def get_person_main(self, personID):
infosets = ('main', 'biography', 'other works')
nl = getLabel(personID, '%snames.index' % self.__db,
'%snames.key' % self.__db)
# No name, no party.
if nl is None:
raise IMDbDataAccessError, 'unable to get personID "%s"' % personID
res = analyze_name(nl)
res.update(getBio(personID, '%sbiographies.index' % self.__db,
'%sbiographies.data' % self.__db))
akas = getAkaNames(personID,
'%saka-names.data' % self.__db,
'%snames.index' % self.__db,
'%snames.key' % self.__db)
if akas: res['akas'] = akas
# XXX: horrible hack! The getBio() function is not able to
# retrieve the movieID!
# A cleaner solution, would be to NOT return Movies object
# at first, from the getBio() function.
# XXX: anyway, this is no more needed, since "guest appearances"
# are gone with the new tv series episodes support.
if res.has_key('notable tv guest appearances'):
nl = []
for m in res['notable tv guest appearances']:
movieID = self._getTitleID(m.get('long imdb canonical title'))
if movieID is None: continue
m.movieID = movieID
nl.append(m)
if nl:
nl.sort()
res['notable tv guest appearances'][:] = nl
else: del res['notable tv guest appearances']
trefs, nrefs = self._extractRefs(res)
return {'data': res, 'info sets': infosets,
'titlesRefs': trefs, 'namesRefs': nrefs}
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 get_person_main(self, personID):
infosets = ('main', 'biography', 'other works')
nl = getLabel(personID, '%snames.index' % self.__db,
'%snames.key' % self.__db)
# No name, no party.
if nl is None:
raise IMDbDataAccessError, 'unable to get personID "%s"' % personID
res = analyze_name(nl)
res.update(getBio(personID, '%sbiographies.index' % self.__db,
'%sbiographies.data' % self.__db))
akas = getAkaNames(personID,
'%saka-names.data' % self.__db,
'%snames.index' % self.__db,
'%snames.key' % self.__db)
if akas: res['akas'] = akas
# XXX: horrible hack! The getBio() function is not able to
# retrieve the movieID!
# A cleaner solution, would be to NOT return Movies object
# at first, from the getBio() function.
# XXX: anyway, this is no more needed, since "guest appearances"
# are gone with the new tv series episodes support.
if res.has_key('notable tv guest appearances'):
nl = []
for m in res['notable tv guest appearances']:
movieID = self._getTitleID(m.get('long imdb canonical title'))
if movieID is None: continue
m.movieID = movieID
nl.append(m)
if nl:
nl.sort()
res['notable tv guest appearances'][:] = nl
else: del res['notable tv guest appearances']
trefs, nrefs = self._extractRefs(res)
return {'data': res, 'info sets': infosets,
'titlesRefs': trefs, 'namesRefs': nrefs}
def _search_movie(self, title, results, _episodes=False):
title = title.strip()
if not title: return []
# Search for these title variations.
if not _episodes:
title1, title2, title3 = titleVariations(title, fromPtdf=1)
else:
title1 = normalizeTitle(title)
title2 = ''
title3 = ''
# XXX: only a guess: results are shrinked, to exclude Adult
# titles and to remove duplicated entries.
resultsST = results * 3
res = _scan_titles('%stitles.key' % self.__db,
title1, title2, title3, resultsST, _episodes)
res[:] = [x[1] for x in res]
# Check for adult movies.
if not self.doAdult:
newlist = []
for entry in res:
genres = getMovieMisc(movieID=entry[0],
dataF='%s%s.data' % (self.__db, 'genres'),
indexF='%s%s.index' % (self.__db, 'genres'),
attrIF='%sattributes.index' % self.__db,
attrKF='%sattributes.key' % self.__db)
if 'Adult' not in genres: newlist.append(entry)
res[:] = newlist
# Get the real name, if this is an AKA.
# XXX: duplicated code!
new_res = []
seen_MID = []
for idx, (movieID, r) in enumerate(res):
# Remove duplicates.
# XXX: find a way to prefer titles with an AKA? Or prefer
# the original title?
if movieID in seen_MID:
continue
else:
seen_MID.append(movieID)
realMID = self._get_real_movieID(movieID)
if movieID == realMID:
new_res.append((movieID, r))
continue
if realMID in seen_MID:
continue
else:
seen_MID.append(realMID)
aka_title = build_title(r, canonical=0)
real_title = getLabel(realMID, '%stitles.index' % self.__db,
'%stitles.key' % self.__db)
if aka_title == real_title:
new_res.append((realMID, r))
continue
new_r = analyze_title(real_title, canonical=1)
new_r['akas'] = [aka_title]
new_res.append((realMID, new_r))
if results > 0: new_res[:] = new_res[:results]
return new_res
def get_imdbPersonID(self, personID):
"""Translate a personID in an imdbID.
Try an Exact Primary Name search on IMDb;
return None if it's unable to get the imdbID.
"""
name = getLabel(personID, '%snames.index' % self.__db,
'%snames.key' % self.__db)
if name is None: return None
return self.name2imdbID(name)
def get_imdbMovieID(self, movieID):
"""Translate a movieID in an imdbID.
Try an Exact Primary Title search on IMDb;
return None if it's unable to get the imdbID.
"""
titline = getLabel(movieID, '%stitles.index' % self.__db,
'%stitles.key' % self.__db)
if titline is None: return None
return self.title2imdbID(titline)
def get_imdbPersonID(self, personID):
"""Translate a personID in an imdbID.
Try an Exact Primary Name search on IMDb;
return None if it's unable to get the imdbID.
"""
name = getLabel(personID, '%snames.index' % self.__db,
'%snames.key' % self.__db)
if name is None: return None
return self.name2imdbID(name)
def get_imdbMovieID(self, movieID):
"""Translate a movieID in an imdbID.
Try an Exact Primary Title search on IMDb;
return None if it's unable to get the imdbID.
"""
titline = getLabel(movieID, '%stitles.index' % self.__db,
'%stitles.key' % self.__db)
if titline is None: return None
return self.title2imdbID(titline)
def _get_top_bottom_movies(self, kind):
if kind == 'top':
kind = 'top 250 rank'
elif kind == 'bottom':
kind = 'bottom 10 rank'
else:
return []
info = getTopBottomList(kind, '%stopbottom.db' % self.__db)
if not info:
return []
res = []
for d in info:
if not 'movieID' in d:
continue
movieID = d['movieID']
del d['movieID']
minfo = analyze_title(getLabel(movieID,
'%stitles.index' % self.__db,
'%stitles.key' % self.__db))
minfo.update(d)
res.append((movieID, minfo))
return res
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 _search_person(self, name, results):
name = name.strip()
if not name: return []
name1, name2, name3 = nameVariations(name)
res = _scan_names('%snames.key' % self.__db,
name1, name2, name3, results)
res[:] = [x[1] for x in res]
new_res = []
seen_PID = []
for idx, (personID, r) in enumerate(res):
# Remove duplicates.
# XXX: find a way to prefer names with an AKA? Or prefer
# the original name?
if personID in seen_PID:
continue
else:
seen_PID.append(personID)
realPID = self._get_real_personID(personID)
if personID == realPID:
new_res.append((personID, r))
continue
if realPID in seen_PID:
continue
else:
seen_PID.append(realPID)
aka_name = build_name(r, canonical=1)
real_name = getLabel(realPID, '%snames.index' % self.__db,
'%snames.key' % self.__db)
if aka_name == real_name:
new_res.append((realPID, r))
continue
new_r = analyze_name(real_name, canonical=1)
new_r['akas'] = [aka_name]
new_res.append((realPID, new_r))
if results > 0: new_res[:] = new_res[:results]
return new_res
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_keyword(self, keyword, results):
return [(movieID, analyze_title(getLabel(movieID,
'%stitles.index' % self.__db, '%stitles.key' % self.__db)))
for movieID in getKeywordMovies(keyword,
'%skeywords.data' % self.__db)][:results]
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}
