-# encoding: utf-8
+# coding: utf-8
from __future__ import unicode_literals
import re
unescapeHTML,
unsmuggle_url,
update_url_query,
+ clean_html,
+ mimetype2ext,
)
params = {}
- playerID = find_param('playerID')
+ playerID = find_param('playerID') or find_param('playerId')
if playerID is None:
raise ExtractorError('Cannot find player ID')
params['playerID'] = playerID
# // build Brightcove <object /> XML
# }
m = re.search(
- r'''(?x)customBC.\createVideo\(
+ r'''(?x)customBC\.createVideo\(
.*? # skipping width and height
["\'](?P<playerID>\d+)["\']\s*,\s* # playerID
["\'](?P<playerKey>AQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters
"""Return a list of all Brightcove URLs from the webpage """
url_m = re.search(
- r'<meta\s+property=[\'"]og:video[\'"]\s+content=[\'"](https?://(?:secure|c)\.brightcove.com/[^\'"]+)[\'"]',
- webpage)
+ r'''(?x)
+ <meta\s+
+ (?:property|itemprop)=([\'"])(?:og:video|embedURL)\1[^>]+
+ content=([\'"])(?P<url>https?://(?:secure|c)\.brightcove.com/(?:(?!\2).)+)\2
+ ''', webpage)
if url_m:
- url = unescapeHTML(url_m.group(1))
+ url = unescapeHTML(url_m.group('url'))
# Some sites don't add it, we can't download with this url, for example:
# http://www.ktvu.com/videos/news/raw-video-caltrain-releases-video-of-man-almost/vCTZdY/
- if 'playerKey' in url or 'videoId' in url:
+ if 'playerKey' in url or 'videoId' in url or 'idVideo' in url:
return [url]
matches = re.findall(
url, smuggled_data = unsmuggle_url(url, {})
# Change the 'videoId' and others field to '@videoPlayer'
- url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
+ url = re.sub(r'(?<=[?&])(videoI(d|D)|idVideo|bctid)', '%40videoPlayer', url)
# Change bckey (used by bcove.me urls) to playerKey
url = re.sub(r'(?<=[?&])bckey', 'playerKey', url)
mobj = re.match(self._VALID_URL, url)
formats = []
for source in json_data.get('sources', []):
container = source.get('container')
- source_type = source.get('type')
+ ext = mimetype2ext(source.get('type'))
src = source.get('src')
- if source_type == 'application/x-mpegURL' or container == 'M2TS':
+ if ext == 'ism' or container == 'WVM':
+ 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 source_type == 'application/dash+xml':
+ elif ext == 'mpd':
if not src:
continue
formats.extend(self._extract_mpd_formats(src, video_id, 'dash', fatal=False))
'tbr': tbr,
'filesize': int_or_none(source.get('size')),
'container': container,
- 'ext': container.lower(),
+ 'ext': ext or container.lower(),
}
if width == 0 and height == 0:
f.update({
'url': text_track['src'],
})
+ is_live = False
+ duration = float_or_none(json_data.get('duration'), 1000)
+ if duration and duration < 0:
+ is_live = True
+
return {
'id': video_id,
- 'title': title,
- 'description': json_data.get('description'),
+ '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': float_or_none(json_data.get('duration'), 1000),
+ 'duration': duration,
'timestamp': parse_iso8601(json_data.get('published_at')),
'uploader_id': account_id,
'formats': formats,
'subtitles': subtitles,
'tags': json_data.get('tags', []),
+ 'is_live': is_live,
}