]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/orf.py
[cbs] extract smpte and vtt subtitles
[youtube-dl.git] / youtube_dl / extractor / orf.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import compat_str
8 from ..utils import (
9     determine_ext,
10     float_or_none,
11     HEADRequest,
12     int_or_none,
13     orderedSet,
14     remove_end,
15     strip_jsonp,
16     unescapeHTML,
17     unified_strdate,
18     url_or_none,
19 )
20
21
22 class ORFTVthekIE(InfoExtractor):
23     IE_NAME = 'orf:tvthek'
24     IE_DESC = 'ORF TVthek'
25     _VALID_URL = r'https?://tvthek\.orf\.at/(?:[^/]+/)+(?P<id>\d+)'
26
27     _TESTS = [{
28         'url': 'http://tvthek.orf.at/program/Aufgetischt/2745173/Aufgetischt-Mit-der-Steirischen-Tafelrunde/8891389',
29         'playlist': [{
30             'md5': '2942210346ed779588f428a92db88712',
31             'info_dict': {
32                 'id': '8896777',
33                 'ext': 'mp4',
34                 'title': 'Aufgetischt: Mit der Steirischen Tafelrunde',
35                 'description': 'md5:c1272f0245537812d4e36419c207b67d',
36                 'duration': 2668,
37                 'upload_date': '20141208',
38             },
39         }],
40         'skip': 'Blocked outside of Austria / Germany',
41     }, {
42         'url': 'http://tvthek.orf.at/topic/Im-Wandel-der-Zeit/8002126/Best-of-Ingrid-Thurnher/7982256',
43         'info_dict': {
44             'id': '7982259',
45             'ext': 'mp4',
46             'title': 'Best of Ingrid Thurnher',
47             'upload_date': '20140527',
48             'description': 'Viele Jahre war Ingrid Thurnher das "Gesicht" der ZIB 2. Vor ihrem Wechsel zur ZIB 2 im Jahr 1995 moderierte sie unter anderem "Land und Leute", "Österreich-Bild" und "Niederösterreich heute".',
49         },
50         'params': {
51             'skip_download': True,  # rtsp downloads
52         },
53         'skip': 'Blocked outside of Austria / Germany',
54     }, {
55         'url': 'http://tvthek.orf.at/topic/Fluechtlingskrise/10463081/Heimat-Fremde-Heimat/13879132/Senioren-betreuen-Migrantenkinder/13879141',
56         'only_matching': True,
57     }, {
58         'url': 'http://tvthek.orf.at/profile/Universum/35429',
59         'only_matching': True,
60     }]
61
62     def _real_extract(self, url):
63         playlist_id = self._match_id(url)
64         webpage = self._download_webpage(url, playlist_id)
65
66         data_jsb = self._parse_json(
67             self._search_regex(
68                 r'<div[^>]+class=(["\']).*?VideoPlaylist.*?\1[^>]+data-jsb=(["\'])(?P<json>.+?)\2',
69                 webpage, 'playlist', group='json'),
70             playlist_id, transform_source=unescapeHTML)['playlist']['videos']
71
72         entries = []
73         for sd in data_jsb:
74             video_id, title = sd.get('id'), sd.get('title')
75             if not video_id or not title:
76                 continue
77             video_id = compat_str(video_id)
78             formats = []
79             for fd in sd['sources']:
80                 src = url_or_none(fd.get('src'))
81                 if not src:
82                     continue
83                 format_id_list = []
84                 for key in ('delivery', 'quality', 'quality_string'):
85                     value = fd.get(key)
86                     if value:
87                         format_id_list.append(value)
88                 format_id = '-'.join(format_id_list)
89                 if determine_ext(fd['src']) == 'm3u8':
90                     formats.extend(self._extract_m3u8_formats(
91                         fd['src'], video_id, 'mp4', m3u8_id=format_id))
92                 elif determine_ext(fd['src']) == 'f4m':
93                     formats.extend(self._extract_f4m_formats(
94                         fd['src'], video_id, f4m_id=format_id))
95                 else:
96                     formats.append({
97                         'format_id': format_id,
98                         'url': src,
99                         'protocol': fd.get('protocol'),
100                     })
101
102             # Check for geoblocking.
103             # There is a property is_geoprotection, but that's always false
104             geo_str = sd.get('geoprotection_string')
105             if geo_str:
106                 try:
107                     http_url = next(
108                         f['url']
109                         for f in formats
110                         if re.match(r'^https?://.*\.mp4$', f['url']))
111                 except StopIteration:
112                     pass
113                 else:
114                     req = HEADRequest(http_url)
115                     self._request_webpage(
116                         req, video_id,
117                         note='Testing for geoblocking',
118                         errnote=((
119                             'This video seems to be blocked outside of %s. '
120                             'You may want to try the streaming-* formats.')
121                             % geo_str),
122                         fatal=False)
123
124             self._check_formats(formats, video_id)
125             self._sort_formats(formats)
126
127             subtitles = {}
128             for sub in sd.get('subtitles', []):
129                 sub_src = sub.get('src')
130                 if not sub_src:
131                     continue
132                 subtitles.setdefault(sub.get('lang', 'de-AT'), []).append({
133                     'url': sub_src,
134                 })
135
136             upload_date = unified_strdate(sd.get('created_date'))
137             entries.append({
138                 '_type': 'video',
139                 'id': video_id,
140                 'title': title,
141                 'formats': formats,
142                 'subtitles': subtitles,
143                 'description': sd.get('description'),
144                 'duration': int_or_none(sd.get('duration_in_seconds')),
145                 'upload_date': upload_date,
146                 'thumbnail': sd.get('image_full_url'),
147             })
148
149         return {
150             '_type': 'playlist',
151             'entries': entries,
152             'id': playlist_id,
153         }
154
155
156 class ORFRadioIE(InfoExtractor):
157     def _real_extract(self, url):
158         mobj = re.match(self._VALID_URL, url)
159         station = mobj.group('station')
160         show_date = mobj.group('date')
161         show_id = mobj.group('show')
162
163         if station == 'fm4':
164             show_id = '4%s' % show_id
165
166         data = self._download_json(
167             'http://audioapi.orf.at/%s/api/json/current/broadcast/%s/%s' % (station, show_id, show_date),
168             show_id
169         )
170
171         def extract_entry_dict(info, title, subtitle):
172             return {
173                 'id': info['loopStreamId'].replace('.mp3', ''),
174                 'url': 'http://loopstream01.apa.at/?channel=%s&id=%s' % (station, info['loopStreamId']),
175                 'title': title,
176                 'description': subtitle,
177                 'duration': (info['end'] - info['start']) / 1000,
178                 'timestamp': info['start'] / 1000,
179                 'ext': 'mp3',
180                 'series': data.get('programTitle')
181             }
182
183         entries = [extract_entry_dict(t, data['title'], data['subtitle']) for t in data['streams']]
184
185         return {
186             '_type': 'playlist',
187             'id': show_id,
188             'title': data['title'],
189             'description': data['subtitle'],
190             'entries': entries
191         }
192
193
194 class ORFFM4IE(ORFRadioIE):
195     IE_NAME = 'orf:fm4'
196     IE_DESC = 'radio FM4'
197     _VALID_URL = r'https?://(?P<station>fm4)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
198
199     _TEST = {
200         'url': 'http://fm4.orf.at/player/20170107/CC',
201         'md5': '2b0be47375432a7ef104453432a19212',
202         'info_dict': {
203             'id': '2017-01-07_2100_tl_54_7DaysSat18_31295',
204             'ext': 'mp3',
205             'title': 'Solid Steel Radioshow',
206             'description': 'Die Mixshow von Coldcut und Ninja Tune.',
207             'duration': 3599,
208             'timestamp': 1483819257,
209             'upload_date': '20170107',
210         },
211         'skip': 'Shows from ORF radios are only available for 7 days.'
212     }
213
214
215 class ORFOE1IE(ORFRadioIE):
216     IE_NAME = 'orf:oe1'
217     IE_DESC = 'Radio Ã–sterreich 1'
218     _VALID_URL = r'https?://(?P<station>oe1)\.orf\.at/player/(?P<date>[0-9]+)/(?P<show>\w+)'
219
220     _TEST = {
221         'url': 'http://oe1.orf.at/player/20170108/456544',
222         'md5': '34d8a6e67ea888293741c86a099b745b',
223         'info_dict': {
224             'id': '2017-01-08_0759_tl_51_7DaysSun6_256141',
225             'ext': 'mp3',
226             'title': 'Morgenjournal',
227             'duration': 609,
228             'timestamp': 1483858796,
229             'upload_date': '20170108',
230         },
231         'skip': 'Shows from ORF radios are only available for 7 days.'
232     }
233
234
235 class ORFIPTVIE(InfoExtractor):
236     IE_NAME = 'orf:iptv'
237     IE_DESC = 'iptv.ORF.at'
238     _VALID_URL = r'https?://iptv\.orf\.at/(?:#/)?stories/(?P<id>\d+)'
239
240     _TEST = {
241         'url': 'http://iptv.orf.at/stories/2275236/',
242         'md5': 'c8b22af4718a4b4af58342529453e3e5',
243         'info_dict': {
244             'id': '350612',
245             'ext': 'flv',
246             'title': 'Weitere Evakuierungen um Vulkan Calbuco',
247             'description': 'md5:d689c959bdbcf04efeddedbf2299d633',
248             'duration': 68.197,
249             'thumbnail': r're:^https?://.*\.jpg$',
250             'upload_date': '20150425',
251         },
252     }
253
254     def _real_extract(self, url):
255         story_id = self._match_id(url)
256
257         webpage = self._download_webpage(
258             'http://iptv.orf.at/stories/%s' % story_id, story_id)
259
260         video_id = self._search_regex(
261             r'data-video(?:id)?="(\d+)"', webpage, 'video id')
262
263         data = self._download_json(
264             'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
265             video_id)[0]
266
267         duration = float_or_none(data['duration'], 1000)
268
269         video = data['sources']['default']
270         load_balancer_url = video['loadBalancerUrl']
271         abr = int_or_none(video.get('audioBitrate'))
272         vbr = int_or_none(video.get('bitrate'))
273         fps = int_or_none(video.get('videoFps'))
274         width = int_or_none(video.get('videoWidth'))
275         height = int_or_none(video.get('videoHeight'))
276         thumbnail = video.get('preview')
277
278         rendition = self._download_json(
279             load_balancer_url, video_id, transform_source=strip_jsonp)
280
281         f = {
282             'abr': abr,
283             'vbr': vbr,
284             'fps': fps,
285             'width': width,
286             'height': height,
287         }
288
289         formats = []
290         for format_id, format_url in rendition['redirect'].items():
291             if format_id == 'rtmp':
292                 ff = f.copy()
293                 ff.update({
294                     'url': format_url,
295                     'format_id': format_id,
296                 })
297                 formats.append(ff)
298             elif determine_ext(format_url) == 'f4m':
299                 formats.extend(self._extract_f4m_formats(
300                     format_url, video_id, f4m_id=format_id))
301             elif determine_ext(format_url) == 'm3u8':
302                 formats.extend(self._extract_m3u8_formats(
303                     format_url, video_id, 'mp4', m3u8_id=format_id))
304             else:
305                 continue
306         self._sort_formats(formats)
307
308         title = remove_end(self._og_search_title(webpage), ' - iptv.ORF.at')
309         description = self._og_search_description(webpage)
310         upload_date = unified_strdate(self._html_search_meta(
311             'dc.date', webpage, 'upload date'))
312
313         return {
314             'id': video_id,
315             'title': title,
316             'description': description,
317             'duration': duration,
318             'thumbnail': thumbnail,
319             'upload_date': upload_date,
320             'formats': formats,
321         }
322
323
324 class ORFFM4StoryIE(InfoExtractor):
325     IE_NAME = 'orf:fm4:story'
326     IE_DESC = 'fm4.orf.at stories'
327     _VALID_URL = r'https?://fm4\.orf\.at/stories/(?P<id>\d+)'
328
329     _TEST = {
330         'url': 'http://fm4.orf.at/stories/2865738/',
331         'playlist': [{
332             'md5': 'e1c2c706c45c7b34cf478bbf409907ca',
333             'info_dict': {
334                 'id': '547792',
335                 'ext': 'flv',
336                 'title': 'Manu Delago und Inner Tongue live',
337                 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.',
338                 'duration': 1748.52,
339                 'thumbnail': r're:^https?://.*\.jpg$',
340                 'upload_date': '20170913',
341             },
342         }, {
343             'md5': 'c6dd2179731f86f4f55a7b49899d515f',
344             'info_dict': {
345                 'id': '547798',
346                 'ext': 'flv',
347                 'title': 'Manu Delago und Inner Tongue live (2)',
348                 'duration': 1504.08,
349                 'thumbnail': r're:^https?://.*\.jpg$',
350                 'upload_date': '20170913',
351                 'description': 'Manu Delago und Inner Tongue haben bei der FM4 Soundpark Session live alles gegeben. Hier gibt es Fotos und die gesamte Session als Video.',
352             },
353         }],
354     }
355
356     def _real_extract(self, url):
357         story_id = self._match_id(url)
358         webpage = self._download_webpage(url, story_id)
359
360         entries = []
361         all_ids = orderedSet(re.findall(r'data-video(?:id)?="(\d+)"', webpage))
362         for idx, video_id in enumerate(all_ids):
363             data = self._download_json(
364                 'http://bits.orf.at/filehandler/static-api/json/current/data.json?file=%s' % video_id,
365                 video_id)[0]
366
367             duration = float_or_none(data['duration'], 1000)
368
369             video = data['sources']['q8c']
370             load_balancer_url = video['loadBalancerUrl']
371             abr = int_or_none(video.get('audioBitrate'))
372             vbr = int_or_none(video.get('bitrate'))
373             fps = int_or_none(video.get('videoFps'))
374             width = int_or_none(video.get('videoWidth'))
375             height = int_or_none(video.get('videoHeight'))
376             thumbnail = video.get('preview')
377
378             rendition = self._download_json(
379                 load_balancer_url, video_id, transform_source=strip_jsonp)
380
381             f = {
382                 'abr': abr,
383                 'vbr': vbr,
384                 'fps': fps,
385                 'width': width,
386                 'height': height,
387             }
388
389             formats = []
390             for format_id, format_url in rendition['redirect'].items():
391                 if format_id == 'rtmp':
392                     ff = f.copy()
393                     ff.update({
394                         'url': format_url,
395                         'format_id': format_id,
396                     })
397                     formats.append(ff)
398                 elif determine_ext(format_url) == 'f4m':
399                     formats.extend(self._extract_f4m_formats(
400                         format_url, video_id, f4m_id=format_id))
401                 elif determine_ext(format_url) == 'm3u8':
402                     formats.extend(self._extract_m3u8_formats(
403                         format_url, video_id, 'mp4', m3u8_id=format_id))
404                 else:
405                     continue
406             self._sort_formats(formats)
407
408             title = remove_end(self._og_search_title(webpage), ' - fm4.ORF.at')
409             if idx >= 1:
410                 # Titles are duplicates, make them unique
411                 title += ' (' + str(idx + 1) + ')'
412             description = self._og_search_description(webpage)
413             upload_date = unified_strdate(self._html_search_meta(
414                 'dc.date', webpage, 'upload date'))
415
416             entries.append({
417                 'id': video_id,
418                 'title': title,
419                 'description': description,
420                 'duration': duration,
421                 'thumbnail': thumbnail,
422                 'upload_date': upload_date,
423                 'formats': formats,
424             })
425
426         return self.playlist_result(entries)