]>
Commit | Line | Data |
---|---|---|
c5469e04 S |
1 | from __future__ import unicode_literals |
2 | ||
b4444d5c | 3 | import re |
082b1155 | 4 | import itertools |
b4444d5c JMF |
5 | |
6 | from .common import InfoExtractor | |
1cc79574 | 7 | from ..compat import ( |
cbf915f3 | 8 | compat_str, |
b00ca882 | 9 | compat_urlparse, |
1cc79574 PH |
10 | ) |
11 | from ..utils import ( | |
cbf915f3 | 12 | find_xpath_attr, |
64ccbf18 | 13 | xpath_attr, |
cbf915f3 | 14 | xpath_with_ns, |
64ccbf18 | 15 | xpath_text, |
16 | orderedSet, | |
17 | int_or_none, | |
18 | float_or_none, | |
19 | parse_iso8601, | |
20 | determine_ext, | |
b00ca882 | 21 | ) |
b4444d5c JMF |
22 | |
23 | ||
24 | class LivestreamIE(InfoExtractor): | |
c5469e04 | 25 | IE_NAME = 'livestream' |
64ccbf18 | 26 | _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+))?' |
22a6f150 | 27 | _TESTS = [{ |
c5469e04 S |
28 | 'url': 'http://new.livestream.com/CoheedandCambria/WebsterHall/videos/4719370', |
29 | 'md5': '53274c76ba7754fb0e8d072716f2292b', | |
30 | 'info_dict': { | |
31 | 'id': '4719370', | |
32 | 'ext': 'mp4', | |
33 | 'title': 'Live from Webster Hall NYC', | |
64ccbf18 | 34 | 'timestamp': 1350008072, |
c5469e04 | 35 | 'upload_date': '20121012', |
64ccbf18 | 36 | 'duration': 5968.0, |
cbf915f3 PH |
37 | 'like_count': int, |
38 | 'view_count': int, | |
39 | 'thumbnail': 're:^http://.*\.jpg$' | |
b4444d5c | 40 | } |
22a6f150 PH |
41 | }, { |
42 | 'url': 'http://new.livestream.com/tedx/cityenglish', | |
43 | 'info_dict': { | |
44 | 'title': 'TEDCity2.0 (English)', | |
1def5f35 | 45 | 'id': '2245590', |
22a6f150 PH |
46 | }, |
47 | 'playlist_mincount': 4, | |
082b1155 JMF |
48 | }, { |
49 | 'url': 'http://new.livestream.com/chess24/tatasteelchess', | |
50 | 'info_dict': { | |
51 | 'title': 'Tata Steel Chess', | |
52 | 'id': '3705884', | |
53 | }, | |
54 | 'playlist_mincount': 60, | |
af63fed7 PH |
55 | }, { |
56 | 'url': 'https://new.livestream.com/accounts/362/events/3557232/videos/67864563/player?autoPlay=false&height=360&mute=false&width=640', | |
57 | 'only_matching': True, | |
4a20c9f6 YCH |
58 | }, { |
59 | 'url': 'http://livestream.com/bsww/concacafbeachsoccercampeonato2015', | |
60 | 'only_matching': True, | |
22a6f150 | 61 | }] |
64ccbf18 | 62 | _API_URL_TEMPLATE = 'http://livestream.com/api/accounts/%s/events/%s' |
63 | ||
64 | def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): | |
5b025168 | 65 | base_ele = find_xpath_attr( |
66 | smil, self._xpath_ns('.//meta', namespace), 'name', 'httpBase') | |
e1dd521e | 67 | base = base_ele.get('content') if base_ele is not None else 'http://livestreamvod-f.akamaihd.net/' |
b4444d5c | 68 | |
8f3034d8 | 69 | formats = [] |
64ccbf18 | 70 | video_nodes = smil.findall(self._xpath_ns('.//video', namespace)) |
8f3034d8 PH |
71 | |
72 | for vn in video_nodes: | |
64ccbf18 | 73 | tbr = int_or_none(vn.attrib.get('system-bitrate'), 1000) |
8f3034d8 | 74 | furl = ( |
64ccbf18 | 75 | '%s%s?v=3.0.3&fp=WIN%%2014,0,0,145' % (base, vn.attrib['src'])) |
8f3034d8 PH |
76 | if 'clipBegin' in vn.attrib: |
77 | furl += '&ssek=' + vn.attrib['clipBegin'] | |
78 | formats.append({ | |
79 | 'url': furl, | |
80 | 'format_id': 'smil_%d' % tbr, | |
81 | 'ext': 'flv', | |
82 | 'tbr': tbr, | |
83 | 'preference': -1000, | |
84 | }) | |
85 | return formats | |
86 | ||
b4444d5c | 87 | def _extract_video_info(self, video_data): |
cbf915f3 PH |
88 | video_id = compat_str(video_data['id']) |
89 | ||
90 | FORMAT_KEYS = ( | |
91 | ('sd', 'progressive_url'), | |
92 | ('hd', 'progressive_url_hd'), | |
72e785f3 | 93 | ) |
64ccbf18 | 94 | |
95 | formats = [] | |
96 | for format_id, key in FORMAT_KEYS: | |
97 | video_url = video_data.get(key) | |
98 | if video_url: | |
99 | ext = determine_ext(video_url) | |
7b813165 | 100 | if ext == 'm3u8': |
101 | continue | |
5b025168 | 102 | bitrate = int_or_none(self._search_regex( |
103 | r'(\d+)\.%s' % ext, video_url, 'bitrate', default=None)) | |
64ccbf18 | 104 | formats.append({ |
105 | 'url': video_url, | |
106 | 'format_id': format_id, | |
107 | 'tbr': bitrate, | |
108 | 'ext': ext, | |
109 | }) | |
cbf915f3 PH |
110 | |
111 | smil_url = video_data.get('smil_url') | |
112 | if smil_url: | |
7e5edcfd | 113 | formats.extend(self._extract_smil_formats(smil_url, video_id)) |
64ccbf18 | 114 | |
115 | m3u8_url = video_data.get('m3u8_url') | |
116 | if m3u8_url: | |
7e5edcfd S |
117 | formats.extend(self._extract_m3u8_formats( |
118 | m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) | |
64ccbf18 | 119 | |
120 | f4m_url = video_data.get('f4m_url') | |
121 | if f4m_url: | |
7e5edcfd S |
122 | formats.extend(self._extract_f4m_formats( |
123 | f4m_url, video_id, f4m_id='hds', fatal=False)) | |
cbf915f3 PH |
124 | self._sort_formats(formats) |
125 | ||
64ccbf18 | 126 | comments = [{ |
127 | 'author_id': comment.get('author_id'), | |
128 | 'author': comment.get('author', {}).get('full_name'), | |
129 | 'id': comment.get('id'), | |
130 | 'text': comment['text'], | |
131 | 'timestamp': parse_iso8601(comment.get('created_at')), | |
132 | } for comment in video_data.get('comments', {}).get('data', [])] | |
133 | ||
c5469e04 | 134 | return { |
cbf915f3 PH |
135 | 'id': video_id, |
136 | 'formats': formats, | |
c5469e04 | 137 | 'title': video_data['caption'], |
64ccbf18 | 138 | 'description': video_data.get('description'), |
cbf915f3 | 139 | 'thumbnail': video_data.get('thumbnail_url'), |
64ccbf18 | 140 | 'duration': float_or_none(video_data.get('duration'), 1000), |
141 | 'timestamp': parse_iso8601(video_data.get('publish_at')), | |
cbf915f3 | 142 | 'like_count': video_data.get('likes', {}).get('total'), |
64ccbf18 | 143 | 'comment_count': video_data.get('comments', {}).get('total'), |
cbf915f3 | 144 | 'view_count': video_data.get('views'), |
64ccbf18 | 145 | 'comments': comments, |
146 | } | |
147 | ||
148 | def _extract_stream_info(self, stream_info): | |
149 | broadcast_id = stream_info['broadcast_id'] | |
150 | is_live = stream_info.get('is_live') | |
151 | ||
152 | formats = [] | |
153 | smil_url = stream_info.get('play_url') | |
154 | if smil_url: | |
7e5edcfd | 155 | formats.extend(self._extract_smil_formats(smil_url, broadcast_id)) |
64ccbf18 | 156 | |
5b025168 | 157 | entry_protocol = 'm3u8' if is_live else 'm3u8_native' |
64ccbf18 | 158 | m3u8_url = stream_info.get('m3u8_url') |
159 | if m3u8_url: | |
7e5edcfd S |
160 | formats.extend(self._extract_m3u8_formats( |
161 | m3u8_url, broadcast_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False)) | |
64ccbf18 | 162 | |
163 | rtsp_url = stream_info.get('rtsp_url') | |
164 | if rtsp_url: | |
165 | formats.append({ | |
166 | 'url': rtsp_url, | |
167 | 'format_id': 'rtsp', | |
168 | }) | |
169 | self._sort_formats(formats) | |
170 | ||
171 | return { | |
172 | 'id': broadcast_id, | |
173 | 'formats': formats, | |
174 | 'title': self._live_title(stream_info['stream_title']) if is_live else stream_info['stream_title'], | |
175 | 'thumbnail': stream_info.get('thumbnail_url'), | |
176 | 'is_live': is_live, | |
c5469e04 | 177 | } |
b4444d5c | 178 | |
64ccbf18 | 179 | def _extract_event(self, event_data): |
180 | event_id = compat_str(event_data['id']) | |
181 | account_id = compat_str(event_data['owner_account_id']) | |
182 | feed_root_url = self._API_URL_TEMPLATE % (account_id, event_id) + '/feed.json' | |
183 | ||
184 | stream_info = event_data.get('stream_info') | |
185 | if stream_info: | |
186 | return self._extract_stream_info(stream_info) | |
187 | ||
188 | last_video = None | |
189 | entries = [] | |
190 | for i in itertools.count(1): | |
191 | if last_video is None: | |
192 | info_url = feed_root_url | |
193 | else: | |
194 | info_url = '{root}?&id={id}&newer=-1&type=video'.format( | |
195 | root=feed_root_url, id=last_video) | |
5b025168 | 196 | videos_info = self._download_json( |
197 | info_url, event_id, 'Downloading page {0}'.format(i))['data'] | |
64ccbf18 | 198 | videos_info = [v['data'] for v in videos_info if v['type'] == 'video'] |
199 | if not videos_info: | |
200 | break | |
201 | for v in videos_info: | |
202 | entries.append(self.url_result( | |
203 | 'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v['id']), | |
204 | 'Livestream', v['id'], v['caption'])) | |
205 | last_video = videos_info[-1]['id'] | |
206 | return self.playlist_result(entries, event_id, event_data['full_name']) | |
082b1155 | 207 | |
b4444d5c JMF |
208 | def _real_extract(self, url): |
209 | mobj = re.match(self._VALID_URL, url) | |
210 | video_id = mobj.group('id') | |
64ccbf18 | 211 | event = mobj.group('event_id') or mobj.group('event_name') |
212 | account = mobj.group('account_id') or mobj.group('account_name') | |
213 | api_url = self._API_URL_TEMPLATE % (account, event) | |
214 | if video_id: | |
5b025168 | 215 | video_data = self._download_json( |
216 | api_url + '/videos/%s' % video_id, video_id) | |
64ccbf18 | 217 | return self._extract_video_info(video_data) |
22a6f150 | 218 | else: |
64ccbf18 | 219 | event_data = self._download_json(api_url, video_id) |
220 | return self._extract_event(event_data) | |
c66d2baa JMF |
221 | |
222 | ||
223 | # The original version of Livestream uses a different system | |
224 | class LivestreamOriginalIE(InfoExtractor): | |
c5469e04 | 225 | IE_NAME = 'livestream:original' |
a9055266 | 226 | _VALID_URL = r'''(?x)https?://original\.livestream\.com/ |
5b025168 | 227 | (?P<user>[^/\?#]+)(?:/(?P<type>video|folder) |
228 | (?:(?:\?.*?Id=|/)(?P<id>.*?)(&|$))?)? | |
78338f71 | 229 | ''' |
22a6f150 | 230 | _TESTS = [{ |
a9055266 | 231 | 'url': 'http://original.livestream.com/dealbook/video?clipId=pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', |
c5469e04 S |
232 | 'info_dict': { |
233 | 'id': 'pla_8aa4a3f1-ba15-46a4-893b-902210e138fb', | |
883340c1 | 234 | 'ext': 'mp4', |
c5469e04 | 235 | 'title': 'Spark 1 (BitCoin) with Cameron Winklevoss & Tyler Winklevoss of Winklevoss Capital', |
64ccbf18 | 236 | 'duration': 771.301, |
237 | 'view_count': int, | |
c66d2baa | 238 | }, |
22a6f150 | 239 | }, { |
a9055266 | 240 | 'url': 'https://original.livestream.com/newplay/folder?dirId=a07bf706-d0e4-4e75-a747-b021d84f2fd3', |
22a6f150 PH |
241 | 'info_dict': { |
242 | 'id': 'a07bf706-d0e4-4e75-a747-b021d84f2fd3', | |
243 | }, | |
244 | 'playlist_mincount': 4, | |
5b025168 | 245 | }, { |
246 | # live stream | |
c71d2e20 | 247 | 'url': 'http://original.livestream.com/znsbahamas', |
5b025168 | 248 | 'only_matching': True, |
22a6f150 | 249 | }] |
c66d2baa | 250 | |
5b025168 | 251 | def _extract_video_info(self, user, video_id): |
252 | api_url = 'http://x%sx.api.channel.livestream.com/2.0/clipdetails?extendedInfo=true&id=%s' % (user, video_id) | |
e26f8712 | 253 | info = self._download_xml(api_url, video_id) |
5b025168 | 254 | |
c66d2baa | 255 | item = info.find('channel').find('item') |
5b025168 | 256 | title = xpath_text(item, 'title') |
64ccbf18 | 257 | media_ns = {'media': 'http://search.yahoo.com/mrss'} |
5b025168 | 258 | thumbnail_url = xpath_attr( |
259 | item, xpath_with_ns('media:thumbnail', media_ns), 'url') | |
260 | duration = float_or_none(xpath_attr( | |
261 | item, xpath_with_ns('media:content', media_ns), 'duration')) | |
64ccbf18 | 262 | ls_ns = {'ls': 'http://api.channel.livestream.com/2.0'} |
5b025168 | 263 | view_count = int_or_none(xpath_text( |
264 | item, xpath_with_ns('ls:viewsCount', ls_ns))) | |
64ccbf18 | 265 | |
5b025168 | 266 | return { |
267 | 'id': video_id, | |
268 | 'title': title, | |
269 | 'thumbnail': thumbnail_url, | |
270 | 'duration': duration, | |
271 | 'view_count': view_count, | |
272 | } | |
273 | ||
274 | def _extract_video_formats(self, video_data, video_id, entry_protocol): | |
275 | formats = [] | |
276 | ||
277 | progressive_url = video_data.get('progressiveUrl') | |
278 | if progressive_url: | |
279 | formats.append({ | |
280 | 'url': progressive_url, | |
281 | 'format_id': 'http', | |
282 | }) | |
64ccbf18 | 283 | |
5b025168 | 284 | m3u8_url = video_data.get('httpUrl') |
64ccbf18 | 285 | if m3u8_url: |
7e5edcfd S |
286 | formats.extend(self._extract_m3u8_formats( |
287 | m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False)) | |
64ccbf18 | 288 | |
5b025168 | 289 | rtsp_url = video_data.get('rtspUrl') |
64ccbf18 | 290 | if rtsp_url: |
291 | formats.append({ | |
292 | 'url': rtsp_url, | |
293 | 'format_id': 'rtsp', | |
294 | }) | |
c66d2baa | 295 | |
5b025168 | 296 | self._sort_formats(formats) |
297 | return formats | |
78338f71 JMF |
298 | |
299 | def _extract_folder(self, url, folder_id): | |
300 | webpage = self._download_webpage(url, folder_id) | |
22a6f150 PH |
301 | paths = orderedSet(re.findall( |
302 | r'''(?x)(?: | |
303 | <li\s+class="folder">\s*<a\s+href="| | |
304 | <a\s+href="(?=https?://livestre\.am/) | |
305 | )([^"]+)"''', webpage)) | |
78338f71 | 306 | |
64ccbf18 | 307 | entries = [{ |
308 | '_type': 'url', | |
309 | 'url': compat_urlparse.urljoin(url, p), | |
310 | } for p in paths] | |
311 | ||
312 | return self.playlist_result(entries, folder_id) | |
78338f71 JMF |
313 | |
314 | def _real_extract(self, url): | |
315 | mobj = re.match(self._VALID_URL, url) | |
78338f71 JMF |
316 | user = mobj.group('user') |
317 | url_type = mobj.group('type') | |
5b025168 | 318 | content_id = mobj.group('id') |
78338f71 | 319 | if url_type == 'folder': |
5b025168 | 320 | return self._extract_folder(url, content_id) |
78338f71 | 321 | else: |
5b025168 | 322 | # this url is used on mobile devices |
323 | stream_url = 'http://x%sx.api.channel.livestream.com/3.0/getstream.json' % user | |
324 | info = {} | |
325 | if content_id: | |
326 | stream_url += '?id=%s' % content_id | |
327 | info = self._extract_video_info(user, content_id) | |
328 | else: | |
329 | content_id = user | |
330 | webpage = self._download_webpage(url, content_id) | |
331 | info = { | |
332 | 'title': self._og_search_title(webpage), | |
333 | 'description': self._og_search_description(webpage), | |
334 | 'thumbnail': self._search_regex(r'channelLogo.src\s*=\s*"([^"]+)"', webpage, 'thumbnail', None), | |
335 | } | |
336 | video_data = self._download_json(stream_url, content_id) | |
337 | is_live = video_data.get('isLive') | |
338 | entry_protocol = 'm3u8' if is_live else 'm3u8_native' | |
339 | info.update({ | |
340 | 'id': content_id, | |
341 | 'title': self._live_title(info['title']) if is_live else info['title'], | |
342 | 'formats': self._extract_video_formats(video_data, content_id, entry_protocol), | |
343 | 'is_live': is_live, | |
344 | }) | |
345 | return info | |
78338f71 JMF |
346 | |
347 | ||
348 | # The server doesn't support HEAD request, the generic extractor can't detect | |
349 | # the redirection | |
350 | class LivestreamShortenerIE(InfoExtractor): | |
351 | IE_NAME = 'livestream:shortener' | |
352 | IE_DESC = False # Do not list | |
353 | _VALID_URL = r'https?://livestre\.am/(?P<id>.+)' | |
354 | ||
355 | def _real_extract(self, url): | |
356 | mobj = re.match(self._VALID_URL, url) | |
357 | id = mobj.group('id') | |
358 | webpage = self._download_webpage(url, id) | |
359 | ||
360 | return { | |
361 | '_type': 'url', | |
362 | 'url': self._og_search_url(webpage), | |
363 | } |