import re
import json
-import xml.etree.ElementTree
from .common import InfoExtractor
from ..compat import (
+ compat_etree_fromstring,
compat_parse_qs,
compat_str,
- compat_urllib_parse,
compat_urllib_parse_urlparse,
- compat_urllib_request,
compat_urlparse,
+ compat_xml_parse_error,
+ compat_HTTPError,
)
from ..utils import (
determine_ext,
ExtractorError,
find_xpath_attr,
fix_xml_ampersands,
+ float_or_none,
+ js_to_json,
+ int_or_none,
+ parse_iso8601,
unescapeHTML,
unsmuggle_url,
+ update_url_query,
)
-class BrightcoveIE(InfoExtractor):
+class BrightcoveLegacyIE(InfoExtractor):
+ IE_NAME = 'brightcove:legacy'
_VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)'
- _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
+ _FEDERATED_URL = 'http://c.brightcove.com/services/viewer/htmlFederated'
_TESTS = [
{
'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',
+ 'timestamp': 1368213670,
+ 'upload_date': '20130510',
+ 'uploader_id': 1589608506001,
}
},
{
'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
'uploader': 'Oracle',
+ 'timestamp': 1344975024,
+ 'upload_date': '20120814',
+ 'uploader_id': 1460825906,
},
},
{
'title': 'This Bracelet Acts as a Personal Thermostat',
'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
'uploader': 'Mashable',
+ 'timestamp': 1382041798,
+ 'upload_date': '20131017',
+ 'uploader_id': 1130468786001,
},
},
{
{
# 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',
+ 'url': 'http://c.brightcove.com/services/viewer/htmlFederated?%40videoPlayer=ref%3Aevent-stream-356&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',
+ 'id': '3750436379001',
'ext': 'flv',
'title': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
- 'uploader': 'Red Bull TV',
+ 'uploader': 'RBTV Old (do not use)',
'description': 'UCI MTB World Cup 2014: Fort William, UK - Downhill Finals',
+ 'timestamp': 1409122195,
+ 'upload_date': '20140827',
+ 'uploader_id': 710858724001,
},
},
{
'playlist_mincount': 7,
},
]
+ FLV_VCODECS = {
+ 1: 'SORENSON',
+ 2: 'ON2',
+ 3: 'H264',
+ 4: 'VP8',
+ }
@classmethod
def _build_brighcove_url(cls, object_str):
object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str)
object_str = fix_xml_ampersands(object_str)
- object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
+ try:
+ object_doc = compat_etree_fromstring(object_str.encode('utf-8'))
+ except compat_xml_parse_error:
+ return
fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
if fv_el is not None:
else:
flashvars = {}
+ data_url = object_doc.attrib.get('data', '')
+ data_url_params = compat_parse_qs(compat_urllib_parse_urlparse(data_url).query)
+
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
+ return data_url_params.get(name)
params = {}
# Not all pages define this value
if playerKey is not None:
params['playerKey'] = playerKey
- # The three fields hold the id of the video
- videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID')
+ # These fields hold the id of the video
+ videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
if videoPlayer is not None:
params['@videoPlayer'] = videoPlayer
linkBase = find_param('linkBaseURL')
if linkBase is not None:
params['linkBaseURL'] = linkBase
- data = compat_urllib_parse.urlencode(params)
- return cls._FEDERATED_URL_TEMPLATE % data
+ return cls._make_brightcove_url(params)
+
+ @classmethod
+ def _build_brighcove_url_from_js(cls, object_js):
+ # The layout of JS is as follows:
+ # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) {
+ # // build Brightcove <object /> XML
+ # }
+ m = re.search(
+ 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
+ # in length, however it's appended to itself
+ # in places, so truncate
+ ["\'](?P<videoID>\d+)["\'] # @videoPlayer
+ ''', object_js)
+ if m:
+ return cls._make_brightcove_url(m.groupdict())
+
+ @classmethod
+ def _make_brightcove_url(cls, params):
+ return update_url_query(cls._FEDERATED_URL, params)
@classmethod
def _extract_brightcove_url(cls, webpage):
"""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/[^"]+)"',
+ r'<meta\s+property=[\'"]og:video[\'"]\s+content=[\'"](https?://(?:secure|c)\.brightcove.com/[^\'"]+)[\'"]',
webpage)
if url_m:
url = unescapeHTML(url_m.group(1))
[^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
).+?>\s*</object>''',
webpage)
- return [cls._build_brighcove_url(m) for m in matches]
+ if matches:
+ return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
+
+ return list(filter(None, [
+ cls._build_brighcove_url_from_js(custom_bc)
+ for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)]))
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
# We set the original url as the default 'Referer' header
referer = smuggled_data.get('Referer', url)
return self._get_video_info(
- videoPlayer[0], query_str, query, referer=referer)
+ videoPlayer[0], query, referer=referer)
elif 'playerKey' in query:
player_key = query['playerKey']
return self._get_playlist_info(player_key[0])
'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
expected=True)
- def _get_video_info(self, video_id, query_str, query, referer=None):
- request_url = self._FEDERATED_URL_TEMPLATE % query_str
- req = compat_urllib_request.Request(request_url)
+ def _get_video_info(self, video_id, query, referer=None):
+ headers = {}
linkBase = query.get('linkBaseURL')
if linkBase is not None:
referer = linkBase[0]
if referer is not None:
- req.add_header('Referer', referer)
- webpage = self._download_webpage(req, video_id)
+ headers['Referer'] = referer
+ webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query)
error_msg = self._html_search_regex(
r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
'description': video_info.get('shortDescription'),
'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
'uploader': video_info.get('publisherName'),
+ 'uploader_id': video_info.get('publisherId'),
+ 'duration': float_or_none(video_info.get('length'), 1000),
+ 'timestamp': int_or_none(video_info.get('creationDate'), 1000),
}
- renditions = video_info.get('renditions')
+ renditions = video_info.get('renditions', []) + video_info.get('IOSRenditions', [])
if renditions:
formats = []
for rend in renditions:
ext = 'flv'
if ext is None:
ext = determine_ext(url)
- size = rend.get('size')
- formats.append({
+ tbr = int_or_none(rend.get('encodingRate'), 1000),
+ a_format = {
+ 'format_id': 'http%s' % ('-%s' % tbr if tbr else ''),
'url': url,
'ext': ext,
- 'height': rend.get('frameHeight'),
- 'width': rend.get('frameWidth'),
- 'filesize': size if size != 0 else None,
- })
+ 'filesize': int_or_none(rend.get('size')) or None,
+ 'tbr': tbr,
+ }
+ if rend.get('audioOnly'):
+ a_format.update({
+ 'vcodec': 'none',
+ })
+ else:
+ a_format.update({
+ 'height': int_or_none(rend.get('frameHeight')),
+ 'width': int_or_none(rend.get('frameWidth')),
+ 'vcodec': rend.get('videoCodec'),
+ })
+
+ # m3u8 manifests with remote == false are media playlists
+ # Not calling _extract_m3u8_formats here to save network traffic
+ if ext == 'm3u8':
+ a_format.update({
+ 'format_id': 'hls%s' % ('-%s' % tbr if tbr else ''),
+ 'ext': 'mp4',
+ 'protocol': 'm3u8',
+ })
+
+ formats.append(a_format)
self._sort_formats(formats)
info['formats'] = formats
elif video_info.get('FLVFullLengthURL') is not None:
info.update({
'url': video_info['FLVFullLengthURL'],
+ 'vcodec': self.FLV_VCODECS.get(video_info.get('FLVFullCodec')),
+ 'filesize': int_or_none(video_info.get('FLVFullSize')),
})
if self._downloader.params.get('include_ads', False):
if 'url' not in info and not info.get('formats'):
raise ExtractorError('Unable to extract video url for %s' % info['id'])
return info
+
+
+class BrightcoveNewIE(InfoExtractor):
+ IE_NAME = 'brightcove:new'
+ _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+|ref:[^&]+)'
+ _TESTS = [{
+ 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
+ 'md5': 'c8100925723840d4b0d243f7025703be',
+ 'info_dict': {
+ 'id': '4463358922001',
+ 'ext': 'mp4',
+ 'title': 'Meet the man behind Popcorn Time',
+ 'description': 'md5:eac376a4fe366edc70279bfb681aea16',
+ 'duration': 165.768,
+ 'timestamp': 1441391203,
+ 'upload_date': '20150904',
+ 'uploader_id': '929656772001',
+ 'formats': 'mincount:22',
+ },
+ }, {
+ # with rtmp streams
+ 'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001',
+ 'info_dict': {
+ 'id': '4279049078001',
+ 'ext': 'mp4',
+ 'title': 'Titansgrave: Chapter 0',
+ 'description': 'Titansgrave: Chapter 0',
+ 'duration': 1242.058,
+ 'timestamp': 1433556729,
+ 'upload_date': '20150606',
+ 'uploader_id': '4036320279001',
+ 'formats': 'mincount:41',
+ },
+ 'params': {
+ # m3u8 download
+ 'skip_download': True,
+ }
+ }, {
+ # ref: prefixed video id
+ 'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442',
+ 'only_matching': True,
+ }, {
+ # non numeric ref: prefixed video id
+ 'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
+ 'only_matching': True,
+ }]
+
+ @staticmethod
+ def _extract_url(webpage):
+ urls = BrightcoveNewIE._extract_urls(webpage)
+ return urls[0] if urls else None
+
+ @staticmethod
+ def _extract_urls(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#setvideousingjavascript
+ # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
+ # 4. 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 embed_in_page embeds [2]
+ for video_id, account_id, player_id, embed in re.findall(
+ # According to examples from [3] it's unclear whether video id
+ # may be optional and what to do when it is
+ # According to [4] data-video-id may be prefixed with ref:
+ r'''(?sx)
+ <video[^>]+
+ data-video-id=["\'](\d+|ref:[^"\']+)["\'][^>]*>.*?
+ </video>.*?
+ <script[^>]+
+ src=["\'](?:https?:)?//players\.brightcove\.net/
+ (\d+)/([^/]+)_([^/]+)/index(?:\.min)?\.js
+ ''', webpage):
+ entries.append(
+ 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
+ % (account_id, player_id, embed, video_id))
+
+ return entries
+
+ def _real_extract(self, url):
+ account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
+
+ 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')
+
+ api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' % (account_id, video_id)
+ try:
+ json_data = self._download_json(api_url, video_id, headers={
+ 'Accept': 'application/json;pk=%s' % policy_key
+ })
+ except ExtractorError as e:
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+ json_data = self._parse_json(e.cause.read().decode(), video_id)
+ raise ExtractorError(json_data[0]['message'], expected=True)
+ raise
+
+ title = json_data['name']
+
+ formats = []
+ for source in json_data.get('sources', []):
+ container = source.get('container')
+ source_type = source.get('type')
+ src = source.get('src')
+ if source_type == 'application/x-mpegURL' or container == 'M2TS':
+ if not src:
+ continue
+ formats.extend(self._extract_m3u8_formats(
+ src, video_id, 'mp4', m3u8_id='hls', fatal=False))
+ elif source_type == 'application/dash+xml':
+ 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': 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)
+ self._sort_formats(formats)
+
+ description = json_data.get('description')
+ thumbnail = json_data.get('thumbnail')
+ timestamp = parse_iso8601(json_data.get('published_at'))
+ duration = float_or_none(json_data.get('duration'), 1000)
+ tags = json_data.get('tags', [])
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'description': description,
+ 'thumbnail': thumbnail,
+ 'duration': duration,
+ 'timestamp': timestamp,
+ 'uploader_id': account_id,
+ 'formats': formats,
+ 'tags': tags,
+ }
projects / youtube-dl / blobdiff