- if mobj is None:
- raise ExtractorError(u'Invalid URL: %s' % url)
- video_id = mobj.group('id')
- webpage = self._download_webpage(url, video_id)
- m_id = re.search(r'YUI\.namespace\("Media"\)\.CONTENT_ID = "(?P<new_id>.+?)";', webpage)
-
- if m_id is None:
- # TODO: Check which url parameters are required
- info_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;lmsoverride=1;outputformat=mrss;cb=974419660;id=%s;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
- webpage = self._download_webpage(info_url, video_id, u'Downloading info webpage')
- info_re = r'''<title><!\[CDATA\[(?P<title>.*?)\]\]></title>.*
- <description><!\[CDATA\[(?P<description>.*?)\]\]></description>.*
- <media:pubStart><!\[CDATA\[(?P<date>.*?)\ .*\]\]></media:pubStart>.*
- <media:content\ medium="image"\ url="(?P<thumb>.*?)"\ name="LARGETHUMB"
- '''
- self.report_extraction(video_id)
- m_info = re.search(info_re, webpage, re.VERBOSE|re.DOTALL)
- if m_info is None:
- raise ExtractorError(u'Unable to extract video info')
- video_title = m_info.group('title')
- video_description = m_info.group('description')
- video_thumb = m_info.group('thumb')
- video_date = m_info.group('date')
- video_date = datetime.datetime.strptime(video_date, '%m/%d/%Y').strftime('%Y%m%d')
-
- # TODO: Find a way to get mp4 videos
- rest_url = 'http://cosmos.bcst.yahoo.com/rest/v2/pops;element=stream;outputformat=mrss;id=%s;lmsoverride=1;bw=375;dynamicstream=1;cb=83521105;tech=flv,mp4;rd=news.yahoo.com;datacontext=mdb;lg=KCa2IihxG3qE60vQ7HtyUy' % video_id
- webpage = self._download_webpage(rest_url, video_id, u'Downloading video url webpage')
- m_rest = re.search(r'<media:content url="(?P<url>.*?)" path="(?P<path>.*?)"', webpage)
- video_url = m_rest.group('url')
- video_path = m_rest.group('path')
- if m_rest is None:
- raise ExtractorError(u'Unable to extract video url')
-
- else: # We have to use a different method if another id is defined
- long_id = m_id.group('new_id')
- info_url = 'http://video.query.yahoo.com/v1/public/yql?q=SELECT%20*%20FROM%20yahoo.media.video.streams%20WHERE%20id%3D%22' + long_id + '%22%20AND%20format%3D%22mp4%2Cflv%22%20AND%20protocol%3D%22rtmp%2Chttp%22%20AND%20plrs%3D%2286Gj0vCaSzV_Iuf6hNylf2%22%20AND%20acctid%3D%22389%22%20AND%20plidl%3D%22%22%20AND%20pspid%3D%22792700001%22%20AND%20offnetwork%3D%22false%22%20AND%20site%3D%22ivy%22%20AND%20lang%3D%22en-US%22%20AND%20region%3D%22US%22%20AND%20override%3D%22none%22%3B&env=prod&format=json&callback=YUI.Env.JSONP.yui_3_8_1_1_1368368376830_335'
- webpage = self._download_webpage(info_url, video_id, u'Downloading info json')
- json_str = re.search(r'YUI.Env.JSONP.yui.*?\((.*?)\);', webpage).group(1)
- info = json.loads(json_str)
- res = info[u'query'][u'results'][u'mediaObj'][0]
- stream = res[u'streams'][0]
- video_path = stream[u'path']
- video_url = stream[u'host']
- meta = res[u'meta']
- video_title = meta[u'title']
- video_description = meta[u'description']
- video_thumb = meta[u'thumbnail']
- video_date = None # I can't find it
-
- info_dict = {
- 'id': video_id,
- 'url': video_url,
- 'play_path': video_path,
- 'title':video_title,
- 'description': video_description,
- 'thumbnail': video_thumb,
- 'upload_date': video_date,
- 'ext': 'flv',
- }
- return info_dict
+ display_id = mobj.group('display_id') or self._match_id(url)
+ page_id = mobj.group('id')
+ url = mobj.group('url')
+ host = mobj.group('host')
+ webpage, urlh = self._download_webpage_handle(url, display_id)
+ if 'err=404' in urlh.geturl():
+ raise ExtractorError('Video gone', expected=True)
+
+ # Look for iframed media first
+ entries = []
+ iframe_urls = re.findall(r'<iframe[^>]+src="(/video/.+?-\d+\.html\?format=embed.*?)"', webpage)
+ for idx, iframe_url in enumerate(iframe_urls):
+ iframepage = self._download_webpage(
+ host + iframe_url, display_id,
+ note='Downloading iframe webpage for video #%d' % idx)
+ items_json = self._search_regex(
+ r'mediaItems: (\[.+?\])$', iframepage, 'items', flags=re.MULTILINE, default=None)
+ if items_json:
+ items = json.loads(items_json)
+ video_id = items[0]['id']
+ entries.append(self._get_info(video_id, display_id, webpage))
+ if entries:
+ return self.playlist_result(entries, page_id)
+
+ # Look for NBCSports iframes
+ nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
+ if nbc_sports_url:
+ return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key())
+
+ # Look for Brightcove New Studio embeds
+ bc_url = BrightcoveNewIE._extract_url(webpage)
+ if bc_url:
+ return self.url_result(bc_url, BrightcoveNewIE.ie_key())
+
+ # Query result is often embedded in webpage as JSON. Sometimes explicit requests
+ # to video API results in a failure with geo restriction reason therefore using
+ # embedded query result when present sounds reasonable.
+ config_json = self._search_regex(
+ r'window\.Af\.bootstrap\[[^\]]+\]\s*=\s*({.*?"applet_type"\s*:\s*"td-applet-videoplayer".*?});(?:</script>|$)',
+ webpage, 'videoplayer applet', default=None)
+ if config_json:
+ config = self._parse_json(config_json, display_id, fatal=False)
+ if config:
+ sapi = config.get('models', {}).get('applet_model', {}).get('data', {}).get('sapi')
+ if sapi and 'query' in sapi:
+ return self._extract_info(display_id, sapi, webpage)
+
+ items_json = self._search_regex(
+ r'mediaItems: ({.*?})$', webpage, 'items', flags=re.MULTILINE,
+ default=None)
+ if items_json is None:
+ alias = self._search_regex(
+ r'"aliases":{"video":"(.*?)"', webpage, 'alias', default=None)
+ if alias is not None:
+ alias_info = self._download_json(
+ 'https://www.yahoo.com/_td/api/resource/VideoService.videos;video_aliases=["%s"]' % alias,
+ display_id, 'Downloading alias info')
+ video_id = alias_info[0]['id']
+ else:
+ CONTENT_ID_REGEXES = [
+ r'YUI\.namespace\("Media"\)\.CONTENT_ID\s*=\s*"([^"]+)"',
+ r'root\.App\.Cache\.context\.videoCache\.curVideo = \{"([^"]+)"',
+ r'"first_videoid"\s*:\s*"([^"]+)"',
+ r'%s[^}]*"ccm_id"\s*:\s*"([^"]+)"' % re.escape(page_id),
+ r'<article[^>]data-uuid=["\']([^"\']+)',
+ r'yahoo://article/view\?.*\buuid=([^&"\']+)',
+ ]
+ video_id = self._search_regex(
+ CONTENT_ID_REGEXES, webpage, 'content ID')
+ else:
+ items = json.loads(items_json)
+ info = items['mediaItems']['query']['results']['mediaObj'][0]
+ # The 'meta' field is not always in the video webpage, we request it
+ # from another page
+ video_id = info['id']
+ return self._get_info(video_id, display_id, webpage)
+
+ def _extract_info(self, display_id, query, webpage):
+ info = query['query']['results']['mediaObj'][0]
+ meta = info.get('meta')
+ video_id = info.get('id')
+
+ if not meta:
+ msg = info['status'].get('msg')
+ if msg:
+ raise ExtractorError(
+ '%s returned error: %s' % (self.IE_NAME, msg), expected=True)
+ raise ExtractorError('Unable to extract media object meta')
+
+ formats = []
+ for s in info['streams']:
+ format_info = {
+ 'width': int_or_none(s.get('width')),
+ 'height': int_or_none(s.get('height')),
+ 'tbr': int_or_none(s.get('bitrate')),
+ }
+
+ host = s['host']
+ path = s['path']
+ if host.startswith('rtmp'):
+ format_info.update({
+ 'url': host,
+ 'play_path': path,
+ 'ext': 'flv',
+ })
+ else:
+ if s.get('format') == 'm3u8_playlist':
+ format_info['protocol'] = 'm3u8_native'
+ format_info['ext'] = 'mp4'
+ format_url = compat_urlparse.urljoin(host, path)
+ format_info['url'] = format_url
+ formats.append(format_info)
+
+ self._sort_formats(formats)
+
+ closed_captions = self._html_search_regex(
+ r'"closedcaptions":(\[[^\]]+\])', webpage, 'closed captions',
+ default='[]')
+
+ cc_json = self._parse_json(closed_captions, video_id, fatal=False)
+ subtitles = {}
+ if cc_json:
+ for closed_caption in cc_json:
+ lang = closed_caption['lang']
+ if lang not in subtitles:
+ subtitles[lang] = []
+ subtitles[lang].append({
+ 'url': closed_caption['url'],
+ 'ext': mimetype2ext(closed_caption['content_type']),
+ })
+
+ return {
+ 'id': video_id,
+ 'display_id': display_id,
+ 'title': unescapeHTML(meta['title']),
+ 'formats': formats,
+ 'description': clean_html(meta['description']),
+ 'thumbnail': meta['thumbnail'] if meta.get('thumbnail') else self._og_search_thumbnail(webpage),
+ 'duration': int_or_none(meta.get('duration')),
+ 'subtitles': subtitles,
+ }
+
+ def _get_info(self, video_id, display_id, webpage):
+ region = self._search_regex(
+ r'\\?"region\\?"\s*:\s*\\?"([^"]+?)\\?"',
+ webpage, 'region', fatal=False, default='US')
+ data = compat_urllib_parse_urlencode({
+ 'protocol': 'http',
+ 'region': region.upper(),
+ })
+ query_url = (
+ 'https://video.media.yql.yahoo.com/v1/video/sapi/streams/'
+ '{id}?{data}'.format(id=video_id, data=data))
+ query_result = self._download_json(
+ query_url, display_id, 'Downloading video info')
+ return self._extract_info(display_id, query_result, webpage)
+
Rapsys Git / youtubedl / blobdiff