]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/prosiebensat1.py
Imported Upstream version 2014.06.19
[youtubedl] / youtube_dl / extractor / prosiebensat1.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from hashlib import sha1
7 from .common import InfoExtractor
8 from ..utils import (
9 compat_urllib_parse,
10 unified_strdate,
11 )
12
13
14 class ProSiebenSat1IE(InfoExtractor):
15 IE_NAME = 'prosiebensat1'
16 IE_DESC = 'ProSiebenSat.1 Digital'
17 _VALID_URL = r'https?://(?:www\.)?(?:(?:prosieben|prosiebenmaxx|sixx|sat1|kabeleins|ran|the-voice-of-germany)\.de|fem\.com)/(?P<id>.+)'
18
19 _TESTS = [
20 {
21 'url': 'http://www.prosieben.de/tv/circus-halligalli/videos/218-staffel-2-episode-18-jahresrueckblick-ganze-folge',
22 'info_dict': {
23 'id': '2104602',
24 'ext': 'mp4',
25 'title': 'Staffel 2, Episode 18 - Jahresrückblick',
26 'description': 'md5:8733c81b702ea472e069bc48bb658fc1',
27 'upload_date': '20131231',
28 'duration': 5845.04,
29 },
30 'params': {
31 # rtmp download
32 'skip_download': True,
33 },
34 },
35 {
36 'url': 'http://www.prosieben.de/videokatalog/Gesellschaft/Leben/Trends/video-Lady-Umstyling-f%C3%BCr-Audrina-Rebekka-Audrina-Fergen-billig-aussehen-Battal-Modica-700544.html',
37 'info_dict': {
38 'id': '2570327',
39 'ext': 'mp4',
40 'title': 'Lady-Umstyling für Audrina',
41 'description': 'md5:4c16d0c17a3461a0d43ea4084e96319d',
42 'upload_date': '20131014',
43 'duration': 606.76,
44 },
45 'params': {
46 # rtmp download
47 'skip_download': True,
48 },
49 'skip': 'Seems to be broken',
50 },
51 {
52 'url': 'http://www.prosiebenmaxx.de/tv/experience/video/144-countdown-fuer-die-autowerkstatt-ganze-folge',
53 'info_dict': {
54 'id': '2429369',
55 'ext': 'mp4',
56 'title': 'Countdown für die Autowerkstatt',
57 'description': 'md5:809fc051a457b5d8666013bc40698817',
58 'upload_date': '20140223',
59 'duration': 2595.04,
60 },
61 'params': {
62 # rtmp download
63 'skip_download': True,
64 },
65 },
66 {
67 'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
68 'info_dict': {
69 'id': '2904997',
70 'ext': 'mp4',
71 'title': 'Sexy laufen in Ugg Boots',
72 'description': 'md5:edf42b8bd5bc4e5da4db4222c5acb7d6',
73 'upload_date': '20140122',
74 'duration': 245.32,
75 },
76 'params': {
77 # rtmp download
78 'skip_download': True,
79 },
80 },
81 {
82 'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
83 'info_dict': {
84 'id': '2906572',
85 'ext': 'mp4',
86 'title': 'Im Interview: Kai Wiesinger',
87 'description': 'md5:e4e5370652ec63b95023e914190b4eb9',
88 'upload_date': '20140225',
89 'duration': 522.56,
90 },
91 'params': {
92 # rtmp download
93 'skip_download': True,
94 },
95 },
96 {
97 'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
98 'info_dict': {
99 'id': '2992323',
100 'ext': 'mp4',
101 'title': 'Jagd auf Fertigkost im Elsthal - Teil 2',
102 'description': 'md5:2669cde3febe9bce13904f701e774eb6',
103 'upload_date': '20140225',
104 'duration': 2410.44,
105 },
106 'params': {
107 # rtmp download
108 'skip_download': True,
109 },
110 },
111 {
112 'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
113 'info_dict': {
114 'id': '3004256',
115 'ext': 'mp4',
116 'title': 'Schalke: Tönnies möchte Raul zurück',
117 'description': 'md5:4b5b271d9bcde223b54390754c8ece3f',
118 'upload_date': '20140226',
119 'duration': 228.96,
120 },
121 'params': {
122 # rtmp download
123 'skip_download': True,
124 },
125 },
126 {
127 'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
128 'info_dict': {
129 'id': '2572814',
130 'ext': 'mp4',
131 'title': 'Andreas Kümmert: Rocket Man',
132 'description': 'md5:6ddb02b0781c6adf778afea606652e38',
133 'upload_date': '20131017',
134 'duration': 469.88,
135 },
136 'params': {
137 # rtmp download
138 'skip_download': True,
139 },
140 },
141 {
142 'url': 'http://www.fem.com/wellness/videos/wellness-video-clip-kurztripps-zum-valentinstag.html',
143 'info_dict': {
144 'id': '2156342',
145 'ext': 'mp4',
146 'title': 'Kurztrips zum Valentinstag',
147 'description': 'md5:8ba6301e70351ae0bedf8da00f7ba528',
148 'upload_date': '20130206',
149 'duration': 307.24,
150 },
151 'params': {
152 # rtmp download
153 'skip_download': True,
154 },
155 },
156 ]
157
158 _CLIPID_REGEXES = [
159 r'"clip_id"\s*:\s+"(\d+)"',
160 r'clipid: "(\d+)"',
161 r'clip[iI]d=(\d+)',
162 ]
163 _TITLE_REGEXES = [
164 r'<h2 class="subtitle" itemprop="name">\s*(.+?)</h2>',
165 r'<header class="clearfix">\s*<h3>(.+?)</h3>',
166 r'<!-- start video -->\s*<h1>(.+?)</h1>',
167 r'<h1 class="att-name">\s*(.+?)</h1>',
168 ]
169 _DESCRIPTION_REGEXES = [
170 r'<p itemprop="description">\s*(.+?)</p>',
171 r'<div class="videoDecription">\s*<p><strong>Beschreibung</strong>: (.+?)</p>',
172 r'<div class="g-plusone" data-size="medium"></div>\s*</div>\s*</header>\s*(.+?)\s*<footer>',
173 r'<p class="att-description">\s*(.+?)\s*</p>',
174 ]
175 _UPLOAD_DATE_REGEXES = [
176 r'<meta property="og:published_time" content="(.+?)">',
177 r'<span>\s*(\d{2}\.\d{2}\.\d{4} \d{2}:\d{2}) \|\s*<span itemprop="duration"',
178 r'<footer>\s*(\d{2}\.\d{2}\.\d{4}) \d{2}:\d{2} Uhr',
179 r'<span style="padding-left: 4px;line-height:20px; color:#404040">(\d{2}\.\d{2}\.\d{4})</span>',
180 r'(\d{2}\.\d{2}\.\d{4}) \| \d{2}:\d{2} Min<br/>',
181 ]
182
183 def _real_extract(self, url):
184 mobj = re.match(self._VALID_URL, url)
185 video_id = mobj.group('id')
186
187 page = self._download_webpage(url, video_id, 'Downloading page')
188
189 clip_id = self._html_search_regex(self._CLIPID_REGEXES, page, 'clip id')
190
191 access_token = 'testclient'
192 client_name = 'kolibri-1.2.5'
193 client_location = url
194
195 videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse.urlencode({
196 'access_token': access_token,
197 'client_location': client_location,
198 'client_name': client_name,
199 'ids': clip_id,
200 })
201
202 videos = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')
203
204 duration = float(videos[0]['duration'])
205 source_ids = [source['id'] for source in videos[0]['sources']]
206 source_ids_str = ','.join(map(str, source_ids))
207
208 g = '01!8d8F_)r9]4s[qeuXfP%'
209
210 client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
211 .encode('utf-8')).hexdigest()
212
213 sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse.urlencode({
214 'access_token': access_token,
215 'client_id': client_id,
216 'client_location': client_location,
217 'client_name': client_name,
218 }))
219
220 sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
221 server_id = sources['server_id']
222
223 client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
224 client_location, source_ids_str, g, client_name])
225 .encode('utf-8')).hexdigest()
226
227 url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse.urlencode({
228 'access_token': access_token,
229 'client_id': client_id,
230 'client_location': client_location,
231 'client_name': client_name,
232 'server_id': server_id,
233 'source_ids': source_ids_str,
234 }))
235
236 urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
237
238 title = self._html_search_regex(self._TITLE_REGEXES, page, 'title')
239 description = self._html_search_regex(self._DESCRIPTION_REGEXES, page, 'description', fatal=False)
240 thumbnail = self._og_search_thumbnail(page)
241
242 upload_date = unified_strdate(self._html_search_regex(
243 self._UPLOAD_DATE_REGEXES, page, 'upload date', fatal=False))
244
245 formats = []
246
247 urls_sources = urls['sources']
248 if isinstance(urls_sources, dict):
249 urls_sources = urls_sources.values()
250
251 def fix_bitrate(bitrate):
252 return bitrate / 1000 if bitrate % 1000 == 0 else bitrate
253
254 for source in urls_sources:
255 protocol = source['protocol']
256 if protocol == 'rtmp' or protocol == 'rtmpe':
257 mobj = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', source['url'])
258 if not mobj:
259 continue
260 formats.append({
261 'url': mobj.group('url'),
262 'app': mobj.group('app'),
263 'play_path': mobj.group('playpath'),
264 'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
265 'page_url': 'http://www.prosieben.de',
266 'vbr': fix_bitrate(source['bitrate']),
267 'ext': 'mp4',
268 'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
269 })
270 else:
271 formats.append({
272 'url': source['url'],
273 'vbr': fix_bitrate(source['bitrate']),
274 })
275
276 self._sort_formats(formats)
277
278 return {
279 'id': clip_id,
280 'title': title,
281 'description': description,
282 'thumbnail': thumbnail,
283 'upload_date': upload_date,
284 'duration': duration,
285 'formats': formats,
286 }