]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/livestream.py
bc7894bf13ed29963aa1dad7880cf8549be1ca77
[youtubedl] / youtube_dl / extractor / livestream.py
1 from __future__ import unicode_literals
2
3 import re
4 import itertools
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_str,
9 compat_urlparse,
10 )
11 from ..utils import (
12 find_xpath_attr,
13 xpath_attr,
14 xpath_with_ns,
15 xpath_text,
16 orderedSet,
17 update_url_query,
18 int_or_none,
19 float_or_none,
20 parse_iso8601,
21 determine_ext,
22 )
23
24
25 class LivestreamIE(InfoExtractor):
26 IE_NAME = 'livestream'
27 _VALID_URL = r'https?://(?:new\.)?livestream\.com/(?:accounts/(?P<account_id>\d+)|(?P<account_name>[^/]+))/(?:events/(?P<event_id>\d+)|(?P<event_name>[^/]+))(?:/videos/(?P<id>\d+))?'
28 _TESTS = [{
29 'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370',
30 'md5': '53274c76ba7754fb0e8d072716f2292b',
31 'info_dict': {
32 'id': '4719370',
33 'ext': 'mp4',
34 'title': 'Live from Webster Hall NYC',
35 'timestamp': 1350008072,
36 'upload_date': '20121012',
37 'duration': 5968.0,
38 'like_count': int,
39 'view_count': int,
40 'thumbnail': 're:^http://.*\.jpg$'
41 }
42 }, {
43 'url': 'http://new.livestream.com/tedx/cityenglish',
44 'info_dict': {
45 'title': 'TEDCity2.0 (English)',
46 'id': '2245590',
47 },
48 'playlist_mincount': 4,
49 }, {
50 'url': 'http://new.livestream.com/chess24/tatasteelchess',
51 'info_dict': {
52 'title': 'Tata Steel Chess',
53 'id': '3705884',
54 },
55 'playlist_mincount': 60,
56 }, {
57 'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640',
58 'only_matching': True,
59 }, {
60 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015',
61 'only_matching': True,
62 }]
63 _API_URL_TEMPLATE = 'http://livestream.com/api/accounts/%s/events/%s'
64
65 def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None):
66 base_ele = find_xpath_attr(
67 smil, self._xpath_ns('.//meta', namespace), 'name', 'httpBase')
68 base = base_ele.get('content') if base_ele is not None else 'http://livestreamvod-f.akamaihd.net/'
69
70 formats = []
71 video_nodes = smil.findall(self._xpath_ns('.//video', namespace))
72
73 for vn in video_nodes:
74 tbr = int_or_none(vn.attrib.get('system-bitrate'), 1000)
75 furl = (
76 update_url_query(compat_urlparse.urljoin(base, vn.attrib['src']), {
77 'v': '3.0.3',
78 'fp': 'WIN% 14,0,0,145',
79 }))
80 if 'clipBegin' in vn.attrib:
81 furl += '&ssek=' + vn.attrib['clipBegin']
82 formats.append({
83 'url': furl,
84 'format_id': 'smil_%d' % tbr,
85 'ext': 'flv',
86 'tbr': tbr,
87 'preference': -1000,
88 })
89 return formats
90
91 def _extract_video_info(self, video_data):
92 video_id = compat_str(video_data['id'])
93
94 FORMAT_KEYS = (
95 ('sd', 'progressive_url'),
96 ('hd', 'progressive_url_hd'),
97 )
98
99 formats = []
100 for format_id, key in FORMAT_KEYS:
101 video_url = video_data.get(key)
102 if video_url:
103 ext = determine_ext(video_url)
104 if ext == 'm3u8':
105 continue
106 bitrate = int_or_none(self._search_regex(
107 r'(\d+)\.%s' % ext, video_url, 'bitrate', default=None))
108 formats.append({
109 'url': video_url,
110 'format_id': format_id,
111 'tbr': bitrate,
112 'ext': ext,
113 })
114
115 smil_url = video_data.get('smil_url')
116 if smil_url:
117 formats.extend(self._extract_smil_formats(smil_url, video_id))
118
119 m3u8_url = video_data.get('m3u8_url')
120 if m3u8_url:
121 formats.extend(self._extract_m3u8_formats(
122 m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
123
124 f4m_url = video_data.get('f4m_url')
125 if f4m_url:
126 formats.extend(self._extract_f4m_formats(
127 f4m_url, video_id, f4m_id='hds', fatal=False))
128 self._sort_formats(formats)
129
130 comments = [{
131 'author_id': comment.get('author_id'),
132 'author': comment.get('author', {}).get('full_name'),
133 'id': comment.get('id'),
134 'text': comment['text'],
135 'timestamp': parse_iso8601(comment.get('created_at')),
136 } for comment in video_data.get('comments', {}).get('data', [])]
137
138 return {
139 'id': video_id,
140 'formats': formats,
141 'title': video_data['caption'],
142 'description': video_data.get('description'),
143 'thumbnail': video_data.get('thumbnail_url'),
144 'duration': float_or_none(video_data.get('duration'), 1000),
145 'timestamp': parse_iso8601(video_data.get('publish_at')),
146 'like_count': video_data.get('likes', {}).get('total'),
147 'comment_count': video_data.get('comments', {}).get('total'),
148 'view_count': video_data.get('views'),
149 'comments': comments,
150 }
151
152 def _extract_stream_info(self, stream_info):
153 broadcast_id = compat_str(stream_info['broadcast_id'])
154 is_live = stream_info.get('is_live')
155
156 formats = []
157 smil_url = stream_info.get('play_url')
158 if smil_url:
159 formats.extend(self._extract_smil_formats(smil_url, broadcast_id))
160
161 entry_protocol = 'm3u8' if is_live else 'm3u8_native'
162 m3u8_url = stream_info.get('m3u8_url')
163 if m3u8_url:
164 formats.extend(self._extract_m3u8_formats(
165 m3u8_url, broadcast_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False))
166
167 rtsp_url = stream_info.get('rtsp_url')
168 if rtsp_url:
169 formats.append({
170 'url': rtsp_url,
171 'format_id': 'rtsp',
172 })
173 self._sort_formats(formats)
174
175 return {
176 'id': broadcast_id,
177 'formats': formats,
178 'title': self._live_title(stream_info['stream_title']) if is_live else stream_info['stream_title'],
179 'thumbnail': stream_info.get('thumbnail_url'),
180 'is_live': is_live,
181 }
182
183 def _extract_event(self, event_data):
184 event_id = compat_str(event_data['id'])
185 account_id = compat_str(event_data['owner_account_id'])
186 feed_root_url = self._API_URL_TEMPLATE % (account_id, event_id) + '/feed.json'
187
188 stream_info = event_data.get('stream_info')
189 if stream_info:
190 return self._extract_stream_info(stream_info)
191
192 last_video = None
193 entries = []
194 for i in itertools.count(1):
195 if last_video is None:
196 info_url = feed_root_url
197 else:
198 info_url = '{root}?&id={id}&newer=-1&type=video'.format(
199 root=feed_root_url, id=last_video)
200 videos_info = self._download_json(
201 info_url, event_id, 'Downloading page {0}'.format(i))['data']
202 videos_info = [v['data'] for v in videos_info if v['type'] == 'video']
203 if not videos_info:
204 break
205 for v in videos_info:
206 v_id = compat_str(v['id'])
207 entries.append(self.url_result(
208 'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v_id),
209 'Livestream', v_id, v.get('caption')))
210 last_video = videos_info[-1]['id']
211 return self.playlist_result(entries, event_id, event_data['full_name'])
212
213 def _real_extract(self, url):
214 mobj = re.match(self._VALID_URL, url)
215 video_id = mobj.group('id')
216 event = mobj.group('event_id') or mobj.group('event_name')
217 account = mobj.group('account_id') or mobj.group('account_name')
218 api_url = self._API_URL_TEMPLATE % (account, event)
219 if video_id:
220 video_data = self._download_json(
221 api_url + '/videos/%s' % video_id, video_id)
222 return self._extract_video_info(video_data)
223 else:
224 event_data = self._download_json(api_url, video_id)
225 return self._extract_event(event_data)
226
227
228 # The original version of Livestream uses a different system
229 class LivestreamOriginalIE(InfoExtractor):
230 IE_NAME = 'livestream:original'
231 _VALID_URL = r'''(?x)https?://original\.livestream\.com/
232 (?P<user>[^/\?#]+)(?:/(?P<type>video|folder)
233 (?:(?:\?.*?Id=|/)(?P<id>.*?)(&|$))?)?
234 '''
235 _TESTS = [{
236 'url': 'http://original.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
237 'info_dict': {
238 'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb',
239 'ext': 'mp4',
240 'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital',
241 'duration': 771.301,
242 'view_count': int,
243 },
244 }, {
245 'url': 'https://original.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3',
246 'info_dict': {
247 'id': 'a07bf706-d0e4-4e75-a747-b021d84f2fd3',
248 },
249 'playlist_mincount': 4,
250 }, {
251 # live stream
252 'url': 'http://original.livestream.com/znsbahamas',
253 'only_matching': True,
254 }]
255
256 def _extract_video_info(self, user, video_id):
257 api_url = 'http://x%sx.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id=%s' % (user, video_id)
258 info = self._download_xml(api_url, video_id)
259
260 item = info.find('channel').find('item')
261 title = xpath_text(item, 'title')
262 media_ns = {'media': 'http://search.yahoo.com/mrss'}
263 thumbnail_url = xpath_attr(
264 item, xpath_with_ns('media:thumbnail', media_ns), 'url')
265 duration = float_or_none(xpath_attr(
266 item, xpath_with_ns('media:content', media_ns), 'duration'))
267 ls_ns = {'ls': 'http://api.channel.livestream.com/2.0'}
268 view_count = int_or_none(xpath_text(
269 item, xpath_with_ns('ls:viewsCount', ls_ns)))
270
271 return {
272 'id': video_id,
273 'title': title,
274 'thumbnail': thumbnail_url,
275 'duration': duration,
276 'view_count': view_count,
277 }
278
279 def _extract_video_formats(self, video_data, video_id, entry_protocol):
280 formats = []
281
282 progressive_url = video_data.get('progressiveUrl')
283 if progressive_url:
284 formats.append({
285 'url': progressive_url,
286 'format_id': 'http',
287 })
288
289 m3u8_url = video_data.get('httpUrl')
290 if m3u8_url:
291 formats.extend(self._extract_m3u8_formats(
292 m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False))
293
294 rtsp_url = video_data.get('rtspUrl')
295 if rtsp_url:
296 formats.append({
297 'url': rtsp_url,
298 'format_id': 'rtsp',
299 })
300
301 self._sort_formats(formats)
302 return formats
303
304 def _extract_folder(self, url, folder_id):
305 webpage = self._download_webpage(url, folder_id)
306 paths = orderedSet(re.findall(
307 r'''(?x)(?:
308 <li\s+class="folder">\s*<a\s+href="|
309 <a\s+href="(?=https?://livestre\.am/)
310 )([^"]+)"''', webpage))
311
312 entries = [{
313 '_type': 'url',
314 'url': compat_urlparse.urljoin(url, p),
315 } for p in paths]
316
317 return self.playlist_result(entries, folder_id)
318
319 def _real_extract(self, url):
320 mobj = re.match(self._VALID_URL, url)
321 user = mobj.group('user')
322 url_type = mobj.group('type')
323 content_id = mobj.group('id')
324 if url_type == 'folder':
325 return self._extract_folder(url, content_id)
326 else:
327 # this url is used on mobile devices
328 stream_url = 'http://x%sx.api.channel.livestream.com/3.0/getstream.json' % user
329 info = {}
330 if content_id:
331 stream_url += '?id=%s' % content_id
332 info = self._extract_video_info(user, content_id)
333 else:
334 content_id = user
335 webpage = self._download_webpage(url, content_id)
336 info = {
337 'title': self._og_search_title(webpage),
338 'description': self._og_search_description(webpage),
339 'thumbnail': self._search_regex(r'channelLogo.src\s*=\s*"([^"]+)"', webpage, 'thumbnail', None),
340 }
341 video_data = self._download_json(stream_url, content_id)
342 is_live = video_data.get('isLive')
343 entry_protocol = 'm3u8' if is_live else 'm3u8_native'
344 info.update({
345 'id': content_id,
346 'title': self._live_title(info['title']) if is_live else info['title'],
347 'formats': self._extract_video_formats(video_data, content_id, entry_protocol),
348 'is_live': is_live,
349 })
350 return info
351
352
353 # The server doesn't support HEAD request, the generic extractor can't detect
354 # the redirection
355 class LivestreamShortenerIE(InfoExtractor):
356 IE_NAME = 'livestream:shortener'
357 IE_DESC = False # Do not list
358 _VALID_URL = r'https?://livestre\.am/(?P<id>.+)'
359
360 def _real_extract(self, url):
361 mobj = re.match(self._VALID_URL, url)
362 id = mobj.group('id')
363 webpage = self._download_webpage(url, id)
364
365 return {
366 '_type': 'url',
367 'url': self._og_search_url(webpage),
368 }