new backend allocine, in development

modules/allocine/__init__.py

@@ -0,0 +1,22 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 Julien Veyssier
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+from .backend import AllocineBackend
+
+__all__ = ['AllocineBackend']

modules/allocine/backend.py

@@ -0,0 +1,120 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 Julien Veyssier
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+from weboob.capabilities.cinema import ICapCinema, Person, Movie
+from weboob.tools.backend import BaseBackend
+
+from .browser import AllocineBrowser
+
+from urllib import quote_plus
+
+__all__ = ['AllocineBackend']
+
+
+class AllocineBackend(BaseBackend, ICapCinema):
+    NAME = 'allocine'
+    MAINTAINER = u'Julien Veyssier'
+    EMAIL = 'julien.veyssier@aiur.fr'
+    VERSION = '0.f'
+    DESCRIPTION = 'Allocine french cinema database service'
+    LICENSE = 'AGPLv3+'
+    BROWSER = AllocineBrowser
+
+    def create_default_browser(self):
+        return self.create_browser()
+
+    def get_movie(self, id):
+        return self.browser.get_movie(id)
+
+    def get_person(self, id):
+        return self.browser.get_person(id)
+
+    def iter_movies(self, pattern):
+        return self.browser.iter_movies(quote_plus(pattern.encode('utf-8')))
+
+    def iter_persons(self, pattern):
+        return self.browser.iter_persons(quote_plus(pattern.encode('utf-8')))
+
+    def iter_movie_persons(self, id, role=None):
+        return self.browser.iter_movie_persons(id, role)
+
+    def iter_person_movies(self, id, role=None):
+        return self.browser.iter_person_movies(id, role)
+
+    def iter_person_movies_ids(self, id):
+        return self.browser.iter_person_movies_ids(id)
+
+    def iter_movie_persons_ids(self, id):
+        return self.browser.iter_movie_persons_ids(id)
+
+    def get_person_biography(self, id):
+        return self.browser.get_person_biography(id)
+
+    def get_movie_releases(self, id, country=None):
+        return self.browser.get_movie_releases(id, country)
+
+    def fill_person(self, person, fields):
+        if 'real_name' in fields or 'birth_place' in fields\
+            or 'death_date' in fields or 'nationality' in fields\
+            or 'short_biography' in fields or 'roles' in fields\
+            or 'birth_date' in fields or 'thumbnail_url' in fields\
+                or 'gender' in fields or fields is None:
+            per = self.get_person(person.id)
+            person.real_name = per.real_name
+            person.birth_date = per.birth_date
+            person.death_date = per.death_date
+            person.birth_place = per.birth_place
+            person.gender = per.gender
+            person.nationality = per.nationality
+            person.short_biography = per.short_biography
+            person.short_description = per.short_description
+            person.roles = per.roles
+            person.thumbnail_url = per.thumbnail_url
+
+        if 'biography' in fields:
+            person.biography = self.get_person_biography(person.id)
+
+        return person
+
+    def fill_movie(self, movie, fields):
+        if 'other_titles' in fields or 'release_date' in fields\
+            or 'duration' in fields or 'country' in fields\
+            or 'roles' in fields or 'note' in fields\
+                or 'thumbnail_url' in fields:
+            mov = self.get_movie(movie.id)
+            movie.other_titles = mov.other_titles
+            movie.release_date = mov.release_date
+            movie.duration = mov.duration
+            movie.pitch = mov.pitch
+            movie.country = mov.country
+            movie.note = mov.note
+            movie.roles = mov.roles
+            movie.genres = mov.genres
+            movie.short_description = mov.short_description
+            movie.thumbnail_url = mov.thumbnail_url
+
+        if 'all_release_dates' in fields:
+            movie.all_release_dates = self.get_movie_releases(movie.id)
+
+        return movie
+
+    OBJECTS = {
+        Person: fill_person,
+        Movie: fill_movie
+    }

modules/allocine/browser.py

@@ -0,0 +1,238 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 Julien Veyssier
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+import HTMLParser
+from weboob.tools.browser import BaseBrowser, BrowserHTTPNotFound
+from weboob.capabilities.base import NotAvailable, NotLoaded
+from weboob.capabilities.cinema import Movie, Person
+from weboob.tools.json import json
+
+from .pages import PersonPage, MovieCrewPage, BiographyPage, FilmographyPage, ReleasePage
+
+from datetime import datetime
+
+__all__ = ['AllocineBrowser']
+
+
+class AllocineBrowser(BaseBrowser):
+    DOMAIN = 'api.allocine.fr'
+    PROTOCOL = 'http'
+    ENCODING = 'utf-8'
+    USER_AGENT = BaseBrowser.USER_AGENTS['wget']
+    #PAGES = {
+    #    'http://www.imdb.com/title/tt[0-9]*/fullcredits.*': MovieCrewPage,
+    #    'http://www.imdb.com/title/tt[0-9]*/releaseinfo.*': ReleasePage,
+    #    'http://www.imdb.com/name/nm[0-9]*/*': PersonPage,
+    #    'http://www.imdb.com/name/nm[0-9]*/bio.*': BiographyPage,
+    #    'http://www.imdb.com/name/nm[0-9]*/filmo.*': FilmographyPage,
+    #}
+
+    def iter_movies(self, pattern):
+        res = self.readurl('http://api.allocine.fr/rest/v3/search?partner=YW5kcm9pZC12M3M&filter=movie&q=%s&format=json' % pattern.encode('utf-8'))
+        jres = json.loads(res)
+        for m in jres['feed']['movie']:
+            tdesc = u''
+            if 'title' in m:
+                tdesc += '%s' % m['title']
+            if 'productionYear' in m:
+                tdesc += ' ; %s' % m['productionYear']
+            elif 'release' in m:
+                tdesc += ' ; %s' % m['release']['releaseDate']
+            short_description = tdesc.strip('; ')
+            movie = Movie(m['code'], unicode(m['originalTitle']))
+            movie.other_titles = NotLoaded
+            movie.release_date = NotLoaded
+            movie.duration = NotLoaded
+            movie.short_description = short_description
+            movie.pitch = NotLoaded
+            movie.country = NotLoaded
+            movie.note = NotLoaded
+            movie.roles = NotLoaded
+            movie.all_release_dates = NotLoaded
+            movie.thumbnail_url = NotLoaded
+            yield movie
+
+    def iter_persons(self, pattern):
+        res = self.readurl('http://api.allocine.fr/rest/v3/search?partner=YW5kcm9pZC12M3M&filter=person&q=%s&format=json' % pattern.encode('utf-8'))
+        jres = json.loads(res)
+        for p in jres['feed']['person']:
+            thumbnail_url = NotAvailable
+            if 'picture' in p:
+                thumbnail_url = unicode(p['picture']['href'])
+            person = Person(p['code'], unicode(p['name']))
+            desc = u''
+            if 'birthDate' in p:
+                desc += '(%s), ' % p['birthDate']
+            if 'activity' in p:
+                for a in p['activity']:
+                    desc += '%s, ' % a['$']
+            person.real_name = NotLoaded
+            person.birth_place = NotLoaded
+            person.birth_date = NotLoaded
+            person.death_date = NotLoaded
+            person.gender = NotLoaded
+            person.nationality = NotLoaded
+            person.short_biography = NotLoaded
+            person.short_description = desc.strip(', ')
+            person.roles = NotLoaded
+            person.thumbnail_url = thumbnail_url
+            yield person
+
+    def get_movie(self, id):
+        res = self.readurl(
+                'http://api.allocine.fr/rest/v3/movie?partner=YW5kcm9pZC12M3M&code=%s&profile=large&mediafmt=mp4-lc&format=json&filter=movie&striptags=synopsis,synopsisshort' % id)
+        if res is not None:
+            jres = json.loads(res)['movie']
+        else:
+            return None
+        title = NotAvailable
+        duration = NotAvailable
+        release_date = NotAvailable
+        pitch = NotAvailable
+        country = NotAvailable
+        note = NotAvailable
+        short_description = NotAvailable
+        thumbnail_url = NotAvailable
+        other_titles = []
+        genres = []
+        roles = {}
+
+        if 'originalTitle' not in jres:
+            return
+        title = unicode(jres['originalTitle'].strip())
+        if 'picture' in jres:
+            thumbnail_url = unicode(jres['picture']['href'])
+        if 'genre' in jres:
+            for g in jres['genre']:
+                genres.append(g['$'])
+        if 'runtime' in jres:
+            nbsecs = jres['runtime']
+            duration = nbsecs / 60
+        #if 'also_known_as' in jres:
+        #    for other_t in jres['also_known_as']:
+        #        if 'country' in other_t and 'title' in other_t:
+        #            other_titles.append('%s : %s' % (other_t['country'], htmlparser.unescape(other_t['title'])))
+        if 'release' in jres:
+            dstr = str(jres['release']['releaseDate'])
+            tdate = dstr.split('-')
+            day = 1
+            month = 1
+            year = 1901
+            if len(tdate) > 2:
+                year = int(tdate[0])
+                month = int(tdate[1])
+                day = int(tdate[2])
+            release_date = datetime(year, month, day)
+        if 'nationality' in jres:
+            country = u''
+            for c in jres['nationality']:
+                country += '%s, ' % c['$']
+            country = country.strip(', ')
+        if 'synopsis' in jres:
+            pitch = unicode(jres['synopsis'])
+        if 'statistics' in jres and 'userRating' in jres['statistics']:
+            note = u'%s/10 (%s votes)' % (jres['statistics']['userRating'], jres['statistics']['userReviewCount'])
+        if 'castMember' in jres:
+            for cast in jres['castMember']:
+                if cast['activity']['$'] not in roles:
+                    roles[cast['activity']['$']] = []
+                roles[cast['activity']['$']].append(cast['person']['name'])
+
+        movie = Movie(id, title)
+        movie.other_titles = other_titles
+        movie.release_date = release_date
+        movie.duration = duration
+        movie.genres = genres
+        movie.pitch = pitch
+        movie.country = country
+        movie.note = note
+        movie.roles = roles
+        movie.short_description = short_description
+        movie.all_release_dates = NotLoaded
+        movie.thumbnail_url = thumbnail_url
+        return movie
+
+    def get_person(self, id):
+        try:
+            self.location('http://www.imdb.com/name/%s' % id)
+        except BrowserHTTPNotFound:
+            return
+        assert self.is_on_page(PersonPage)
+        return self.page.get_person(id)
+
+    def get_person_biography(self, id):
+        self.location('http://www.imdb.com/name/%s/bio' % id)
+        assert self.is_on_page(BiographyPage)
+        return self.page.get_biography()
+
+    def iter_movie_persons(self, movie_id, role):
+        self.location('http://www.imdb.com/title/%s/fullcredits' % movie_id)
+        assert self.is_on_page(MovieCrewPage)
+        for p in self.page.iter_persons(role):
+            yield p
+
+    def iter_person_movies(self, person_id, role):
+        self.location('http://www.imdb.com/name/%s/filmotype' % person_id)
+        assert self.is_on_page(FilmographyPage)
+        return self.page.iter_movies(role)
+
+    def iter_person_movies_ids(self, person_id):
+        self.location('http://www.imdb.com/name/%s/filmotype' % person_id)
+        assert self.is_on_page(FilmographyPage)
+        for movie in self.page.iter_movies_ids():
+            yield movie
+
+    def iter_movie_persons_ids(self, movie_id):
+        self.location('http://www.imdb.com/title/%s/fullcredits' % movie_id)
+        assert self.is_on_page(MovieCrewPage)
+        for person in self.page.iter_persons_ids():
+            yield person
+
+    def get_movie_releases(self, id, country):
+        return
+        self.location('http://www.imdb.com/title/%s/releaseinfo' % id)
+        assert self.is_on_page(ReleasePage)
+        return self.page.get_movie_releases(country)
+
+
+dict_hex = {'&#xE1;': u'á',
+            '&#xE9;': u'é',
+            '&#xE8;': u'è',
+            '&#xED;': u'í',
+            '&#xF1;': u'ñ',
+            '&#xF3;': u'ó',
+            '&#xFA;': u'ú',
+            '&#xFC;': u'ü',
+            '&#x26;': u'&',
+            '&#x27;': u"'",
+            '&#xE0;': u'à',
+            '&#xC0;': u'À',
+            '&#xE2;': u'â',
+            '&#xC9;': u'É',
+            '&#xEB;': u'ë',
+            '&#xF4;': u'ô',
+            '&#xE7;': u'ç'
+            }
+
+
+def latin2unicode(word):
+    for key in dict_hex.keys():
+        word = word.replace(key, dict_hex[key])
+    return unicode(word)

modules/allocine/pages.py

@@ -0,0 +1,231 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 Julien Veyssier
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+
+from weboob.capabilities.cinema import Person, Movie
+from weboob.capabilities.base import NotAvailable, NotLoaded
+from weboob.tools.browser import BasePage
+
+from datetime import datetime
+
+
+__all__ = ['PersonPage', 'MovieCrewPage', 'BiographyPage', 'FilmographyPage', 'ReleasePage']
+
+
+class ReleasePage(BasePage):
+    ''' Page containing releases of a movie
+    '''
+    def get_movie_releases(self, country_filter):
+        result = unicode()
+        links = self.parser.select(self.document.getroot(), 'b a')
+        for a in links:
+            href = a.attrib.get('href', '')
+            if href.strip('/').split('/')[0] == 'calendar' and\
+                    (country_filter is None or href.split('region=')[-1].lower() == country_filter):
+                country = a.text
+                td_date = self.parser.select(a.getparent().getparent().getparent(), 'td')[1]
+                date_links = self.parser.select(td_date, 'a')
+                if len(date_links) > 1:
+                    date = date_links[1].attrib.get('href', '').strip('/').split('/')[-1]
+                    date += '-'+date_links[0].attrib.get('href', '').strip('/').split('/')[-1]
+                else:
+                    date = unicode(self.parser.select(a.getparent().getparent().getparent(), 'td')[1].text_content())
+                result += '%s : %s\n' % (country, date)
+        if result == u'':
+            result = NotAvailable
+        else:
+            result = result.strip()
+        return result
+
+
+class BiographyPage(BasePage):
+    ''' Page containing biography of a person
+    '''
+    def get_biography(self):
+        bio = unicode()
+        tn = self.parser.select(self.document.getroot(), 'div#tn15content', 1)
+        # we only read paragraphs, titles and links
+        for ch in tn.getchildren():
+            if ch.tag in ['p', 'h5', 'a']:
+                bio += '%s\n\n' % ch.text_content().strip()
+        if bio == u'':
+            bio = NotAvailable
+        return bio
+
+
+class MovieCrewPage(BasePage):
+    ''' Page listing all the persons related to a movie
+    '''
+    def iter_persons(self, role_filter=None):
+        if (role_filter is None or (role_filter is not None and role_filter == 'actor')):
+            tables = self.parser.select(self.document.getroot(), 'table.cast')
+            if len(tables) > 0:
+                table = tables[0]
+                tds = self.parser.select(table, 'td.nm')
+                for td in tds:
+                    id = td.find('a').attrib.get('href', '').strip('/').split('/')[-1]
+                    name = unicode(td.find('a').text)
+                    char_name = unicode(self.parser.select(td.getparent(), 'td.char', 1).text_content())
+                    person = Person(id, name)
+                    person.short_description = char_name
+                    person.real_name = NotLoaded
+                    person.birth_place = NotLoaded
+                    person.birth_date = NotLoaded
+                    person.death_date = NotLoaded
+                    person.gender = NotLoaded
+                    person.nationality = NotLoaded
+                    person.short_biography = NotLoaded
+                    person.roles = NotLoaded
+                    person.thumbnail_url = NotLoaded
+                    yield person
+
+        for gloss_link in self.parser.select(self.document.getroot(), 'table[cellspacing=1] h5 a'):
+            role = gloss_link.attrib.get('name', '').rstrip('s')
+            if (role_filter is None or (role_filter is not None and role == role_filter)):
+                tbody = gloss_link.getparent().getparent().getparent().getparent()
+                for line in self.parser.select(tbody, 'tr')[1:]:
+                    for a in self.parser.select(line, 'a'):
+                        role_detail = NotAvailable
+                        href = a.attrib.get('href', '')
+                        if '/name/nm' in href:
+                            id = href.strip('/').split('/')[-1]
+                            name = unicode(a.text)
+                        if 'glossary' in href:
+                            role_detail = unicode(a.text)
+                        person = Person(id, name)
+                        person.short_description = role_detail
+                        yield person
+                            # yield self.browser.get_person(id)
+
+    def iter_persons_ids(self):
+        tables = self.parser.select(self.document.getroot(), 'table.cast')
+        if len(tables) > 0:
+            table = tables[0]
+            tds = self.parser.select(table, 'td.nm')
+            for td in tds:
+                id = td.find('a').attrib.get('href', '').strip('/').split('/')[-1]
+                yield id
+
+
+class PersonPage(BasePage):
+    ''' Page giving informations about a person
+    It is used to build a Person instance and to get the movie list related to a person
+    '''
+    def get_person(self, id):
+        name = NotAvailable
+        short_biography = NotAvailable
+        short_description = NotAvailable
+        birth_place = NotAvailable
+        birth_date = NotAvailable
+        death_date = NotAvailable
+        real_name = NotAvailable
+        gender = NotAvailable
+        thumbnail_url = NotAvailable
+        roles = {}
+        nationality = NotAvailable
+        td_overview = self.parser.select(self.document.getroot(), 'td#overview-top', 1)
+        descs = self.parser.select(td_overview, 'span[itemprop=description]')
+        if len(descs) > 0:
+            short_biography = unicode(descs[0].text)
+        rname_block = self.parser.select(td_overview, 'div.txt-block h4.inline')
+        if len(rname_block) > 0 and "born" in rname_block[0].text.lower():
+            links = self.parser.select(rname_block[0].getparent(), 'a')
+            for a in links:
+                href = a.attrib.get('href', '').strip()
+                if href == 'bio':
+                    real_name = unicode(a.text.strip())
+                elif 'birth_place' in href:
+                    birth_place = unicode(a.text.lower().strip())
+        names = self.parser.select(td_overview, 'h1[itemprop=name]')
+        if len(names) > 0:
+            name = unicode(names[0].text.strip())
+        times = self.parser.select(td_overview, 'time[itemprop=birthDate]')
+        if len(times) > 0:
+            time = times[0].attrib.get('datetime', '').split('-')
+            if len(time) == 3 and int(time[0]) >= 1900:
+                birth_date = datetime(int(time[0]), int(time[1]), int(time[2]))
+        dtimes = self.parser.select(td_overview, 'time[itemprop=deathDate]')
+        if len(dtimes) > 0:
+            dtime = dtimes[0].attrib.get('datetime', '').split('-')
+            if len(dtime) == 3 and int(dtime[0]) >= 1900:
+                death_date = datetime(int(dtime[0]), int(dtime[1]), int(dtime[2]))
+        img_thumbnail = self.parser.select(self.document.getroot(), 'td#img_primary img')
+        if len(img_thumbnail) > 0:
+            thumbnail_url = unicode(img_thumbnail[0].attrib.get('src', ''))
+
+        # go to the filmography page
+        self.browser.location('http://www.imdb.com/name/%s/filmotype' % id)
+        assert self.browser.is_on_page(FilmographyPage)
+        roles = self.browser.page.get_roles()
+
+        person = Person(id, name)
+        person.real_name = real_name
+        person.birth_date = birth_date
+        person.death_date = death_date
+        person.birth_place = birth_place
+        person.gender = gender
+        person.nationality = nationality
+        person.short_biography = short_biography
+        person.short_description = short_description
+        person.roles = roles
+        person.thumbnail_url = thumbnail_url
+        return person
+
+
+class FilmographyPage(BasePage):
+    ''' Page of detailed filmography of a person, sorted by type of role
+    This page is easier to parse than the main person page filmography
+    '''
+    def iter_movies_ids(self):
+        for role_div in self.parser.select(self.document.getroot(), 'div.filmo'):
+            for a in self.parser.select(role_div, 'ol > li > a'):
+                id = a.attrib.get('href', '').strip('/').split('/')[-1]
+                if id.startswith('tt'):
+                    yield id
+
+    def get_roles(self):
+        roles = {}
+        for role_div in self.parser.select(self.document.getroot(), 'div.filmo'):
+            role = self.parser.select(role_div, 'h5 a', 1).text.replace(':', '')
+            roles[role] = []
+            for a in self.parser.select(role_div, 'ol > li > a'):
+                id = a.attrib.get('href', '').strip('/').split('/')[-1]
+                if id.startswith('tt'):
+                    if '(' in a.tail and ')' in a.tail:
+                        between_p = a.tail.split(')')[0].split('(')[1]
+                    else:
+                        between_p = '????'
+                    roles[role].append('(%s) %s' % (between_p, a.text))
+        return roles
+
+    def iter_movies(self, role_filter=None):
+        for role_div in self.parser.select(self.document.getroot(), 'div.filmo'):
+            role = self.parser.select(role_div, 'h5 a', 1).text.replace(':', '')
+            if (role_filter is None or (role_filter is not None and role.lower().strip() == role_filter))\
+                    and role != 'In Development':
+                for a in self.parser.select(role_div, 'ol > li > a'):
+                    id = a.attrib.get('href', '').strip('/').split('/')[-1]
+                    if id.startswith('tt'):
+                        title = unicode(a.text)
+                        role_detail = NotAvailable
+                        if len(a.tail) > 0:
+                            role_detail = unicode(' '.join(a.tail.replace('..', '').split()))
+                        movie = Movie(id, title)
+                        movie.short_description = role_detail
+                        yield movie

modules/allocine/test.py

@@ -0,0 +1,67 @@
+# -*- coding: utf-8 -*-
+
+# Copyright(C) 2013 Julien Veyssier
+#
+# This file is part of weboob.
+#
+# weboob is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# weboob is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with weboob. If not, see <http://www.gnu.org/licenses/>.
+
+from weboob.tools.test import BackendTest
+
+
+class ImdbTest(BackendTest):
+    BACKEND = 'imdb'
+
+    def test_search_movie(self):
+        movies = list(self.backend.iter_movies('spiderman'))
+        for movie in movies:
+            assert movie.id
+
+    def test_get_movie(self):
+        movie = self.backend.get_movie('tt0079980')
+        assert movie.id
+        assert movie.original_title
+
+    def test_search_person(self):
+        persons = list(self.backend.iter_persons('dewaere'))
+        for person in persons:
+            assert person.id
+
+    def test_get_person(self):
+        person = self.backend.get_person('nm0223033')
+        assert person.id
+        assert person.name
+        assert person.birth_date
+
+    def test_movie_persons(self):
+        persons = list(self.backend.iter_movie_persons('tt0079980'))
+        for person in persons:
+            assert person.id
+            assert person.name
+
+    def test_person_movies(self):
+        movies = list(self.backend.iter_person_movies('nm0223033'))
+        for movie in movies:
+            assert movie.id
+            assert movie.original_title
+
+    def test_get_person_biography(self):
+        bio = self.backend.get_person_biography('nm0223033')
+        assert bio != ''
+        assert bio is not None
+
+    def test_get_movie_releases(self):
+        rel = self.backend.get_movie_releases('tt0079980')
+        assert rel != ''
+        assert rel is not None