+ @staticmethod
+ def _extract_url(ie, webpage):
+ urls = BrightcoveNewIE._extract_urls(ie, webpage)
+ return urls[0] if urls else None
+
+ @staticmethod
+ def _extract_urls(ie, webpage):
+ # Reference:
+ # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
+ # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#tag
+ # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
+ # 4. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/in-page-embed-player-implementation.html
+ # 5. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
+
+ entries = []
+
+ # Look for iframe embeds [1]
+ for _, url in re.findall(
+ r'<iframe[^>]+src=(["\'])((?:https?:)?//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
+ entries.append(url if url.startswith('http') else 'http:' + url)
+
+ # Look for <video> tags [2] and embed_in_page embeds [3]
+ # [2] looks like:
+ for video, script_tag, account_id, player_id, embed in re.findall(
+ r'''(?isx)
+ (<video\s+[^>]*\bdata-video-id\s*=\s*['"]?[^>]+>)
+ (?:.*?
+ (<script[^>]+
+ src=["\'](?:https?:)?//players\.brightcove\.net/
+ (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
+ )
+ )?
+ ''', webpage):
+ attrs = extract_attributes(video)
+
+ # According to examples from [4] it's unclear whether video id
+ # may be optional and what to do when it is
+ video_id = attrs.get('data-video-id')
+ if not video_id:
+ continue
+
+ account_id = account_id or attrs.get('data-account')
+ if not account_id:
+ continue
+
+ player_id = player_id or attrs.get('data-player') or 'default'
+ embed = embed or attrs.get('data-embed') or 'default'
+
+ bc_url = 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' % (
+ account_id, player_id, embed, video_id)
+
+ # Some brightcove videos may be embedded with video tag only and
+ # without script tag or any mentioning of brightcove at all. Such
+ # embeds are considered ambiguous since they are matched based only
+ # on data-video-id and data-account attributes and in the wild may
+ # not be brightcove embeds at all. Let's check reconstructed
+ # brightcove URLs in case of such embeds and only process valid
+ # ones. By this we ensure there is indeed a brightcove embed.
+ if not script_tag and not ie._is_valid_url(
+ bc_url, video_id, 'possible brightcove video'):
+ continue
+
+ entries.append(bc_url)
+
+ return entries
+
+ def _parse_brightcove_metadata(self, json_data, video_id, headers={}):
+ title = json_data['name'].strip()
+
+ formats = []
+ for source in json_data.get('sources', []):
+ container = source.get('container')
+ ext = mimetype2ext(source.get('type'))
+ src = source.get('src')
+ # https://support.brightcove.com/playback-api-video-fields-reference#key_systems_object
+ if ext == 'ism' or container == 'WVM' or source.get('key_systems'):
+ continue
+ elif ext == 'm3u8' or container == 'M2TS':
+ if not src:
+ continue
+ formats.extend(self._extract_m3u8_formats(
+ src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+ elif ext == 'mpd':
+ if not src:
+ continue
+ formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
+ else:
+ streaming_src = source.get('streaming_src')
+ stream_name, app_name = source.get('stream_name'), source.get('app_name')
+ if not src and not streaming_src and (not stream_name or not app_name):
+ continue
+ tbr = float_or_none(source.get('avg_bitrate'), 1000)
+ height = int_or_none(source.get('height'))
+ width = int_or_none(source.get('width'))
+ f = {
+ 'tbr': tbr,
+ 'filesize': int_or_none(source.get('size')),
+ 'container': container,
+ 'ext': ext or container.lower(),
+ }
+ if width == 0 and height == 0:
+ f.update({
+ 'vcodec': 'none',
+ })
+ else:
+ f.update({
+ 'width': width,
+ 'height': height,
+ 'vcodec': source.get('codec'),
+ })
+
+ def build_format_id(kind):
+ format_id = kind
+ if tbr:
+ format_id += '-%dk' % int(tbr)
+ if height:
+ format_id += '-%dp' % height
+ return format_id
+
+ if src or streaming_src:
+ f.update({
+ 'url': src or streaming_src,
+ 'format_id': build_format_id('http' if src else 'http-streaming'),
+ 'source_preference': 0 if src else -1,
+ })
+ else:
+ f.update({
+ 'url': app_name,
+ 'play_path': stream_name,
+ 'format_id': build_format_id('rtmp'),
+ })
+ formats.append(f)
+ if not formats:
+ # for sonyliv.com DRM protected videos
+ s3_source_url = json_data.get('custom_fields', {}).get('s3sourceurl')
+ if s3_source_url:
+ formats.append({
+ 'url': s3_source_url,
+ 'format_id': 'source',
+ })
+
+ errors = json_data.get('errors')
+ if not formats and errors:
+ error = errors[0]
+ raise ExtractorError(
+ error.get('message') or error.get('error_subcode') or error['error_code'], expected=True)
+
+ self._sort_formats(formats)
+
+ for f in formats:
+ f.setdefault('http_headers', {}).update(headers)
+
+ subtitles = {}
+ for text_track in json_data.get('text_tracks', []):
+ if text_track.get('src'):
+ subtitles.setdefault(text_track.get('srclang'), []).append({
+ 'url': text_track['src'],
+ })
+
+ is_live = False
+ duration = float_or_none(json_data.get('duration'), 1000)
+ if duration is not None and duration <= 0:
+ is_live = True
+
+ return {
+ 'id': video_id,
+ 'title': self._live_title(title) if is_live else title,
+ 'description': clean_html(json_data.get('description')),
+ 'thumbnail': json_data.get('thumbnail') or json_data.get('poster'),
+ 'duration': duration,
+ 'timestamp': parse_iso8601(json_data.get('published_at')),
+ 'uploader_id': json_data.get('account_id'),
+ 'formats': formats,
+ 'subtitles': subtitles,
+ 'tags': json_data.get('tags', []),
+ 'is_live': is_live,
+ }
+
+ def _real_extract(self, url):
+ url, smuggled_data = unsmuggle_url(url, {})
+ self._initialize_geo_bypass({
+ 'countries': smuggled_data.get('geo_countries'),
+ 'ip_blocks': smuggled_data.get('geo_ip_blocks'),
+ })
+
+ account_id, player_id, embed, content_type, video_id = re.match(self._VALID_URL, url).groups()
+
+ policy_key_id = '%s_%s' % (account_id, player_id)
+ policy_key = self._downloader.cache.load('brightcove', policy_key_id)
+ policy_key_extracted = False
+ store_pk = lambda x: self._downloader.cache.store('brightcove', policy_key_id, x)
+
+ def extract_policy_key():
+ webpage = self._download_webpage(
+ 'http://players.brightcove.net/%s/%s_%s/index.min.js'
+ % (account_id, player_id, embed), video_id)
+
+ policy_key = None
+
+ catalog = self._search_regex(
+ r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
+ if catalog:
+ catalog = self._parse_json(
+ js_to_json(catalog), video_id, fatal=False)
+ if catalog:
+ policy_key = catalog.get('policyKey')
+
+ if not policy_key:
+ policy_key = self._search_regex(
+ r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
+ webpage, 'policy key', group='pk')
+
+ store_pk(policy_key)
+ return policy_key
+
+ api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/%ss/%s' % (account_id, content_type, video_id)
+ headers = {}
+ referrer = smuggled_data.get('referrer')
+ if referrer:
+ headers.update({
+ 'Referer': referrer,
+ 'Origin': re.search(r'https?://[^/]+', referrer).group(0),
Rapsys Git / youtubedl / blobdiff