Prepare to upload

[youtubedl] / youtube_dl / extractor / francetv.py

diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py
index f3e0f38b7200a70c897dd561b45a275cf42f7193..095bb3954c523852be8d7d09d498d324d429c826 100644 (file)
--- a/youtube_dl/extractor/francetv.py
+++ b/youtube_dl/extractor/francetv.py
@@ -1,54 +1,208 @@
-# encoding: utf-8
+# coding: utf-8

 
 from __future__ import unicode_literals
 
 import re
 
 from __future__ import unicode_literals
 
 import re

-import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

+from ..compat import compat_urlparse

 from ..utils import (
 from ..utils import (

-    compat_urlparse,
+    clean_html,
+    ExtractorError,
+    int_or_none,
+    parse_duration,
+    determine_ext,

 )
 )

+from .dailymotion import DailymotionIE

 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):

-    def _extract_video(self, video_id):
-        info = self._download_xml(
-            'http://www.francetvinfo.fr/appftv/webservices/video/'
-            'getInfosOeuvre.php?id-diffusion='
-            + video_id, video_id, 'Downloading XML config')
+    def _extract_video(self, video_id, catalogue=None):
+        info = self._download_json(
+            'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/',
+            video_id, 'Downloading video JSON', query={
+                'idDiffusion': video_id,
+                'catalogue': catalogue or '',
+            })

 
 

-        manifest_url = info.find('videos/video/url').text
-        video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8')
-        video_url = video_url.replace('/z/', '/i/')
-        thumbnail_path = info.find('image').text
+        if info.get('status') == 'NOK':
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, info['message']), expected=True)
+        allowed_countries = info['videos'][0].get('geoblocage')
+        if allowed_countries:
+            georestricted = True
+            geo_info = self._download_json(
+                'http://geo.francetv.fr/ws/edgescape.json', video_id,
+                'Downloading geo restriction info')
+            country = geo_info['reponse']['geo_info']['country_code']
+            if country not in allowed_countries:
+                raise ExtractorError(
+                    'The video is not available from your location',
+                    expected=True)
+        else:
+            georestricted = False
+
+        formats = []
+        for video in info['videos']:
+            if video['statut'] != 'ONLINE':
+                continue
+            video_url = video['url']
+            if not video_url:
+                continue
+            format_id = video['format']
+            ext = determine_ext(video_url)
+            if ext == 'f4m':
+                if georestricted:
+                    # See https://github.com/rg3/youtube-dl/issues/3963
+                    # m3u8 urls work fine
+                    continue
+                f4m_url = self._download_webpage(
+                    'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url,
+                    video_id, 'Downloading f4m manifest token', fatal=False)
+                if f4m_url:
+                    formats.extend(self._extract_f4m_formats(
+                        f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
+                        video_id, f4m_id=format_id, fatal=False))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id=format_id, fatal=False))
+            elif video_url.startswith('rtmp'):
+                formats.append({
+                    'url': video_url,
+                    'format_id': 'rtmp-%s' % format_id,
+                    'ext': 'flv',
+                })
+            else:
+                if self._is_valid_url(video_url, video_id, format_id):
+                    formats.append({
+                        'url': video_url,
+                        'format_id': format_id,
+                    })
+        self._sort_formats(formats)

 
 

-        return {'id': video_id,
-                'ext': 'flv' if video_url.startswith('rtmp') else 'mp4',
-                'url': video_url,
-                'title': info.find('titre').text,
-                'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path),
-                'description': info.find('synopsis').text,
-                }
+        title = info['titre']
+        subtitle = info.get('sous_titre')
+        if subtitle:
+            title += ' - %s' % subtitle
+        title = title.strip()

 
 

+        subtitles = {}
+        subtitles_list = [{
+            'url': subformat['url'],
+            'ext': subformat.get('format'),
+        } for subformat in info.get('subtitles', []) if subformat.get('url')]
+        if subtitles_list:
+            subtitles['fr'] = subtitles_list
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': clean_html(info['synopsis']),
+            'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
+            'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
+            'timestamp': int_or_none(info['diffusion']['timestamp']),
+            'formats': formats,
+            'subtitles': subtitles,
+        }

 
 

-class PluzzIE(FranceTVBaseInfoExtractor):
-    IE_NAME = 'pluzz.francetv.fr'
-    _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'

 
 

-    # Can't use tests, videos expire in 7 days
+class FranceTVIE(FranceTVBaseInfoExtractor):
+    _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)*(?P<id>[^/]+)\.html'
+
+    _TESTS = [{
+        'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
+        'info_dict': {
+            'id': '157550144',
+            'ext': 'mp4',
+            'title': '13h15, le dimanche... - Les mystères de Jésus',
+            'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
+            'timestamp': 1494156300,
+            'upload_date': '20170507',
+        },
+        'params': {
+            # m3u8 downloads
+            'skip_download': True,
+        },
+    }, {
+        # france3
+        'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
+        'only_matching': True,
+    }, {
+        # france4
+        'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
+        'only_matching': True,
+    }, {
+        # france5
+        'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
+        'only_matching': True,
+    }, {
+        # franceo
+        'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
+        'only_matching': True,
+    }, {
+        # france2 live
+        'url': 'https://www.france.tv/france-2/direct.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.france.tv/142749-rouge-sang.html',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        title = re.match(self._VALID_URL, url).group(1)
-        webpage = self._download_webpage(url, title)
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        catalogue = None

         video_id = self._search_regex(
         video_id = self._search_regex(

-            r'data-diffusion="(\d+)"', webpage, 'ID')
-        return self._extract_video(video_id)
+            r'data-main-video=(["\'])(?P<id>(?:(?!\1).)+)\1',
+            webpage, 'video id', default=None, group='id')
+
+        if not video_id:
+            video_id, catalogue = self._html_search_regex(
+                r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
+                webpage, 'video ID').split('@')
+        return self._extract_video(video_id, catalogue)
+
+
+class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
+    _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
+
+    _TEST = {
+        'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
+        'info_dict': {
+            'id': 'NI_983319',
+            'ext': 'mp4',
+            'title': 'Le Pen Reims',
+            'upload_date': '20170505',
+            'timestamp': 1493981780,
+            'duration': 16,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        video = self._download_json(
+            'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
+            video_id)

 
 

+        return self._extract_video(video['video_id'], video.get('catalog'))

 
 

-class FranceTvInfoIE(FranceTVBaseInfoExtractor):
+
+class FranceTVInfoIE(FranceTVBaseInfoExtractor):

     IE_NAME = 'francetvinfo.fr'
     IE_NAME = 'francetvinfo.fr'

-    _VALID_URL = r'https?://www\.francetvinfo\.fr/.*/(?P<title>.+)\.html'
+    _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<title>[^/?#&.]+)'

 
     _TESTS = [{
         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
 
     _TESTS = [{
         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',


@@ -56,8 +210,14 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
             'id': '84981923',
             'ext': 'mp4',
             'title': 'Soir 3',
             'id': '84981923',
             'ext': 'mp4',
             'title': 'Soir 3',

+            'upload_date': '20130826',
+            'timestamp': 1377548400,
+            'subtitles': {
+                'fr': 'mincount:2',
+            },

         },
         'params': {
         },
         'params': {

+            # m3u8 downloads

             'skip_download': True,
         },
     }, {
             'skip_download': True,
         },
     }, {


@@ -70,166 +230,133 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
         },
         'params': {
             'skip_download': 'HLS (reqires ffmpeg)'
         },
         'params': {
             'skip_download': 'HLS (reqires ffmpeg)'

-        }
-    }]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        page_title = mobj.group('title')
-        webpage = self._download_webpage(url, page_title)
-        video_id = self._search_regex(r'id-video=((?:[^0-9]*?_)?[0-9]+)[@"]', webpage, 'video id')
-        return self._extract_video(video_id)
-
-
-class FranceTVIE(FranceTVBaseInfoExtractor):
-    IE_NAME = 'francetv'
-    IE_DESC = 'France 2, 3, 4, 5 and Ô'
-    _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/
-        (?:
-            emissions/.*?/(videos|emissions)/(?P<id>[^/?]+)
-        |   (emissions?|jt)/(?P<key>[^/?]+)
-        )'''
-
-    _TESTS = [
-        # france2
-        {
-            'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
-            'file': '75540104.mp4',
-            'info_dict': {
-                'title': '13h15, le samedi...',
-                'description': 'md5:2e5b58ba7a2d3692b35c792be081a03d',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },

         },
         },

-        # france3
-        {
-            'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
-            'info_dict': {
-                'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
-                'ext': 'flv',
-                'title': 'Le scandale du prix des médicaments',
-                'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
+        'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
+    }, {
+        'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
+        'md5': 'f485bda6e185e7d15dbc69b72bae993e',
+        'info_dict': {
+            'id': 'NI_173343',
+            'ext': 'mp4',
+            'title': 'Les entreprises familiales : le secret de la réussite',
+            'thumbnail': r're:^https?://.*\.jpe?g$',
+            'timestamp': 1433273139,
+            'upload_date': '20150602',

         },
         },

-        # france4
-        {
-            'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
-            'info_dict': {
-                'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
-                'ext': 'flv',
-                'title': 'Hero Corp Making of - Extrait 1',
-                'description': 'md5:c87d54871b1790679aec1197e73d650a',
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
+        'params': {
+            # m3u8 downloads
+            'skip_download': True,

         },
         },

-        # france5
-        {
-            'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
-            'info_dict': {
-                'id': '92837968',
-                'ext': 'mp4',
-                'title': 'C à dire ?!',
-                'description': 'md5:fb1db1cbad784dcce7c7a7bd177c8e2f',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
+    }, {
+        'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
+        'md5': 'f485bda6e185e7d15dbc69b72bae993e',
+        'info_dict': {
+            'id': 'NI_657393',
+            'ext': 'mp4',
+            'title': 'Olivier Monthus, réalisateur de "Bretagne, le choix de l’Armor"',
+            'description': 'md5:a3264114c9d29aeca11ced113c37b16c',
+            'thumbnail': r're:^https?://.*\.jpe?g$',
+            'timestamp': 1458300695,
+            'upload_date': '20160318',

         },
         },

-        # franceo
-        {
-            'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
-            'info_dict': {
-                'id': '92327925',
-                'ext': 'mp4',
-                'title': 'Infô-Afrique',
-                'description': 'md5:ebf346da789428841bee0fd2a935ea55',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-            'skip': 'The id changes frequently',
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # Dailymotion embed
+        'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
+        'md5': 'ee7f1828f25a648addc90cb2687b1f12',
+        'info_dict': {
+            'id': 'x4iiko0',
+            'ext': 'mp4',
+            'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
+            'description': 'Au lendemain de la victoire du "oui" au référendum sur l\'aéroport de Notre-Dame-des-Landes, l\'ancienne ministre écologiste est l\'invitée de Patrick Cohen. Plus d\'info : https://www.franceinter.fr/emissions/le-7-9/le-7-9-27-juin-2016',
+            'timestamp': 1467011958,
+            'upload_date': '20160627',
+            'uploader': 'France Inter',
+            'uploader_id': 'x2q2ez',

         },
         },

-    ]
+        'add_ie': ['Dailymotion'],
+    }, {
+        'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        if mobj.group('key'):
-            webpage = self._download_webpage(url, mobj.group('key'))
-            id_res = [
-                (r'''(?x)<div\s+class="video-player">\s*
-                    <a\s+href="http://videos.francetv.fr/video/([0-9]+)"\s+
-                    class="francetv-video-player">'''),
-                (r'<a id="player_direct" href="http://info\.francetelevisions'
-                 '\.fr/\?id-video=([^"/&]+)'),
-                (r'<a class="video" id="ftv_player_(.+?)"'),
-            ]
-            video_id = self._html_search_regex(id_res, webpage, 'video ID')
-        else:
-            video_id = mobj.group('id')
-        return self._extract_video(video_id)
+        page_title = mobj.group('title')
+        webpage = self._download_webpage(url, page_title)

 
 

+        dailymotion_urls = DailymotionIE._extract_urls(webpage)
+        if dailymotion_urls:
+            return self.playlist_result([
+                self.url_result(dailymotion_url, DailymotionIE.ie_key())
+                for dailymotion_url in dailymotion_urls])

 
 

-class GenerationQuoiIE(InfoExtractor):
-    IE_NAME = 'france2.fr:generation-quoi'
-    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
+        video_id, catalogue = self._search_regex(
+            (r'id-video=([^@]+@[^"]+)',
+             r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'),
+            webpage, 'video id').split('@')
+        return self._extract_video(video_id, catalogue)

 
 

-    _TEST = {
-        'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
-        'file': 'k7FJX8VBcvvLmX4wA5Q.mp4',
+
+class GenerationWhatIE(InfoExtractor):
+    IE_NAME = 'france2.fr:generation-what'
+    _VALID_URL = r'https?://generation-what\.francetv\.fr/[^/]+/video/(?P<id>[^/?#]+)'
+
+    _TESTS = [{
+        'url': 'http://generation-what.francetv.fr/portrait/video/present-arms',

         'info_dict': {
         'info_dict': {

-            'title': 'Génération Quoi - Garde à Vous',
-            'uploader': 'Génération Quoi',
-        },
-        'params': {
-            # It uses Dailymotion
-            'skip_download': True,
+            'id': 'wtvKYUG45iw',
+            'ext': 'mp4',
+            'title': 'Generation What - Garde à vous - FRA',
+            'uploader': 'Generation What',
+            'uploader_id': 'UCHH9p1eetWCgt4kXBYCb3_w',
+            'upload_date': '20160411',

         },
         },

-        'skip': 'Only available from France',
-    }
+    }, {
+        'url': 'http://generation-what.francetv.fr/europe/video/present-arms',
+        'only_matching': True,
+    }]

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VALID_URL, url)
-        name = mobj.group('name')
-        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % name)
-        info_json = self._download_webpage(info_url, name)
-        info = json.loads(info_json)
-        return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
-            ie='Dailymotion')
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        youtube_id = self._search_regex(
+            r"window\.videoURL\s*=\s*'([0-9A-Za-z_-]{11})';",
+            webpage, 'youtube id')
+        return self.url_result(youtube_id, 'Youtube', youtube_id)

 
 
 class CultureboxIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'culturebox.francetvinfo.fr'
 
 
 class CultureboxIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'culturebox.francetvinfo.fr'

-    _VALID_URL = r'https?://culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
+    _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'

 
     _TEST = {
 
     _TEST = {

-        'url': 'http://culturebox.francetvinfo.fr/einstein-on-the-beach-au-theatre-du-chatelet-146813',
+        'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511',
+        'md5': '9b88dc156781c4dbebd4c3e066e0b1d6',

         'info_dict': {
         'info_dict': {

-            'id': 'EV_6785',
-            'ext': 'mp4',
-            'title': 'Einstein on the beach au Théâtre du Châtelet',
-            'description': 'md5:9ce2888b1efefc617b5e58b3f6200eeb',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
+            'id': 'EV_50111',
+            'ext': 'flv',
+            'title': "Le Livre Vermeil de Montserrat à la Cathédrale d'Elne",
+            'description': 'md5:f8a4ad202e8fe533e2c493cc12e739d9',
+            'upload_date': '20150320',
+            'timestamp': 1426892400,
+            'duration': 2760.9,

         },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         name = mobj.group('name')
         },
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         name = mobj.group('name')

+

         webpage = self._download_webpage(url, name)
         webpage = self._download_webpage(url, name)

-        video_id = self._search_regex(r'"http://videos\.francetv\.fr/video/(.*?)"', webpage, 'video id')
-        return self._extract_video(video_id)
+
+        if ">Ce live n'est plus disponible en replay<" in webpage:
+            raise ExtractorError('Video %s is not available' % name, expected=True)
+
+        video_id, catalogue = self._search_regex(
+            r'["\'>]https?://videos\.francetv\.fr/video/([^@]+@.+?)["\'<]',
+            webpage, 'video id').split('@')
+
+        return self._extract_video(video_id, catalogue)