = g[:opi].strip()
cr = u''
cri = g.find('_ (qv), as ')
if cri != -1:
cr = g[cri + 11:].replace('[unknown]', u'').strip()
if cr and cr[-1] == ')':
opi = cr.rfind('(')
if opi != -1:
if note: note += ' '
note += cr[opi:]
cr = cr[:opi].strip()
# As you can see, we've no notion of the movieID, here.
m = Movie(title=titl[0],
currentRole=cr,
notes=note,
accessSystem='local')
rlapp(m)
return rl
def postprocess_data(self, data):
result = {}
for item in ('names refs', 'titles refs', 'characters refs'):
result[item] = {}
for k, v in data.get(item, []):
k = k.strip()
v = v.strip()
if not (k and v):
continue
if not v.endswith('/'): continue
imdbID = analyze_imdbid(v)
if item == 'names refs':
obj = Person(personID=imdbID,
name=k,
accessSystem=self._as,
modFunct=self._modFunct)
elif item == 'titles refs':
obj = Movie(movieID=imdbID,
title=k,
accessSystem=self._as,
modFunct=self._modFunct)
else:
obj = Character(characterID=imdbID,
name=k,
accessSystem=self._as,
modFunct=self._modFunct)
# XXX: companies aren't handled: are they ever found in text,
# as links to their page?
result[item][k] = obj
return result
def _base_person_info(self,
personID,
movies_cache=None,
persons_cache=None):
if movies_cache is None:
movies_cache = {}
if persons_cache is None:
persons_cache = {}
if personID in persons_cache:
return persons_cache[personID]
nb = self.T['name_basics']
person = nb.select(nb.c.nconst == personID).execute().fetchone() or {}
data = self._rename('name_basics', dict(person))
movies = []
for movieID in split_array(data.get('known for') or ''):
if not movieID:
continue
movieID = int(movieID)
movie_data = self._base_title_info(movieID,
movies_cache=movies_cache,
persons_cache=persons_cache)
movie = Movie(movieID=movieID,
data=movie_data,
accessSystem=self.accessSystem)
movies.append(movie)
data['known for'] = movies
self._clean(data,
('ns_soundex', 'sn_soundex', 's_soundex', 'personID'))
persons_cache[personID] = data
return data
def get_movie_data(movieID, kindDict, fromAka=0):
"""Return a dictionary containing data about the given movieID;
if fromAka is true, the AkaTitle table is searched."""
if not fromAka: Table = Title
else: Table = AkaTitle
m = Table.get(movieID)
mdict = {
'title': m.title,
'kind': kindDict[m.kindID],
'year': m.productionYear,
'imdbIndex': m.imdbIndex,
'season': m.seasonNr,
'episode': m.episodeNr
}
if not fromAka:
if m.seriesYears is not None:
mdict['series years'] = unicode(m.seriesYears)
if mdict['imdbIndex'] is None: del mdict['imdbIndex']
if mdict['year'] is None: del mdict['year']
else:
try:
mdict['year'] = int(mdict['year'])
except (TypeError, ValueError):
del mdict['year']
if mdict['season'] is None: del mdict['season']
else:
try:
mdict['season'] = int(mdict['season'])
except:
pass
if mdict['episode'] is None: del mdict['episode']
else:
try:
mdict['episode'] = int(mdict['episode'])
except:
pass
episodeOfID = m.episodeOfID
if episodeOfID is not None:
ser_dict = get_movie_data(episodeOfID, kindDict, fromAka)
mdict['episode of'] = Movie(data=ser_dict,
movieID=episodeOfID,
accessSystem='sql')
if fromAka:
ser_note = AkaTitle.get(episodeOfID).note
if ser_note:
mdict['episode of'].notes = ser_note
return mdict
def end_li(self):
self._in_li = 0
if self._in_episodes:
et = self._cur_episode_title.strip()
minfo = self._misc_info.strip()
if et and self._episode_id:
eps_data = analyze_title(et, canonical=1)
eps_data['kind'] = u'episode'
e = Movie(movieID=str(self._episode_id),
data=eps_data,
accessSystem=self._as,
modFunct=self._modFunct)
e['episode of'] = self._cur_series
if minfo.startswith('('):
pe = minfo.find(')')
if pe != -1:
date = minfo[1:pe]
if date != '????':
e['original air date'] = date
if eps_data.get('year', '????') == '????':
syear = date.split()[-1]
if syear.isdigit():
e['year'] = syear
rolei = minfo.find(' - ')
if rolei != -1:
if not self._got_i_info:
role = u''
role = minfo[rolei + 3:].strip()
notei = role.rfind('(')
note = u''
if notei != -1 and role and role[-1] == ')':
note = role[notei:]
role = role[:notei].strip()
e.notes = note
e.currentRole = role
else:
randn = minfo[rolei + 3:].strip().split()
note = '[%s]' % randn[0]
note += ' '.join(randn[1:])
e.notes = note
self._episodes.setdefault(self._cur_series, []).append(e)
self._cur_episode_title = u''
self._episode_id = None
self._in_misc_info = 0
self._misc_info = u''
def end_li(self):
self._in_li = 0
if self._in_episodes:
et = self._cur_episode_title.strip()
minfo = self._misc_info.strip()
if et and self._episode_id:
eps_data = analyze_title(et, canonical=1)
eps_data['kind'] = u'episode'
e = Movie(movieID=str(self._episode_id), data=eps_data,
accessSystem=self._as, modFunct=self._modFunct)
e['episode of'] = self._cur_series
if minfo.startswith('('):
pe = minfo.find(')')
if pe != -1:
date = minfo[1:pe]
if date != '????':
e['original air date'] = date
if eps_data.get('year', '????') == '????':
syear = date.split()[-1]
if syear.isdigit():
e['year'] = syear
rolei = minfo.find(' - ')
if rolei != -1:
if not self._got_i_info:
role = u''
role = minfo[rolei+3:].strip()
notei = role.rfind('(')
note = u''
if notei != -1 and role and role[-1] == ')':
note = role[notei:]
role = role[:notei].strip()
e.notes = note
e.currentRole = role
else:
randn = minfo[rolei+3:].strip().split()
note = '[%s]' % randn[0]
note += ' '.join(randn[1:])
e.notes = note
self._episodes.setdefault(self._cur_series, []).append(e)
self._cur_episode_title = u''
self._episode_id = None
self._in_misc_info = 0
self._misc_info = u''
def postprocess_data(self, data):
if not data:
return {}
newData = {}
for title in data:
movieID, quotes = data[title]
if movieID is None:
movie = title
else:
movie = Movie(title=title, movieID=movieID,
accessSystem=self._as, modFunct=self._modFunct)
newData[movie] = [quote.split('::') for quote in quotes]
return {'quotes': newData}
def _add_items(self):
self._quotes = [x.replace(':: ', '::').replace(' ::', '::').rstrip(':')
for x in self._quotes]
self._quotes = [x.replace(' ', ' ').replace(' ', ' ').strip()
for x in self._quotes]
self._quotes = filter(None, self._quotes)
if not (self._cur_title and self._cur_titleID and self._quotes):
self._quotes = [u'']
return
movie = Movie(title=self._cur_title, movieID=self._cur_titleID,
accessSystem=self._as, modFunct=self._modFunct)
self._tot_quotes[movie] = self._quotes[:]
self._quotes = [u'']
def test_search(self):
ia = IMDb()
serials = ia.search_movie("good place")
for serial in serials:
if serial.data["kind"] == "tv series":
print(serial.movieID, serial.data["title"], serial.data["kind"])
print(serials)
result = ia.get_movie_episodes("4955642")
print(result)
for season in result["data"]["episodes"]:
for index in result["data"]["episodes"][season]:
item = Movie(result["data"]["episodes"][4][index])
print(f"{season}x{index}", item.myID)
def getFilmography(dataF, indexF, keyF, attrIF, attrKF, offset,
charNF=None, doCast=0, doWriters=0):
"""Gather information from the given files about the
person entry found at offset; return a list of Movie objects,
with the relevant attributes."""
name, res = getRawData(dataF, offset, doCast, doWriters)
resList = []
for movie in res:
title = getLabel(movie['movieID'], indexF, keyF)
if not title: continue
curRole = movie.get('currentRole', u'')
roleID = None
if curRole and charNF:
curRole, roleID = getCharactersIDs(curRole, charNF)
m = Movie(title=title, movieID=movie['movieID'],
currentRole=curRole, roleID=roleID,
accessSystem='local')
if movie.has_key('attributeID'):
attr = getLabel(movie['attributeID'], attrIF, attrKF)
if attr: m.notes = attr
resList.append(m)
return resList
def end_a(self):
if self._in_episode_title:
self._in_episode_title = 0
self._in_misc_info = 1
elif self._in_series_title:
self._in_series_title = 0
st = self._cur_series_title.strip()
if st and self._series_id is not None:
series_data = analyze_title(st, canonical=1)
s = Movie(movieID=str(self._series_id),
data=series_data,
accessSystem=self._as,
modFunct=self._modFunct)
self._cur_series = s
def getMovieLinks(movieID, dataF, movieTitlIF, movieTitlKF):
"""Return a dictionary with movie connections."""
entries = getFullIndex(dataF, movieID, kind='mlinks',
rindex=None, multi=1, default=[])
res = {}
for entry in entries:
title = getLabel(entry[2], movieTitlIF, movieTitlKF)
if not title: continue
m = Movie(title=title, movieID=entry[2],
accessSystem='local')
sect = _links_sect.get(entry[1])
if not sect: continue
res.setdefault(sect, []).append(m)
return res
def _add_ref(self, kind):
"""Add a reference entry to the names and titles dictionaries."""
if kind == 'tt':
if self._titleRefCID and self._titleCN:
if not self._titlesRefs.has_key(self._titleCN):
try:
movie = Movie(movieID=str(self._titleRefCID),
title=self._titleCN,
accessSystem=self._as,
modFunct=self._modFunct)
self._titlesRefs[self._titleCN] = movie
except IMDbParserError:
pass
self._titleRefCID = u''
self._titleCN = u''
self._inTTRef = 0
self._inLinkTTRef = 0
elif kind == 'nm' and self._nameRefCID and self._nameCN:
# XXX: 'Neo' and 'Keanu Reeves' are two separated
# entry in the dictionary. Check the ID value instead
# of the key?
if not self._namesRefs.has_key(self._nameCN):
try:
person = Person(name=self._nameCN,
personID=str(self._nameRefCID),
accessSystem=self._as,
modFunct=self._modFunct)
self._namesRefs[self._nameCN] = person
except IMDbParserError:
pass
self._nameRefCID = u''
self._nameCN = u''
self._inNMRef = 0
elif kind == 'ch' and self._characterRefCID and self._characterCN:
if not self._charactersRefs.has_key(self._characterCN):
try:
character = Character(name=self._characterCN,
characterID=str(
self._characterRefCID),
accessSystem='http')
self._charactersRefs[self._characterCN] = character
except IMDbParserError:
pass
self._characterRefCID = u''
self._characterCN = u''
self._inCHRef = 0
def _findRefs(self, o, trefs, nrefs):
"""Find titles or names references in strings."""
if isinstance(o, (unicode, str)):
for title in re_titleRef.findall(o):
a_title = analyze_title(title, canonical=0)
rtitle = build_title(a_title, ptdf=1)
if trefs.has_key(rtitle): continue
movieID = self._getTitleID(rtitle)
if movieID is None:
movieID = self._getTitleID(title)
if movieID is None:
continue
m = Movie(title=rtitle,
movieID=movieID,
accessSystem=self.accessSystem)
trefs[rtitle] = m
rtitle2 = canonicalTitle(a_title.get('title', u''))
if rtitle2 and rtitle2 != rtitle and rtitle2 != title:
trefs[rtitle2] = m
if title != rtitle:
trefs[title] = m
for name in re_nameRef.findall(o):
a_name = analyze_name(name, canonical=1)
rname = build_name(a_name, canonical=1)
if nrefs.has_key(rname): continue
personID = self._getNameID(rname)
if personID is None:
personID = self._getNameID(name)
if personID is None: continue
p = Person(name=rname,
personID=personID,
accessSystem=self.accessSystem)
nrefs[rname] = p
rname2 = normalizeName(a_name.get('name', u''))
if rname2 and rname2 != rname:
nrefs[rname2] = p
if name != rname and name != rname2:
nrefs[name] = p
elif isinstance(o, (list, tuple)):
for item in o:
self._findRefs(item, trefs, nrefs)
elif isinstance(o, dict):
for value in o.values():
self._findRefs(value, trefs, nrefs)
return (trefs, nrefs)
def postprocess_data(self, data):
if len(data) == 0:
return {}
nd = {}
for key in data.keys():
dom = self.get_dom(key)
link = self.xpath(dom, "//a/@href")[0]
title = self.xpath(dom, "//a/text()")[0][1:-1]
series = Movie(movieID=analyze_imdbid(link),
data=analyze_title(title),
accessSystem=self._as, modFunct=self._modFunct)
nd[series] = []
for episode in data[key]:
# XXX: should we create a copy of 'series', to avoid
# circular references?
episode['episode of'] = series
nd[series].append(episode)
return {'episodes': nd}
def postprocess_data(self, data):
result = {}
for item in ('names refs', 'titles refs'):
result[item] = {}
for k, v in data.get(item, []):
k = k.strip()
v = v.strip()
if not (k and v):
continue
imdbID = analyze_imdbid(v)
if item == 'names refs':
obj = Person(personID=imdbID, name=k,
accessSystem=self._as, modFunct=self._modFunct)
elif item == 'titles refs':
obj = Movie(movieID=imdbID, title=k,
accessSystem=self._as, modFunct=self._modFunct)
result[item][k] = obj
return result
def __get_serial(imdb_id: str, title: str, year: str) -> Serial:
try:
ia = IMDb()
info = ia.get_movie_episodes(imdb_id)
seasons = list()
for season_index in info["data"]["episodes"]:
episodes = list()
for episode_index in info["data"]["episodes"][season_index]:
movie = Movie(info["data"]["episodes"][season_index][episode_index])
episode = Episode(episode_index, movie.myID.data["title"], movie.myID.movieID)
episodes.append(episode)
# print(f"{season_index}x{episode_index}", movie.myID)
season = Season(f"Season {season_index}", episodes)
seasons.append(season)
seasons.sort(key=lambda s: s.title)
serial = Serial(0, title, year, imdb_id, "", seasons)
return serial
except Exception as ex:
print("Exception", ex)
def _build_episode(link, title, minfo, role, roleA, roleAID):
"""Build an Movie object for a given episode of a series."""
episode_id = analyze_imdbid(link)
notes = ''
minidx = minfo.find(' -')
# Sometimes, for some unknown reason, the role is left in minfo.
if minidx != -1:
slfRole = minfo[minidx + 3:].lstrip()
minfo = minfo[:minidx].rstrip()
if slfRole.endswith(')'):
commidx = slfRole.rfind('(')
if commidx != -1:
notes = slfRole[commidx:]
slfRole = slfRole[:commidx]
if slfRole and role is None and roleA is None:
role = slfRole
eps_data = analyze_title(title)
eps_data['kind'] = 'episode'
# FIXME: it's wrong for multiple characters (very rare on tv series?).
if role is None:
role = roleA # At worse, it's None.
if role is None:
roleAID = None
if roleAID is not None:
roleAID = analyze_imdbid(roleAID)
e = Movie(movieID=episode_id,
data=eps_data,
currentRole=role,
roleID=roleAID,
notes=notes)
# XXX: are we missing some notes?
# XXX: does it parse things as "Episode dated 12 May 2005 (12 May 2005)"?
if minfo.startswith('('):
pe = minfo.find(')')
if pe != -1:
date = minfo[1:pe]
if date != '????':
e['original air date'] = date
if eps_data.get('year', '????') == '????':
syear = date.split()[-1]
if syear.isdigit():
e['year'] = int(syear)
return e
def do_br(self, attrs):
if self._in_series_title:
self._in_series_title = 0
st = self._series_title.strip()
if st and self.__seriesID:
d_title = analyze_title(st, canonical=1)
m = Movie(movieID=str(self.__seriesID),
data=d_title,
accessSystem=self._as,
modFunct=self._modFunct)
self._result['kind'] = u'episode'
self._result['episode of'] = m
self._series_title = u''
elif self._in_series_info:
self._in_series_info = 0
si = ' '.join([x for x in self._series_info.split() if x])
if si:
aid = self.re_airdate.findall(si)
if aid and len(aid[0]) == 3:
date, season, episode = aid[0]
date = date.strip()
try:
season = int(season)
except:
pass
try:
episode = int(episode)
except:
pass
if date and date != '????':
self._result['original air date'] = date
# Handle also "episode 0".
if season or type(season) is type(0):
self._result['season'] = season
if episode or type(season) is type(0):
self._result['episode'] = episode
self._series_info = u''
return None
length = convBin(dfptr.read(2), 'longlength')
# Skip character name.
latin2utf(dfptr.read(length))
nrItems = convBin(dfptr.read(3), 'nrCharacterItems')
if limit is not None and nrItems/2 > limit:
nrItems = limit*2
filmography = []
for i in xrange(nrItems/2):
personID = convBin(dfptr.read(3), 'personID')
name = getLabel(personID, personIF, personKF)
movieID = convBin(dfptr.read(3), 'movieID')
title = getLabel(movieID, movieIF, movieKF)
# XXX: notes are not retrieved: they can be found scanning
# actors.list and acresses.list, but it will slow down everything.
m = Movie(title=title, movieID=movieID, currentRole=name,
roleID=personID, roleIsPerson=True, accessSystem='local')
filmography.append(m)
dfptr.close()
return filmography
def _convChID(characterID):
"""Return a numeric value for the given string, or None."""
if characterID is None:
return None
return convBin(characterID, 'characterID')
def getCharactersIDs(names_string, charNF):
"""Returns a tuple (name, roleID) if the supplied string contains
only one character, otherwise returns a tuple of lists:
# Complete cast/crew.
compcast = [
(self._compcast[cc.subjectID], self._compcast[cc.statusID])
for cc in CompleteCast.select(CompleteCast.q.movieID == movieID)
]
if compcast:
for entry in compcast:
val = unicode(entry[1])
res[u'complete %s' % entry[0]] = val
# Movie connections.
mlinks = [[ml.linkedMovieID, self._link[ml.linkTypeID]]
for ml in MovieLink.select(MovieLink.q.movieID == movieID)]
if mlinks:
for ml in mlinks:
lmovieData = get_movie_data(ml[0], self._kind)
m = Movie(movieID=ml[0], data=lmovieData, accessSystem='sql')
ml[0] = m
res['connections'] = {}
mlinks[:] = _groupListBy(mlinks, 1)
for group in mlinks:
lt = group[0][1]
res['connections'][lt] = [i[0] for i in group]
# Episodes.
episodes = {}
eps_list = list(Title.select(Title.q.episodeOfID == movieID))
eps_list.sort()
if eps_list:
ps_data = {
'title': res['title'],
'kind': res['kind'],
'year': res.get('year'),
def get_movie_main(self, movieID):
# Information sets provided by this method.
infosets = ('main', 'vote details')
tl = getLabel(movieID, '%stitles.index' % self.__db,
'%stitles.key' % self.__db)
# No title, no party.
if tl is None:
raise IMDbDataAccessError, 'unable to get movieID "%s"' % movieID
res = analyze_title(tl)
# Build the cast list.
actl = []
for castG in ('actors', 'actresses'):
midx = getFullIndex('%s%s.titles' % (self.__db, castG),
movieID, multi=1)
if midx is not None:
params = {'movieID': movieID,
'dataF': '%s%s.data' % (self.__db, castG),
'indexF': '%snames.index' % self.__db,
'keyF': '%snames.key' % self.__db,
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db,
'charNF': '%scharacter2id.index' % self.__db,
'offsList': midx, 'doCast': 1}
actl += getMovieCast(**params)
if actl:
actl.sort()
res['cast'] = actl
# List of other workers.
works = ('writer', 'cinematographer', 'composer',
'costume-designer', 'director', 'editor', 'miscellaneou',
'producer', 'production-designer', 'cinematographer')
for i in works:
index = getFullIndex('%s%ss.titles' % (self.__db, i),
movieID, multi=1)
if index is not None:
params = {'movieID': movieID,
'dataF': '%s%s.data' % (self.__db, i),
'indexF': '%snames.index' % self.__db,
'keyF': '%snames.key' % self.__db,
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db,
'offsList': index}
name = key = i
if '-' in name:
name = name.replace('-', ' ')
elif name == 'miscellaneou':
name = 'miscellaneous crew'
key = 'miscellaneou'
elif name == 'writer':
params['doWriters'] = 1
params['dataF'] = '%s%ss.data' % (self.__db, key)
data = getMovieCast(**params)
if name == 'writer': data.sort()
res[name] = data
# Rating.
rt = self.get_movie_vote_details(movieID)['data']
if rt: res.update(rt)
# Various information.
miscInfo = (('runtimes', 'running-times'), ('color info', 'color-info'),
('genres', 'genres'), ('distributors', 'distributors'),
('languages', 'language'), ('certificates', 'certificates'),
('special effects companies', 'special-effects-companies'),
('sound mix', 'sound-mix'), ('tech info', 'technical'),
('production companies', 'production-companies'),
('countries', 'countries'))
for name, fname in miscInfo:
params = {'movieID': movieID,
'dataF': '%s%s.data' % (self.__db, fname),
'indexF': '%s%s.index' % (self.__db, fname),
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db}
data = getMovieMisc(**params)
if name in ('distributors', 'special effects companies',
'production companies'):
for nitem in xrange(len(data)):
n, notes = split_company_name_notes(data[nitem])
company = Company(name=n, companyID=getCompanyID(n,
'%scompany2id.index' % self.__db),
notes=notes,
accessSystem=self.accessSystem)
data[nitem] = company
if data: res[name] = data
if res.has_key('runtimes') and len(res['runtimes']) > 0:
rt = res['runtimes'][0]
episodes = re_episodes.findall(rt)
if episodes:
res['runtimes'][0] = re_episodes.sub('', rt)
res['number of episodes'] = episodes[0]
# AKA titles.
akas = getAkaTitles(movieID,
'%saka-titles.data' % self.__db,
'%stitles.index' % self.__db,
'%stitles.key' % self.__db,
'%sattributes.index' % self.__db,
'%sattributes.key' % self.__db)
if akas:
# normalize encoding.
for i in xrange(len(akas)):
ts = akas[i].split('::')
if len(ts) != 2: continue
t = ts[0]
n = ts[1]
nt = self._changeAKAencoding(n, t)
if nt is not None: akas[i] = '%s::%s' % (nt, n)
res['akas'] = akas
if res.get('kind') == 'episode':
# Things to do if this is a tv series episode.
episodeOf = res.get('episode of')
if episodeOf is not None:
parentSeries = Movie(data=res['episode of'],
accessSystem='local')
seriesID = self._getTitleID(parentSeries.get(
'long imdb canonical title'))
parentSeries.movieID = seriesID
res['episode of'] = parentSeries
if not res.get('year'):
year = getFullIndex('%smovies.data' % self.__db,
movieID, kind='moviedata', rindex=1)
if year: res['year'] = year
# MPAA info.
mpaa = getMPAA(movieID, '%smpaa-ratings-reasons.index' % self.__db,
'%smpaa-ratings-reasons.data' % self.__db)
if mpaa: res.update(mpaa)
return {'data': res, 'info sets': infosets}
def build_movie(txt,
movieID=None,
roleID=None,
status=None,
accessSystem='http',
modFunct=None,
_parsingCharacter=False,
_parsingCompany=False,
year=None,
chrRoles=None,
rolesNoChar=None,
additionalNotes=None):
"""Given a string as normally seen on the "categorized" page of
a person on the IMDb's web site, returns a Movie instance."""
# FIXME: Oook, lets face it: build_movie and build_person are now
# two horrible sets of patches to support the new IMDb design. They
# must be rewritten from scratch.
if _parsingCharacter:
_defSep = ' Played by '
elif _parsingCompany:
_defSep = ' ... '
else:
_defSep = ' .... '
title = re_spaces.sub(' ', txt).strip()
# Split the role/notes from the movie title.
tsplit = title.split(_defSep, 1)
role = u''
notes = u''
roleNotes = []
if len(tsplit) == 2:
title = tsplit[0].rstrip()
role = tsplit[1].lstrip()
if title[-9:] == 'TV Series':
title = title[:-9].rstrip()
elif title[-14:] == 'TV mini-series':
title = title[:-14] + ' (mini)'
if title and title.endswith(_defSep.rstrip()):
title = title[:-len(_defSep) + 1]
# Try to understand where the movie title ends.
while True:
if year:
break
if title[-1:] != ')':
# Ignore the silly "TV Series" notice.
if title[-9:] == 'TV Series':
title = title[:-9].rstrip()
continue
else:
# Just a title: stop here.
break
# Try to match paired parentheses; yes: sometimes there are
# parentheses inside comments...
nidx = title.rfind('(')
while (nidx != -1 and \
title[nidx:].count('(') != title[nidx:].count(')')):
nidx = title[:nidx].rfind('(')
# Unbalanced parentheses: stop here.
if nidx == -1: break
# The last item in parentheses seems to be a year: stop here.
first4 = title[nidx + 1:nidx + 5]
if (first4.isdigit() or first4 == '????') and \
title[nidx+5:nidx+6] in (')', '/'):
break
# The last item in parentheses is a known kind: stop here.
if title[nidx + 1:-1] in ('TV', 'V', 'mini', 'VG'): break
# Else, in parentheses there are some notes.
# XXX: should the notes in the role half be kept separated
# from the notes in the movie title half?
if notes: notes = '%s %s' % (title[nidx:], notes)
else: notes = title[nidx:]
title = title[:nidx].rstrip()
if year:
year = year.strip()
if title[-1] == ')':
fpIdx = title.rfind('(')
if fpIdx != -1:
if notes: notes = '%s %s' % (title[fpIdx:], notes)
else: notes = title[fpIdx:]
title = title[:fpIdx].rstrip()
title = u'%s (%s)' % (title, year)
if _parsingCharacter and roleID and not role:
roleID = None
if not roleID:
roleID = None
elif len(roleID) == 1:
roleID = roleID[0]
if not role and chrRoles and isinstance(roleID, (str, unicode)):
roleID = _re_chrIDs.findall(roleID)
role = ' / '.join(filter(None, chrRoles.split('@@')))
# Manages multiple roleIDs.
if isinstance(roleID, list):
tmprole = role.split('/')
role = []
for r in tmprole:
nidx = r.find('(')
if nidx != -1:
role.append(r[:nidx].rstrip())
roleNotes.append(r[nidx:])
else:
role.append(r)
roleNotes.append(None)
lr = len(role)
lrid = len(roleID)
if lr > lrid:
roleID += [None] * (lrid - lr)
elif lr < lrid:
roleID = roleID[:lr]
for i, rid in enumerate(roleID):
if rid is not None:
roleID[i] = str(rid)
if lr == 1:
role = role[0]
roleID = roleID[0]
elif roleID is not None:
roleID = str(roleID)
if movieID is not None:
movieID = str(movieID)
if (not title) or (movieID is None):
_b_m_logger.error('empty title or movieID for "%s"', txt)
if rolesNoChar:
rolesNoChar = filter(None, [x.strip() for x in rolesNoChar.split('/')])
if not role:
role = []
elif not isinstance(role, list):
role = [role]
role += rolesNoChar
notes = notes.strip()
if additionalNotes:
additionalNotes = re_spaces.sub(' ', additionalNotes).strip()
if notes:
notes += u' '
notes += additionalNotes
m = Movie(title=title,
movieID=movieID,
notes=notes,
currentRole=role,
roleID=roleID,
roleIsPerson=_parsingCharacter,
modFunct=modFunct,
accessSystem=accessSystem)
if roleNotes and len(roleNotes) == len(roleID):
for idx, role in enumerate(m.currentRole):
try:
if roleNotes[idx]:
role.notes = roleNotes[idx]
except IndexError:
break
# Status can't be checked here, and must be detected by the parser.
if status:
m['status'] = status
return m
def get_person_filmography(self, personID):
infosets = ('filmography', 'episodes')
res = {}
episodes = {}
works = ('actor', 'actresse', 'producer', 'writer',
'cinematographer', 'composer', 'costume-designer',
'director', 'editor', 'miscellaneou', 'production-designer')
for i in works:
index = getFullIndex('%s%ss.names' % (self.__db, i), personID)
if index is not None:
params = {'offset': index,
'indexF': '%stitles.index' % self.__db,
'keyF': '%stitles.key' % self.__db,
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db,
'charNF': '%scharacter2id.index' % self.__db}
name = key = i
if '-' in name:
name = name.replace('-', ' ')
elif name == 'actresse':
name = 'actress'
params['doCast'] = 1
elif name == 'miscellaneou':
name = 'miscellaneous crew'
key = 'miscellaneou'
elif name == 'actor':
params['doCast'] = 1
elif name == 'writer':
params['doWriters'] = 1
params['dataF'] = '%s%ss.data' % (self.__db, key)
data = getFilmography(**params)
movies = []
eps = []
# Split normal titles from episodes.
for d in data:
if d.get('kind') != 'episode':
movies.append(d)
else:
eps.append(d)
movies.sort()
if movies:
res[name] = movies
for e in eps:
series = Movie(data=e['episode of'], accessSystem='local')
seriesID = self._getTitleID(series.get(
'long imdb canonical title'))
series.movieID = seriesID
if not e.get('year'):
year = getFullIndex('%smovies.data' % self.__db,
e.movieID, kind='moviedata',
rindex=1)
if year: e['year'] = year
if not e.currentRole and name not in ('actor', 'actress'):
if e.notes: e.notes = ' %s' % e.notes
e.notes = '[%s]%s' % (name, e.notes)
episodes.setdefault(series, []).append(e)
if episodes:
for k in episodes:
episodes[k].sort()
episodes[k].reverse()
res['episodes'] = episodes
return {'data': res, 'info sets': tuple(infosets)}
def build_movie(txt,
movieID=None,
roleID=None,
status=None,
accessSystem='http',
modFunct=None,
_parsingCharacter=False,
_parsingCompany=False):
"""Given a string as normally seen on the "categorized" page of
a person on the IMDb's web site, returns a Movie instance."""
if _parsingCharacter:
_defSep = ' Played by '
elif _parsingCompany:
_defSep = ' ... '
else:
_defSep = ' .... '
title = re_spaces.sub(' ', txt).strip()
# Split the role/notes from the movie title.
tsplit = title.split(_defSep, 1)
role = u''
notes = u''
roleNotes = []
if len(tsplit) == 2:
title = tsplit[0].rstrip()
role = tsplit[1].lstrip()
if title[-9:] == 'TV Series':
title = title[:-9].rstrip()
elif title[-14:] == 'TV mini-series':
title = title[:-14] + ' (mini)'
# Try to understand where the movie title ends.
while True:
if title[-1:] != ')':
# Ignore the silly "TV Series" notice.
if title[-9:] == 'TV Series':
title = title[:-9].rstrip()
continue
else:
# Just a title: stop here.
break
# Try to match paired parentheses; yes: sometimes there are
# parentheses inside comments...
nidx = title.rfind('(')
while (nidx != -1 and \
title[nidx:].count('(') != title[nidx:].count(')')):
nidx = title[:nidx].rfind('(')
# Unbalanced parentheses: stop here.
if nidx == -1: break
# The last item in parentheses seems to be a year: stop here.
first4 = title[nidx + 1:nidx + 5]
if (first4.isdigit() or first4 == '????') and \
title[nidx+5:nidx+6] in (')', '/'):
break
# The last item in parentheses is a known kind: stop here.
if title[nidx + 1:-1] in ('TV', 'V', 'mini', 'VG'): break
# Else, in parentheses there are some notes.
# XXX: should the notes in the role half be kept separated
# from the notes in the movie title half?
if notes: notes = '%s %s' % (title[nidx:], notes)
else: notes = title[nidx:]
title = title[:nidx].rstrip()
if _parsingCharacter and roleID and not role:
roleID = None
if not roleID:
roleID = None
elif len(roleID) == 1:
roleID = roleID[0]
# Manages multiple roleIDs.
if isinstance(roleID, list):
tmprole = role.split('/')
role = []
for r in tmprole:
nidx = r.find('(')
if nidx != -1:
role.append(r[:nidx].rstrip())
roleNotes.append(r[nidx:])
else:
role.append(r)
roleNotes.append(None)
lr = len(role)
lrid = len(roleID)
if lr > lrid:
roleID += [None] * (lrid - lr)
elif lr < lrid:
roleID = roleID[:lr]
for i, rid in enumerate(roleID):
if rid is not None:
roleID[i] = str(rid)
if lr == 1:
role = role[0]
roleID = roleID[0]
elif roleID is not None:
roleID = str(roleID)
if movieID is not None:
movieID = str(movieID)
if (not title) or (movieID is None):
_b_m_logger.error('empty title or movieID for "%s"', txt)
m = Movie(title=title,
movieID=movieID,
notes=notes,
currentRole=role,
roleID=roleID,
roleIsPerson=_parsingCharacter,
modFunct=modFunct,
accessSystem=accessSystem)
if roleNotes and len(roleNotes) == len(roleID):
for idx, role in enumerate(m.currentRole):
if roleNotes[idx]:
role.notes = roleNotes[idx]
# Status can't be checked here, and must be detected by the parser.
if status:
m['status'] = status
return m
def get_movie_main(self, movieID):
# Information sets provided by this method.
infosets = ('main', 'vote details')
tl = getLabel(movieID, '%stitles.index' % self.__db,
'%stitles.key' % self.__db)
# No title, no party.
if tl is None:
raise IMDbDataAccessError, 'unable to get movieID "%s"' % movieID
res = analyze_title(tl)
# Build the cast list.
actl = []
for castG in ('actors', 'actresses'):
midx = getFullIndex('%s%s.titles' % (self.__db, castG),
movieID, multi=1)
if midx is not None:
params = {'movieID': movieID,
'dataF': '%s%s.data' % (self.__db, castG),
'indexF': '%snames.index' % self.__db,
'keyF': '%snames.key' % self.__db,
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db,
'charNF': '%scharacter2id.index' % self.__db,
'offsList': midx, 'doCast': 1}
actl += getMovieCast(**params)
if actl:
actl.sort()
res['cast'] = actl
# List of other workers.
works = ('writer', 'cinematographer', 'composer',
'costume-designer', 'director', 'editor', 'miscellaneou',
'producer', 'production-designer', 'cinematographer')
for i in works:
index = getFullIndex('%s%ss.titles' % (self.__db, i),
movieID, multi=1)
if index is not None:
params = {'movieID': movieID,
'dataF': '%s%s.data' % (self.__db, i),
'indexF': '%snames.index' % self.__db,
'keyF': '%snames.key' % self.__db,
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db,
'offsList': index}
name = key = i
if '-' in name:
name = name.replace('-', ' ')
elif name == 'miscellaneou':
name = 'miscellaneous crew'
key = 'miscellaneou'
elif name == 'writer':
params['doWriters'] = 1
params['dataF'] = '%s%ss.data' % (self.__db, key)
data = getMovieCast(**params)
if name == 'writer': data.sort()
res[name] = data
# Rating.
rt = self.get_movie_vote_details(movieID)['data']
if rt: res.update(rt)
# Various information.
miscInfo = (('runtimes', 'running-times'), ('color info', 'color-info'),
('genres', 'genres'), ('distributors', 'distributors'),
('languages', 'language'), ('certificates', 'certificates'),
('special effects companies', 'special-effects-companies'),
('sound mix', 'sound-mix'), ('tech info', 'technical'),
('production companies', 'production-companies'),
('countries', 'countries'))
for name, fname in miscInfo:
params = {'movieID': movieID,
'dataF': '%s%s.data' % (self.__db, fname),
'indexF': '%s%s.index' % (self.__db, fname),
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db}
data = getMovieMisc(**params)
if name in ('distributors', 'special effects companies',
'production companies'):
for nitem in xrange(len(data)):
n, notes = split_company_name_notes(data[nitem])
company = Company(name=n, companyID=getCompanyID(n,
'%scompany2id.index' % self.__db),
notes=notes,
accessSystem=self.accessSystem)
data[nitem] = company
if data: res[name] = data
if res.has_key('runtimes') and len(res['runtimes']) > 0:
rt = res['runtimes'][0]
episodes = re_episodes.findall(rt)
if episodes:
res['runtimes'][0] = re_episodes.sub('', rt)
res['number of episodes'] = episodes[0]
# AKA titles.
akas = getAkaTitles(movieID,
'%saka-titles.data' % self.__db,
'%stitles.index' % self.__db,
'%stitles.key' % self.__db,
'%sattributes.index' % self.__db,
'%sattributes.key' % self.__db)
if akas:
# normalize encoding.
for i in xrange(len(akas)):
ts = akas[i].split('::')
if len(ts) != 2: continue
t = ts[0]
n = ts[1]
nt = self._changeAKAencoding(n, t)
if nt is not None: akas[i] = '%s::%s' % (nt, n)
res['akas'] = akas
if res.get('kind') == 'episode':
# Things to do if this is a tv series episode.
episodeOf = res.get('episode of')
if episodeOf is not None:
parentSeries = Movie(data=res['episode of'],
accessSystem='local')
seriesID = self._getTitleID(parentSeries.get(
'long imdb canonical title'))
parentSeries.movieID = seriesID
res['episode of'] = parentSeries
if not res.get('year'):
year = getFullIndex('%smovies.data' % self.__db,
movieID, kind='moviedata', rindex=1)
if year: res['year'] = year
# MPAA info.
mpaa = getMPAA(movieID, '%smpaa-ratings-reasons.index' % self.__db,
'%smpaa-ratings-reasons.data' % self.__db)
if mpaa: res.update(mpaa)
return {'data': res, 'info sets': infosets}
def get_person_filmography(self, personID):
infosets = ('filmography', 'episodes')
res = {}
episodes = {}
works = ('actor', 'actresse', 'producer', 'writer',
'cinematographer', 'composer', 'costume-designer',
'director', 'editor', 'miscellaneou', 'production-designer')
for i in works:
index = getFullIndex('%s%ss.names' % (self.__db, i), personID)
if index is not None:
params = {'offset': index,
'indexF': '%stitles.index' % self.__db,
'keyF': '%stitles.key' % self.__db,
'attrIF': '%sattributes.index' % self.__db,
'attrKF': '%sattributes.key' % self.__db,
'charNF': '%scharacter2id.index' % self.__db}
name = key = i
if '-' in name:
name = name.replace('-', ' ')
elif name == 'actresse':
name = 'actress'
params['doCast'] = 1
elif name == 'miscellaneou':
name = 'miscellaneous crew'
key = 'miscellaneou'
elif name == 'actor':
params['doCast'] = 1
elif name == 'writer':
params['doWriters'] = 1
params['dataF'] = '%s%ss.data' % (self.__db, key)
data = getFilmography(**params)
movies = []
eps = []
# Split normal titles from episodes.
for d in data:
if d.get('kind') != 'episode':
movies.append(d)
else:
eps.append(d)
movies.sort()
if movies:
res[name] = movies
for e in eps:
series = Movie(data=e['episode of'], accessSystem='local')
seriesID = self._getTitleID(series.get(
'long imdb canonical title'))
series.movieID = seriesID
if not e.get('year'):
year = getFullIndex('%smovies.data' % self.__db,
e.movieID, kind='moviedata',
rindex=1)
if year: e['year'] = year
if not e.currentRole and name not in ('actor', 'actress'):
if e.notes: e.notes = ' %s' % e.notes
e.notes = '[%s]%s' % (name, e.notes)
episodes.setdefault(series, []).append(e)
if episodes:
for k in episodes:
episodes[k].sort()
episodes[k].reverse()
res['episodes'] = episodes
return {'data': res, 'info sets': tuple(infosets)}
# Yes: kindID values are hard-coded in the companies4local.py script.
_kinds = {
0: 'distributors',
1: 'production companies',
2: 'special effect companies',
3: 'miscellaneous companies'
}
for i in xrange(nrItems):
kind = _kinds.get(ord(dfptr.read(1)))
if kind is None:
import warnings
warnings.warn('Unidentified kindID for a company.')
break
movieID = convBin(dfptr.read(3), 'movieID')
title = getLabel(movieID, movieIF, movieKF)
m = Movie(title=title, movieID=movieID, accessSystem='local')
filmography.setdefault(kind, []).append(m)
dfptr.close()
return filmography
def _convChID(companyID):
"""Return a numeric value for the given string, or None."""
if companyID is None:
return None
return convBin(companyID, 'companyID')
def getCompanyID(name, compNF):
"""Return a companyID for a name."""
try:
def get_movie_main(self, movieID):
cont = self._mretrieve(self.urls['movie_main'] % movieID + 'maindetails')
title = _findBetween(cont, '<title>', '</title>', maxRes=1)
if not title:
raise IMDbDataAccessError('unable to get movieID "%s"' % movieID)
title = _unHtml(title[0])
if title.endswith(' - IMDb'):
title = title[:-7]
if cont.find('<span class="tv-extra">TV mini-series</span>') != -1:
title += ' (mini)'
d = analyze_title(title)
kind = d.get('kind')
tv_series = _findBetween(cont, 'TV Series:</h5>', '</a>', maxRes=1)
if tv_series: mid = re_imdbID.findall(tv_series[0])
else: mid = None
if tv_series and mid:
s_title = _unHtml(tv_series[0])
s_data = analyze_title(s_title)
m = Movie(movieID=str(mid[0]), data=s_data,
accessSystem=self.accessSystem,
modFunct=self._defModFunct)
d['kind'] = kind = u'episode'
d['episode of'] = m
if kind in ('tv series', 'tv mini series'):
years = _findBetween(cont, '<h1>', '</h1>', maxRes=1)
if years:
years[:] = _findBetween(years[0], 'TV series', '</span>',
maxRes=1)
if years:
d['series years'] = years[0].strip()
air_date = _findBetween(cont, 'Original Air Date:</h5>', '</div>',
maxRes=1)
if air_date:
air_date = air_date[0]
vi = air_date.find('(')
if vi != -1:
date = _unHtml(air_date[:vi]).strip()
if date != '????':
d['original air date'] = date
air_date = air_date[vi:]
season = _findBetween(air_date, 'Season', ',', maxRes=1)
if season:
season = season[0].strip()
try: season = int(season)
except: pass
if season or type(season) is _inttype:
d['season'] = season
episode = _findBetween(air_date, 'Episode', ')', maxRes=1)
if episode:
episode = episode[0].strip()
try: episode = int(episode)
except: pass
if episode or type(season) is _inttype:
d['episode'] = episode
direct = _findBetween(cont, '<h5>Director', ('</div>', '<br/> <br/>'),
maxRes=1)
if direct:
direct = direct[0]
h5idx = direct.find('/h5>')
if h5idx != -1:
direct = direct[h5idx+4:]
direct = self._getPersons(direct)
if direct: d['director'] = direct
if kind in ('tv series', 'tv mini series', 'episode'):
if kind != 'episode':
seasons = _findBetween(cont, 'Seasons:</h5>', '</div>',
maxRes=1)
if seasons:
d['number of seasons'] = seasons[0].count('|') + 1
creator = _findBetween(cont, 'Created by</h5>', ('class="tn15more"',
'</div>',
'<br/> <br/>'),
maxRes=1)
if not creator:
# They change 'Created by' to 'Creator' and viceversa
# from time to time...
# XXX: is 'Creators' also used?
creator = _findBetween(cont, 'Creator:</h5>',
('class="tn15more"', '</div>',
'<br/> <br/>'), maxRes=1)
if creator:
creator = creator[0]
if creator.find('tn15more'): creator = '%s>' % creator
creator = self._getPersons(creator)
if creator: d['creator'] = creator
writers = _findBetween(cont, '<h5>Writer', ('</div>', '<br/> <br/>'),
maxRes=1)
if writers:
writers = writers[0]
h5idx = writers.find('/h5>')
if h5idx != -1:
writers = writers[h5idx+4:]
writers = self._getPersons(writers)
if writers: d['writer'] = writers
cvurl = _getTagsWith(cont, 'name="poster"', toClosure=True, maxRes=1)
if cvurl:
cvurl = _findBetween(cvurl[0], 'src="', '"', maxRes=1)
if cvurl: d['cover url'] = cvurl[0]
genres = _findBetween(cont, 'href="/genre/', '"')
if genres:
d['genres'] = list(set(genres))
ur = _findBetween(cont, 'id="star-bar-user-rate">', '</div>',
maxRes=1)
if ur:
rat = _findBetween(ur[0], '<b>', '</b>', maxRes=1)
if rat:
if rat:
d['rating'] = rat[0].strip()
else:
self._mobile_logger.warn('wrong rating: %s', rat)
vi = ur[0].rfind('href="ratings"')
if vi != -1 and ur[0][vi+10:].find('await') == -1:
try:
votes = _findBetween(ur[0][vi:], "title='",
" IMDb", maxRes=1)
votes = int(votes[0].replace(',', ''))
d['votes'] = votes
except (ValueError, IndexError):
self._mobile_logger.warn('wrong votes: %s', ur)
top250 = _findBetween(cont, 'href="/chart/top?', '</a>', maxRes=1)
if top250:
fn = top250[0].rfind('#')
if fn != -1:
try:
td = int(top250[0][fn+1:])
d['top 250 rank'] = td
except ValueError:
self._mobile_logger.warn('wrong top250: %s', top250)
castdata = _findBetween(cont, 'Cast overview', '</table>', maxRes=1)
if not castdata:
castdata = _findBetween(cont, 'Credited cast', '</table>', maxRes=1)
if not castdata:
castdata = _findBetween(cont, 'Complete credited cast', '</table>',
maxRes=1)
if not castdata:
castdata = _findBetween(cont, 'Series Cast Summary', '</table>',
maxRes=1)
if not castdata:
castdata = _findBetween(cont, 'Episode Credited cast', '</table>',
maxRes=1)
if castdata:
castdata = castdata[0]
# Reintegrate the fist tag.
fl = castdata.find('href=')
if fl != -1: castdata = '<a ' + castdata[fl:]
# Exclude the 'rest of cast listed alphabetically' row.
smib = castdata.find('<tr><td align="center" colspan="4"><small>')
if smib != -1:
smie = castdata.rfind('</small></td></tr>')
if smie != -1:
castdata = castdata[:smib].strip() + \
castdata[smie+18:].strip()
castdata = castdata.replace('/tr> <tr', '/tr><tr')
cast = self._getPersons(castdata, sep='</tr><tr')
if cast: d['cast'] = cast
akas = _findBetween(cont, 'Also Known As:</h5>', '</div>', maxRes=1)
if akas:
# For some reason, here <br> is still used in place of <br/>.
akas[:] = [x for x in akas[0].split('<br>') if x.strip()]
akas = [_unHtml(x).replace('" - ','::', 1).lstrip('"').strip()
for x in akas]
if 'See more' in akas: akas.remove('See more')
akas[:] = [x for x in akas if x]
if akas:
d['akas'] = akas
mpaa = _findBetween(cont, 'MPAA</a>:', '</div>', maxRes=1)
if mpaa: d['mpaa'] = _unHtml(mpaa[0])
runtimes = _findBetween(cont, 'Runtime:</h5>', '</div>', maxRes=1)
if runtimes:
runtimes = runtimes[0]
runtimes = [x.strip().replace(' min', '').replace(' (', '::(', 1)
for x in runtimes.split('|')]
d['runtimes'] = [_unHtml(x).strip() for x in runtimes]
if kind == 'episode':
# number of episodes.
epsn = _findBetween(cont, 'title="Full Episode List">', '</a>',
maxRes=1)
if epsn:
epsn = epsn[0].replace(' Episodes', '').strip()
if epsn:
try:
epsn = int(epsn)
except:
self._mobile_logger.warn('wrong episodes #: %s', epsn)
d['number of episodes'] = epsn
country = _findBetween(cont, 'Country:</h5>', '</div>', maxRes=1)
if country:
country[:] = country[0].split(' | ')
country[:] = ['<a %s' % x for x in country if x]
country[:] = [_unHtml(x.replace(' <i>', '::')) for x in country]
if country: d['countries'] = country
lang = _findBetween(cont, 'Language:</h5>', '</div>', maxRes=1)
if lang:
lang[:] = lang[0].split(' | ')
lang[:] = ['<a %s' % x for x in lang if x]
lang[:] = [_unHtml(x.replace(' <i>', '::')) for x in lang]
if lang: d['languages'] = lang
col = _findBetween(cont, '"/search/title?colors=', '</div>')
if col:
col[:] = col[0].split(' | ')
col[:] = ['<a %s' % x for x in col if x]
col[:] = [_unHtml(x.replace(' <i>', '::')) for x in col]
if col: d['color info'] = col
sm = _findBetween(cont, '/search/title?sound_mixes=', '</div>',
maxRes=1)
if sm:
sm[:] = sm[0].split(' | ')
sm[:] = ['<a %s' % x for x in sm if x]
sm[:] = [_unHtml(x.replace(' <i>', '::')) for x in sm]
if sm: d['sound mix'] = sm
cert = _findBetween(cont, 'Certification:</h5>', '</div>', maxRes=1)
if cert:
cert[:] = cert[0].split(' | ')
cert[:] = [_unHtml(x.replace(' <i>', '::')) for x in cert]
if cert: d['certificates'] = cert
plotoutline = _findBetween(cont, 'Plot:</h5>', ['<a ', '</div>'],
maxRes=1)
if plotoutline:
plotoutline = plotoutline[0].strip()
plotoutline = plotoutline.rstrip('|').rstrip()
if plotoutline: d['plot outline'] = _unHtml(plotoutline)
aratio = _findBetween(cont, 'Aspect Ratio:</h5>', ['<a ', '</div>'],
maxRes=1)
if aratio:
aratio = aratio[0].strip().replace(' (', '::(', 1)
if aratio:
d['aspect ratio'] = _unHtml(aratio)
return {'data': d}
