from ..utils import (
compat_urllib_parse,
find_xpath_attr,
+ fix_xml_ampersands,
compat_urlparse,
compat_str,
compat_urllib_request,
+ compat_parse_qs,
+ determine_ext,
ExtractorError,
unsmuggle_url,
+ unescapeHTML,
)
class BrightcoveIE(InfoExtractor):
_VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
_FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
- _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
_TESTS = [
{
# From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
- 'file': '2371591881001.mp4',
'md5': '5423e113865d26e40624dce2e4b45d95',
'note': 'Test Brightcove downloads and detection in GenericIE',
'info_dict': {
+ 'id': '2371591881001',
+ 'ext': 'mp4',
'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
'uploader': '8TV',
'description': 'md5:a950cc4285c43e44d763d036710cd9cd',
{
# From http://medianetwork.oracle.com/video/player/1785452137001
'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
- 'file': '1785452137001.flv',
'info_dict': {
+ 'id': '1785452137001',
+ 'ext': 'flv',
'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
'uploader': 'Oracle',
'uploader': 'National Ballet of Canada',
},
},
+ {
+ # test flv videos served by akamaihd.net
+ # From http://www.redbull.com/en/bike/stories/1331655643987/replay-uci-dh-world-cup-2014-from-fort-william
+ 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?%40videoPlayer=ref%3ABC2996102916001&linkBaseURL=http%3A%2F%2Fwww.redbull.com%2Fen%2Fbike%2Fvideos%2F1331655630249%2Freplay-uci-fort-william-2014-dh&playerKey=AQ%7E%7E%2CAAAApYJ7UqE%7E%2Cxqr_zXk0I-zzNndy8NlHogrCb5QdyZRf&playerID=1398061561001#__youtubedl_smuggle=%7B%22Referer%22%3A+%22http%3A%2F%2Fwww.redbull.com%2Fen%2Fbike%2Fstories%2F1331655643987%2Freplay-uci-dh-world-cup-2014-from-fort-william%22%7D',
+ # The md5 checksum changes on each download
+ 'info_dict': {
+ 'id': '2996102916001',
+ 'ext': 'flv',
+ 'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
+ 'uploader': 'Red Bull TV',
+ 'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
+ },
+ },
]
@classmethod
lambda m: m.group(1) + '/>', object_str)
# Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
object_str = object_str.replace('<--', '<!--')
+ object_str = fix_xml_ampersands(object_str)
+
+ object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
+
+ fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
+ if fv_el is not None:
+ flashvars = dict(
+ (k, v[0])
+ for k, v in compat_parse_qs(fv_el.attrib['value']).items())
+ else:
+ flashvars = {}
- object_doc = xml.etree.ElementTree.fromstring(object_str)
- assert 'BrightcoveExperience' in object_doc.attrib['class']
- params = {
- 'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
- }
def find_param(name):
+ if name in flashvars:
+ return flashvars[name]
node = find_xpath_attr(object_doc, './param', 'name', name)
if node is not None:
return node.attrib['value']
return None
+
+ params = {}
+
+ playerID = find_param('playerID')
+ if playerID is None:
+ raise ExtractorError('Cannot find player ID')
+ params['playerID'] = playerID
+
playerKey = find_param('playerKey')
# Not all pages define this value
if playerKey is not None:
@classmethod
def _extract_brightcove_url(cls, webpage):
- """Try to extract the brightcove url from the wepbage, returns None
+ """Try to extract the brightcove url from the webpage, returns None
if it can't be found
"""
- m_brightcove = re.search(
- r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>',
- webpage, re.DOTALL)
- if m_brightcove is not None:
- return cls._build_brighcove_url(m_brightcove.group())
- else:
- return None
+ urls = cls._extract_brightcove_urls(webpage)
+ return urls[0] if urls else None
+
+ @classmethod
+ def _extract_brightcove_urls(cls, webpage):
+ """Return a list of all Brightcove URLs from the webpage """
+
+ url_m = re.search(r'<meta\s+property="og:video"\s+content="(http://c.brightcove.com/[^"]+)"', webpage)
+ if url_m:
+ url = unescapeHTML(url_m.group(1))
+ # 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:
+ return [url]
+
+ matches = re.findall(
+ r'''(?sx)<object
+ (?:
+ [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
+ [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
+ ).+?</object>''',
+ webpage)
+ return [cls._build_brighcove_url(m) for m in matches]
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
webpage = self._download_webpage(req, video_id)
self.report_extraction(video_id)
- info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
+ info = self._search_regex(r'var experienceJSON = ({.*});', webpage, 'json')
info = json.loads(info)['data']
video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+ video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
return self._extract_video_info(video_info)
def _get_playlist_info(self, player_key):
- playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
- player_key, 'Downloading playlist information')
+ info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
+ playlist_info = self._download_webpage(
+ info_url, player_key, 'Downloading playlist information')
json_data = json.loads(playlist_info)
if 'videoList' not in json_data:
def _extract_video_info(self, video_info):
info = {
'id': compat_str(video_info['id']),
- 'title': video_info['displayName'],
+ 'title': video_info['displayName'].strip(),
'description': video_info.get('shortDescription'),
'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
'uploader': video_info.get('publisherName'),
renditions = video_info.get('renditions')
if renditions:
- renditions = sorted(renditions, key=lambda r: r['size'])
- info['formats'] = [{
- 'url': rend['defaultURL'],
- 'height': rend.get('frameHeight'),
- 'width': rend.get('frameWidth'),
- } for rend in renditions]
+ formats = []
+ for rend in renditions:
+ url = rend['defaultURL']
+ if rend['remote']:
+ # This type of renditions are served through akamaihd.net,
+ # but they don't use f4m manifests
+ url = url.replace('control/', '') + '?&v=3.3.0&fp=13&r=FEEFJ&g=RTSJIMBMPFPB'
+ ext = 'flv'
+ else:
+ ext = determine_ext(url)
+ size = rend.get('size')
+ formats.append({
+ 'url': url,
+ 'ext': ext,
+ 'height': rend.get('frameHeight'),
+ 'width': rend.get('frameWidth'),
+ 'filesize': size if size != 0 else None,
+ })
+ self._sort_formats(formats)
+ info['formats'] = formats
elif video_info.get('FLVFullLengthURL') is not None:
info.update({
'url': video_info['FLVFullLengthURL'],
})
- else:
+
+ if self._downloader.params.get('include_ads', False):
+ adServerURL = video_info.get('_youtubedl_adServerURL')
+ if adServerURL:
+ ad_info = {
+ '_type': 'url',
+ 'url': adServerURL,
+ }
+ if 'url' in info:
+ return {
+ '_type': 'playlist',
+ 'title': info['title'],
+ 'entries': [ad_info, info],
+ }
+ else:
+ return ad_info
+
+ if 'url' not in info and not info.get('formats'):
raise ExtractorError('Unable to extract video url for %s' % info['id'])
return info
Rapsys Git / youtubedl / blobdiff