]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/francetv.py
Imported Upstream version 2015.06.04.1
[youtubedl] / youtube_dl / extractor / francetv.py
index 51eb97b2f6290d11c9a2a7d9c4bf8a785f54239b..edf555b2987520618b70bf8bd423c5fc1f60e5a9 100644 (file)
@@ -6,30 +6,90 @@ import re
 import json
 
 from .common import InfoExtractor
 import json
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
+    compat_urllib_parse_urlparse,
     compat_urlparse,
 )
     compat_urlparse,
 )
+from ..utils import (
+    clean_html,
+    ExtractorError,
+    int_or_none,
+    float_or_none,
+    parse_duration,
+    determine_ext,
+)
 
 
 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')
-
-        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
-
-        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,
-                }
+    def _extract_video(self, video_id, catalogue):
+        info = self._download_json(
+            'http://webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/?idDiffusion=%s&catalogue=%s'
+            % (video_id, catalogue),
+            video_id, 'Downloading video JSON')
+
+        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
+                video_url_parsed = compat_urllib_parse_urlparse(video_url)
+                f4m_url = self._download_webpage(
+                    'http://hdfauth.francetv.fr/esi/urltokengen2.html?url=%s' % video_url_parsed.path,
+                    video_id, 'Downloading f4m manifest token', fatal=False)
+                if f4m_url:
+                    formats.extend(self._extract_f4m_formats(f4m_url, video_id, 1, format_id))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(video_url, video_id, 'mp4', m3u8_id=format_id))
+            elif video_url.startswith('rtmp'):
+                formats.append({
+                    'url': video_url,
+                    'format_id': 'rtmp-%s' % format_id,
+                    'ext': 'flv',
+                    'preference': 1,
+                })
+            else:
+                formats.append({
+                    'url': video_url,
+                    'format_id': format_id,
+                    'preference': -1,
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info['titre'],
+            'description': clean_html(info['synopsis']),
+            'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
+            'duration': float_or_none(info.get('real_duration'), 1000) or parse_duration(info['duree']),
+            'timestamp': int_or_none(info['diffusion']['timestamp']),
+            'formats': formats,
+        }
 
 
 class PluzzIE(FranceTVBaseInfoExtractor):
 
 
 class PluzzIE(FranceTVBaseInfoExtractor):
@@ -43,30 +103,43 @@ class PluzzIE(FranceTVBaseInfoExtractor):
         webpage = self._download_webpage(url, title)
         video_id = self._search_regex(
             r'data-diffusion="(\d+)"', webpage, 'ID')
         webpage = self._download_webpage(url, title)
         video_id = self._search_regex(
             r'data-diffusion="(\d+)"', webpage, 'ID')
-        return self._extract_video(video_id)
+        return self._extract_video(video_id, 'Pluzz')
 
 
 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'francetvinfo.fr'
 
 
 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
     IE_NAME = 'francetvinfo.fr'
-    _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+)\.html'
+    _VALID_URL = r'https?://(?:www|mobile)\.francetvinfo\.fr/.*/(?P<title>.+)\.html'
 
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
         'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
-        'file': '84981923.mp4',
         'info_dict': {
         'info_dict': {
+            'id': '84981923',
+            'ext': 'flv',
             'title': 'Soir 3',
             'title': 'Soir 3',
+            'upload_date': '20130826',
+            'timestamp': 1377548400,
+        },
+    }, {
+        'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
+        'info_dict': {
+            'id': 'EV_20019',
+            'ext': 'mp4',
+            'title': 'Débat des candidats à la Commission européenne',
+            'description': 'Débat des candidats à la Commission européenne',
         },
         'params': {
         },
         'params': {
-            'skip_download': True,
+            'skip_download': 'HLS (reqires ffmpeg)'
         },
         },
-    }
+        'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         page_title = mobj.group('title')
         webpage = self._download_webpage(url, page_title)
 
     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=(\d+?)[@"]', webpage, 'video id')
-        return self._extract_video(video_id)
+        video_id, catalogue = self._search_regex(
+            r'id-video=([^@]+@[^"]+)', webpage, 'video id').split('@')
+        return self._extract_video(video_id, catalogue)
 
 
 class FranceTVIE(FranceTVBaseInfoExtractor):
 
 
 class FranceTVIE(FranceTVBaseInfoExtractor):
@@ -82,101 +155,88 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
         # france2
         {
             'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
         # france2
         {
             'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
-            'file': '75540104.mp4',
+            'md5': 'c03fc87cb85429ffd55df32b9fc05523',
             'info_dict': {
             'info_dict': {
-                'title': '13h15, le samedi...',
-                'description': 'md5:2e5b58ba7a2d3692b35c792be081a03d',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
+                'id': '109169362',
+                'ext': 'flv',
+                'title': '13h15, le dimanche...',
+                'description': 'md5:9a0932bb465f22d377a449be9d1a0ff7',
+                'upload_date': '20140914',
+                'timestamp': 1410693600,
             },
         },
         # france3
         {
             'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
             },
         },
         # france3
         {
             'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
+            'md5': '679bb8f8921f8623bd658fa2f8364da0',
             'info_dict': {
                 'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
             'info_dict': {
                 'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Le scandale du prix des médicaments',
                 'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
                 'title': 'Le scandale du prix des médicaments',
                 'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
+                'upload_date': '20131113',
+                'timestamp': 1384380000,
             },
         },
         # france4
         {
             'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
             },
         },
         # france4
         {
             'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
+            'md5': 'a182bf8d2c43d88d46ec48fbdd260c1c',
             'info_dict': {
                 'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
             'info_dict': {
                 'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Hero Corp Making of - Extrait 1',
                 'description': 'md5:c87d54871b1790679aec1197e73d650a',
                 'title': 'Hero Corp Making of - Extrait 1',
                 'description': 'md5:c87d54871b1790679aec1197e73d650a',
-            },
-            'params': {
-                # rtmp download
-                'skip_download': True,
+                'upload_date': '20131106',
+                'timestamp': 1383766500,
             },
         },
         # france5
         {
             'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
             },
         },
         # france5
         {
             'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
+            'md5': '78f0f4064f9074438e660785bbf2c5d9',
             'info_dict': {
             'info_dict': {
-                'id': '92837968',
-                'ext': 'mp4',
+                'id': '108961659',
+                'ext': 'flv',
                 'title': 'C à dire ?!',
                 'title': 'C à dire ?!',
-                'description': 'md5:fb1db1cbad784dcce7c7a7bd177c8e2f',
-            },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
+                'description': 'md5:1a4aeab476eb657bf57c4ff122129f81',
+                'upload_date': '20140915',
+                'timestamp': 1410795000,
             },
         },
         # franceo
         {
             'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
             },
         },
         # franceo
         {
             'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
+            'md5': '52f0bfe202848b15915a2f39aaa8981b',
             'info_dict': {
             'info_dict': {
-                'id': '92327925',
-                'ext': 'mp4',
-                'title': 'Infô-Afrique',
+                'id': '108634970',
+                'ext': 'flv',
+                'title': 'Infô Afrique',
                 'description': 'md5:ebf346da789428841bee0fd2a935ea55',
                 'description': 'md5:ebf346da789428841bee0fd2a935ea55',
+                'upload_date': '20140915',
+                'timestamp': 1410822000,
             },
             },
-            'params': {
-                # m3u8 download
-                'skip_download': True,
-            },
-            'skip': 'The id changes frequently',
         },
     ]
 
     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)
+        webpage = self._download_webpage(url, mobj.group('key') or mobj.group('id'))
+        video_id, catalogue = self._html_search_regex(
+            r'href="http://videos\.francetv\.fr/video/([^@]+@[^"]+)"',
+            webpage, 'video ID').split('@')
+        return self._extract_video(video_id, catalogue)
 
 
 class GenerationQuoiIE(InfoExtractor):
     IE_NAME = 'france2.fr:generation-quoi'
 
 
 class GenerationQuoiIE(InfoExtractor):
     IE_NAME = 'france2.fr:generation-quoi'
-    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
+    _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
 
     _TEST = {
         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
 
     _TEST = {
         'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
-        'file': 'k7FJX8VBcvvLmX4wA5Q.mp4',
         'info_dict': {
         'info_dict': {
+            'id': 'k7FJX8VBcvvLmX4wA5Q',
+            'ext': 'mp4',
             'title': 'Génération Quoi - Garde à Vous',
             'uploader': 'Génération Quoi',
         },
             'title': 'Génération Quoi - Garde à Vous',
             'uploader': 'Génération Quoi',
         },
@@ -184,40 +244,45 @@ class GenerationQuoiIE(InfoExtractor):
             # It uses Dailymotion
             'skip_download': True,
         },
             # It uses Dailymotion
             'skip_download': True,
         },
-        'skip': 'Only available from France',
     }
 
     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)
+        display_id = self._match_id(url)
+        info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
+        info_json = self._download_webpage(info_url, display_id)
         info = json.loads(info_json)
         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
         info = json.loads(info_json)
         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
-            ie='Dailymotion')
+                               ie='Dailymotion')
 
 
 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'"http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video id').split('@')
+
+        return self._extract_video(video_id, catalogue)