[brightcove] some improvements and fixes

- use FFmpeg downloader to download m3u8 formats extracted
from BrightcoveNew(some of the m3u8 media playlists use AES-128)
- update comment and update_url_query to handle url query
This commit is contained in:
remitamine 2016-03-16 09:14:37 +01:00
parent 2cab48704c
commit 48254c3f2c
1 changed files with 14 additions and 18 deletions

View File

@ -24,16 +24,16 @@ from ..utils import (
js_to_json, js_to_json,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
sanitized_Request,
unescapeHTML, unescapeHTML,
unsmuggle_url, unsmuggle_url,
update_url_query,
) )
class BrightcoveLegacyIE(InfoExtractor): class BrightcoveLegacyIE(InfoExtractor):
IE_NAME = 'brightcove:legacy' IE_NAME = 'brightcove:legacy'
_VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)' _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 = [ _TESTS = [
{ {
@ -156,7 +156,7 @@ class BrightcoveLegacyIE(InfoExtractor):
# Not all pages define this value # Not all pages define this value
if playerKey is not None: if playerKey is not None:
params['playerKey'] = playerKey params['playerKey'] = playerKey
# The three fields hold the id of the video # These fields hold the id of the video
videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList') videoPlayer = find_param('@videoPlayer') or find_param('videoId') or find_param('videoID') or find_param('@videoList')
if videoPlayer is not None: if videoPlayer is not None:
params['@videoPlayer'] = videoPlayer params['@videoPlayer'] = videoPlayer
@ -185,8 +185,7 @@ class BrightcoveLegacyIE(InfoExtractor):
@classmethod @classmethod
def _make_brightcove_url(cls, params): def _make_brightcove_url(cls, params):
data = compat_urllib_parse.urlencode(params) return update_url_query(cls._FEDERATED_URL, params)
return cls._FEDERATED_URL_TEMPLATE % data
@classmethod @classmethod
def _extract_brightcove_url(cls, webpage): def _extract_brightcove_url(cls, webpage):
@ -240,7 +239,7 @@ class BrightcoveLegacyIE(InfoExtractor):
# We set the original url as the default 'Referer' header # We set the original url as the default 'Referer' header
referer = smuggled_data.get('Referer', url) referer = smuggled_data.get('Referer', url)
return self._get_video_info( return self._get_video_info(
videoPlayer[0], query_str, query, referer=referer) videoPlayer[0], query, referer=referer)
elif 'playerKey' in query: elif 'playerKey' in query:
player_key = query['playerKey'] player_key = query['playerKey']
return self._get_playlist_info(player_key[0]) return self._get_playlist_info(player_key[0])
@ -249,15 +248,14 @@ class BrightcoveLegacyIE(InfoExtractor):
'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?', 'Cannot find playerKey= variable. Did you forget quotes in a shell invocation?',
expected=True) expected=True)
def _get_video_info(self, video_id, query_str, query, referer=None): def _get_video_info(self, video_id, query, referer=None):
request_url = self._FEDERATED_URL_TEMPLATE % query_str headers = {}
req = sanitized_Request(request_url)
linkBase = query.get('linkBaseURL') linkBase = query.get('linkBaseURL')
if linkBase is not None: if linkBase is not None:
referer = linkBase[0] referer = linkBase[0]
if referer is not None: if referer is not None:
req.add_header('Referer', referer) headers['Referer'] = referer
webpage = self._download_webpage(req, video_id) webpage = self._download_webpage(self._FEDERATED_URL, video_id, headers=headers, query=query)
error_msg = self._html_search_regex( error_msg = self._html_search_regex(
r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage, r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
@ -459,12 +457,11 @@ class BrightcoveNewIE(InfoExtractor):
r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1', r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
webpage, 'policy key', group='pk') webpage, 'policy key', group='pk')
req = sanitized_Request( api_url = 'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s' % (account_id, video_id)
'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s'
% (account_id, video_id),
headers={'Accept': 'application/json;pk=%s' % policy_key})
try: try:
json_data = self._download_json(req, video_id) json_data = self._download_json(api_url, video_id, headers={
'Accept': 'application/json;pk=%s' % policy_key
})
except ExtractorError as e: except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
json_data = self._parse_json(e.cause.read().decode(), video_id) json_data = self._parse_json(e.cause.read().decode(), video_id)
@ -482,8 +479,7 @@ class BrightcoveNewIE(InfoExtractor):
if not src: if not src:
continue continue
formats.extend(self._extract_m3u8_formats( formats.extend(self._extract_m3u8_formats(
src, video_id, 'mp4', entry_protocol='m3u8_native', src, video_id, 'mp4', m3u8_id='hls', fatal=False))
m3u8_id='hls', fatal=False))
elif source_type == 'application/dash+xml': elif source_type == 'application/dash+xml':
if not src: if not src:
continue continue