]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/francetv.py
546d5caa093c456ce68869b711497a0469836fef
[youtubedl] / youtube_dl / extractor / francetv.py
1 # coding: utf-8
2
3 from __future__ import unicode_literals
4
5 import re
6 import json
7
8 from .common import InfoExtractor
9 from ..compat import compat_urlparse
10 from ..utils import (
11 clean_html,
12 ExtractorError,
13 int_or_none,
14 parse_duration,
15 determine_ext,
16 )
17 from .dailymotion import (
18 DailymotionIE,
19 DailymotionCloudIE,
20 )
21
22
23 class FranceTVBaseInfoExtractor(InfoExtractor):
24 def _extract_video(self, video_id, catalogue=None):
25 info = self._download_json(
26 'https://sivideo.webservices.francetelevisions.fr/tools/getInfosOeuvre/v2/',
27 video_id, 'Downloading video JSON', query={
28 'idDiffusion': video_id,
29 'catalogue': catalogue or '',
30 })
31
32 if info.get('status') == 'NOK':
33 raise ExtractorError(
34 '%s returned error: %s' % (self.IE_NAME, info['message']), expected=True)
35 allowed_countries = info['videos'][0].get('geoblocage')
36 if allowed_countries:
37 georestricted = True
38 geo_info = self._download_json(
39 'http://geo.francetv.fr/ws/edgescape.json', video_id,
40 'Downloading geo restriction info')
41 country = geo_info['reponse']['geo_info']['country_code']
42 if country not in allowed_countries:
43 raise ExtractorError(
44 'The video is not available from your location',
45 expected=True)
46 else:
47 georestricted = False
48
49 formats = []
50 for video in info['videos']:
51 if video['statut'] != 'ONLINE':
52 continue
53 video_url = video['url']
54 if not video_url:
55 continue
56 format_id = video['format']
57 ext = determine_ext(video_url)
58 if ext == 'f4m':
59 if georestricted:
60 # See https://github.com/rg3/youtube-dl/issues/3963
61 # m3u8 urls work fine
62 continue
63 f4m_url = self._download_webpage(
64 'http://hdfauth.francetv.fr/esi/TA?url=%s' % video_url,
65 video_id, 'Downloading f4m manifest token', fatal=False)
66 if f4m_url:
67 formats.extend(self._extract_f4m_formats(
68 f4m_url + '&hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
69 video_id, f4m_id=format_id, fatal=False))
70 elif ext == 'm3u8':
71 formats.extend(self._extract_m3u8_formats(
72 video_url, video_id, 'mp4', entry_protocol='m3u8_native',
73 m3u8_id=format_id, fatal=False))
74 elif video_url.startswith('rtmp'):
75 formats.append({
76 'url': video_url,
77 'format_id': 'rtmp-%s' % format_id,
78 'ext': 'flv',
79 })
80 else:
81 if self._is_valid_url(video_url, video_id, format_id):
82 formats.append({
83 'url': video_url,
84 'format_id': format_id,
85 })
86 self._sort_formats(formats)
87
88 title = info['titre']
89 subtitle = info.get('sous_titre')
90 if subtitle:
91 title += ' - %s' % subtitle
92 title = title.strip()
93
94 subtitles = {}
95 subtitles_list = [{
96 'url': subformat['url'],
97 'ext': subformat.get('format'),
98 } for subformat in info.get('subtitles', []) if subformat.get('url')]
99 if subtitles_list:
100 subtitles['fr'] = subtitles_list
101
102 return {
103 'id': video_id,
104 'title': title,
105 'description': clean_html(info['synopsis']),
106 'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', info['image']),
107 'duration': int_or_none(info.get('real_duration')) or parse_duration(info['duree']),
108 'timestamp': int_or_none(info['diffusion']['timestamp']),
109 'formats': formats,
110 'subtitles': subtitles,
111 }
112
113
114 class FranceTVIE(FranceTVBaseInfoExtractor):
115 _VALID_URL = r'https?://(?:(?:www\.)?france\.tv|mobile\.france\.tv)/(?:[^/]+/)+(?P<id>[^/]+)\.html'
116
117 _TESTS = [{
118 'url': 'https://www.france.tv/france-2/13h15-le-dimanche/140921-les-mysteres-de-jesus.html',
119 'info_dict': {
120 'id': '157550144',
121 'ext': 'mp4',
122 'title': '13h15, le dimanche... - Les mystères de Jésus',
123 'description': 'md5:75efe8d4c0a8205e5904498ffe1e1a42',
124 'timestamp': 1494156300,
125 'upload_date': '20170507',
126 },
127 'params': {
128 # m3u8 downloads
129 'skip_download': True,
130 },
131 }, {
132 # france3
133 'url': 'https://www.france.tv/france-3/des-chiffres-et-des-lettres/139063-emission-du-mardi-9-mai-2017.html',
134 'only_matching': True,
135 }, {
136 # france4
137 'url': 'https://www.france.tv/france-4/hero-corp/saison-1/134151-apres-le-calme.html',
138 'only_matching': True,
139 }, {
140 # france5
141 'url': 'https://www.france.tv/france-5/c-a-dire/saison-10/137013-c-a-dire.html',
142 'only_matching': True,
143 }, {
144 # franceo
145 'url': 'https://www.france.tv/france-o/archipels/132249-mon-ancetre-l-esclave.html',
146 'only_matching': True,
147 }, {
148 # france2 live
149 'url': 'https://www.france.tv/france-2/direct.html',
150 'only_matching': True,
151 }, {
152 'url': 'https://www.france.tv/documentaires/histoire/136517-argentine-les-500-bebes-voles-de-la-dictature.html',
153 'only_matching': True,
154 }, {
155 'url': 'https://www.france.tv/jeux-et-divertissements/divertissements/133965-le-web-contre-attaque.html',
156 'only_matching': True,
157 }, {
158 'url': 'https://mobile.france.tv/france-5/c-dans-l-air/137347-emission-du-vendredi-12-mai-2017.html',
159 'only_matching': True,
160 }]
161
162 def _real_extract(self, url):
163 display_id = self._match_id(url)
164
165 webpage = self._download_webpage(url, display_id)
166
167 catalogue = None
168 video_id = self._search_regex(
169 r'data-main-video=(["\'])(?P<id>(?:(?!\1).)+)\1',
170 webpage, 'video id', default=None, group='id')
171
172 if not video_id:
173 video_id, catalogue = self._html_search_regex(
174 r'(?:href=|player\.setVideo\(\s*)"http://videos?\.francetv\.fr/video/([^@]+@[^"]+)"',
175 webpage, 'video ID').split('@')
176 return self._extract_video(video_id, catalogue)
177
178
179 class FranceTVEmbedIE(FranceTVBaseInfoExtractor):
180 _VALID_URL = r'https?://embed\.francetv\.fr/*\?.*?\bue=(?P<id>[^&]+)'
181
182 _TEST = {
183 'url': 'http://embed.francetv.fr/?ue=7fd581a2ccf59d2fc5719c5c13cf6961',
184 'info_dict': {
185 'id': 'NI_983319',
186 'ext': 'mp4',
187 'title': 'Le Pen Reims',
188 'upload_date': '20170505',
189 'timestamp': 1493981780,
190 'duration': 16,
191 },
192 }
193
194 def _real_extract(self, url):
195 video_id = self._match_id(url)
196
197 video = self._download_json(
198 'http://api-embed.webservices.francetelevisions.fr/key/%s' % video_id,
199 video_id)
200
201 return self._extract_video(video['video_id'], video.get('catalog'))
202
203
204 class FranceTVInfoIE(FranceTVBaseInfoExtractor):
205 IE_NAME = 'francetvinfo.fr'
206 _VALID_URL = r'https?://(?:www|mobile|france3-regions)\.francetvinfo\.fr/(?:[^/]+/)*(?P<title>[^/?#&.]+)'
207
208 _TESTS = [{
209 'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
210 'info_dict': {
211 'id': '84981923',
212 'ext': 'mp4',
213 'title': 'Soir 3',
214 'upload_date': '20130826',
215 'timestamp': 1377548400,
216 'subtitles': {
217 'fr': 'mincount:2',
218 },
219 },
220 'params': {
221 # m3u8 downloads
222 'skip_download': True,
223 },
224 }, {
225 'url': 'http://www.francetvinfo.fr/elections/europeennes/direct-europeennes-regardez-le-debat-entre-les-candidats-a-la-presidence-de-la-commission_600639.html',
226 'info_dict': {
227 'id': 'EV_20019',
228 'ext': 'mp4',
229 'title': 'Débat des candidats à la Commission européenne',
230 'description': 'Débat des candidats à la Commission européenne',
231 },
232 'params': {
233 'skip_download': 'HLS (reqires ffmpeg)'
234 },
235 'skip': 'Ce direct est terminé et sera disponible en rattrapage dans quelques minutes.',
236 }, {
237 'url': 'http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html',
238 'md5': 'f485bda6e185e7d15dbc69b72bae993e',
239 'info_dict': {
240 'id': 'NI_173343',
241 'ext': 'mp4',
242 'title': 'Les entreprises familiales : le secret de la réussite',
243 'thumbnail': r're:^https?://.*\.jpe?g$',
244 'timestamp': 1433273139,
245 'upload_date': '20150602',
246 },
247 'params': {
248 # m3u8 downloads
249 'skip_download': True,
250 },
251 }, {
252 'url': 'http://france3-regions.francetvinfo.fr/bretagne/cotes-d-armor/thalassa-echappee-breizh-ce-venredi-dans-les-cotes-d-armor-954961.html',
253 'md5': 'f485bda6e185e7d15dbc69b72bae993e',
254 'info_dict': {
255 'id': 'NI_657393',
256 'ext': 'mp4',
257 'title': 'Olivier Monthus, réalisateur de "Bretagne, le choix de l’Armor"',
258 'description': 'md5:a3264114c9d29aeca11ced113c37b16c',
259 'thumbnail': r're:^https?://.*\.jpe?g$',
260 'timestamp': 1458300695,
261 'upload_date': '20160318',
262 },
263 'params': {
264 'skip_download': True,
265 },
266 }, {
267 # Dailymotion embed
268 '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',
269 'md5': 'ee7f1828f25a648addc90cb2687b1f12',
270 'info_dict': {
271 'id': 'x4iiko0',
272 'ext': 'mp4',
273 'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
274 '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',
275 'timestamp': 1467011958,
276 'upload_date': '20160627',
277 'uploader': 'France Inter',
278 'uploader_id': 'x2q2ez',
279 },
280 'add_ie': ['Dailymotion'],
281 }, {
282 'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin',
283 'only_matching': True,
284 }]
285
286 def _real_extract(self, url):
287 mobj = re.match(self._VALID_URL, url)
288 page_title = mobj.group('title')
289 webpage = self._download_webpage(url, page_title)
290
291 dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
292 if dmcloud_url:
293 return self.url_result(dmcloud_url, DailymotionCloudIE.ie_key())
294
295 dailymotion_urls = DailymotionIE._extract_urls(webpage)
296 if dailymotion_urls:
297 return self.playlist_result([
298 self.url_result(dailymotion_url, DailymotionIE.ie_key())
299 for dailymotion_url in dailymotion_urls])
300
301 video_id, catalogue = self._search_regex(
302 (r'id-video=([^@]+@[^"]+)',
303 r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'),
304 webpage, 'video id').split('@')
305 return self._extract_video(video_id, catalogue)
306
307
308 class GenerationQuoiIE(InfoExtractor):
309 IE_NAME = 'france2.fr:generation-quoi'
310 _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<id>[^/?#]+)'
311
312 _TEST = {
313 'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
314 'info_dict': {
315 'id': 'k7FJX8VBcvvLmX4wA5Q',
316 'ext': 'mp4',
317 'title': 'Génération Quoi - Garde à Vous',
318 'uploader': 'Génération Quoi',
319 },
320 'params': {
321 # It uses Dailymotion
322 'skip_download': True,
323 },
324 }
325
326 def _real_extract(self, url):
327 display_id = self._match_id(url)
328 info_url = compat_urlparse.urljoin(url, '/medias/video/%s.json' % display_id)
329 info_json = self._download_webpage(info_url, display_id)
330 info = json.loads(info_json)
331 return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
332 ie='Dailymotion')
333
334
335 class CultureboxIE(FranceTVBaseInfoExtractor):
336 IE_NAME = 'culturebox.francetvinfo.fr'
337 _VALID_URL = r'https?://(?:m\.)?culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
338
339 _TEST = {
340 'url': 'http://culturebox.francetvinfo.fr/live/musique/musique-classique/le-livre-vermeil-de-montserrat-a-la-cathedrale-delne-214511',
341 'md5': '9b88dc156781c4dbebd4c3e066e0b1d6',
342 'info_dict': {
343 'id': 'EV_50111',
344 'ext': 'flv',
345 'title': "Le Livre Vermeil de Montserrat à la Cathédrale d'Elne",
346 'description': 'md5:f8a4ad202e8fe533e2c493cc12e739d9',
347 'upload_date': '20150320',
348 'timestamp': 1426892400,
349 'duration': 2760.9,
350 },
351 }
352
353 def _real_extract(self, url):
354 mobj = re.match(self._VALID_URL, url)
355 name = mobj.group('name')
356
357 webpage = self._download_webpage(url, name)
358
359 if ">Ce live n'est plus disponible en replay<" in webpage:
360 raise ExtractorError('Video %s is not available' % name, expected=True)
361
362 video_id, catalogue = self._search_regex(
363 r'"http://videos\.francetv\.fr/video/([^@]+@[^"]+)"', webpage, 'video id').split('@')
364
365 return self._extract_video(video_id, catalogue)