import re
import struct
-from .common import InfoExtractor
from .adobepass import AdobePassIE
+from .common import InfoExtractor
from ..compat import (
compat_etree_fromstring,
+ compat_HTTPError,
compat_parse_qs,
compat_urllib_parse_urlparse,
compat_urlparse,
compat_xml_parse_error,
- compat_HTTPError,
)
from ..utils import (
- ExtractorError,
+ clean_html,
extract_attributes,
+ ExtractorError,
find_xpath_attr,
fix_xml_ampersands,
float_or_none,
- js_to_json,
int_or_none,
+ js_to_json,
+ mimetype2ext,
parse_iso8601,
smuggle_url,
+ str_or_none,
+ try_get,
unescapeHTML,
unsmuggle_url,
- update_url_query,
- clean_html,
- mimetype2ext,
UnsupportedError,
+ update_url_query,
+ url_or_none,
)
]
@classmethod
- def _build_brighcove_url(cls, object_str):
+ def _build_brightcove_url(cls, object_str):
"""
Build a Brightcove url from a xml string containing
<object class="BrightcoveExperience">{params}</object>
return cls._make_brightcove_url(params)
@classmethod
- def _build_brighcove_url_from_js(cls, object_js):
+ def _build_brightcove_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
).+?>\s*</object>''',
webpage)
if matches:
- return list(filter(None, [cls._build_brighcove_url(m) for m in matches]))
+ return list(filter(None, [cls._build_brightcove_url(m) for m in matches]))
matches = re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)
if matches:
return list(filter(None, [
- cls._build_brighcove_url_from_js(custom_bc)
+ cls._build_brightcove_url_from_js(custom_bc)
for custom_bc in matches]))
return [src for _, src in re.findall(
r'<iframe[^>]+src=([\'"])((?:https?:)?//link\.brightcove\.com/services/player/(?!\1).+)\1', webpage)]
# [2] looks like:
for video, script_tag, account_id, player_id, embed in re.findall(
r'''(?isx)
- (<video\s+[^>]*\bdata-video-id\s*=\s*['"]?[^>]+>)
+ (<video(?:-js)?\s+[^>]*\bdata-video-id\s*=\s*['"]?[^>]+>)
(?:.*?
(<script[^>]+
src=["\'](?:https?:)?//players\.brightcove\.net/
def _parse_brightcove_metadata(self, json_data, video_id, headers={}):
title = json_data['name'].strip()
+ num_drm_sources = 0
formats = []
- for source in json_data.get('sources', []):
+ sources = json_data.get('sources') or []
+ for source in 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'):
+ if container == 'WVM' or source.get('key_systems'):
+ num_drm_sources += 1
+ continue
+ elif ext == 'ism':
continue
elif ext == 'm3u8' or container == 'M2TS':
if not src:
'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)
+ if not formats:
+ errors = json_data.get('errors')
+ if errors:
+ error = errors[0]
+ raise ExtractorError(
+ error.get('message') or error.get('error_subcode') or error['error_code'], expected=True)
+ if sources and num_drm_sources == len(sources):
+ raise ExtractorError('This video is DRM protected.', expected=True)
self._sort_formats(formats)
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'],
- })
+ if text_track.get('kind') != 'captions':
+ continue
+ text_track_url = url_or_none(text_track.get('src'))
+ if not text_track_url:
+ continue
+ lang = (str_or_none(text_track.get('srclang'))
+ or str_or_none(text_track.get('label')) or 'en').lower()
+ subtitles.setdefault(lang, []).append({
+ 'url': text_track_url,
+ })
is_live = False
duration = float_or_none(json_data.get('duration'), 1000)
policy_key_id = '%s_%s' % (account_id, player_id)
policy_key = self._downloader.cache.load('brightcove', policy_key_id)
- if not policy_key:
- webpage = self._download_webpage(
- 'http://players.brightcove.net/%s/%s_%s/index.min.js'
- % (account_id, player_id, embed), video_id)
-
- catalog = self._search_regex(
- r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
- if catalog:
- catalog = self._parse_json(
- js_to_json(catalog), video_id, fatal=False)
+ policy_key_extracted = False
+ store_pk = lambda x: self._downloader.cache.store('brightcove', policy_key_id, x)
+
+ def extract_policy_key():
+ base_url = 'http://players.brightcove.net/%s/%s_%s/' % (account_id, player_id, embed)
+ config = self._download_json(
+ base_url + 'config.json', video_id, fatal=False) or {}
+ policy_key = try_get(
+ config, lambda x: x['video_cloud']['policy_key'])
+ if not policy_key:
+ webpage = self._download_webpage(
+ base_url + 'index.min.js', video_id)
+
+ catalog = self._search_regex(
+ r'catalog\(({.+?})\);', webpage, 'catalog', default=None)
if catalog:
- policy_key = catalog.get('policyKey')
+ 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')
- self._downloader.cache.store('brightcove', policy_key_id, policy_key)
+ 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 = {
- 'Accept': 'application/json;pk=%s' % policy_key,
- }
+ headers = {}
referrer = smuggled_data.get('referrer')
if referrer:
headers.update({
'Referer': referrer,
'Origin': re.search(r'https?://[^/]+', referrer).group(0),
})
- try:
- json_data = self._download_json(api_url, video_id, headers=headers)
- 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)[0]
- message = json_data.get('message') or json_data['error_code']
- if json_data.get('error_subcode') == 'CLIENT_GEO':
- self.raise_geo_restricted(msg=message)
- raise ExtractorError(message, expected=True)
- raise
+
+ for _ in range(2):
+ if not policy_key:
+ policy_key = extract_policy_key()
+ policy_key_extracted = True
+ headers['Accept'] = 'application/json;pk=%s' % policy_key
+ try:
+ json_data = self._download_json(api_url, video_id, headers=headers)
+ break
+ except ExtractorError as e:
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
+ json_data = self._parse_json(e.cause.read().decode(), video_id)[0]
+ message = json_data.get('message') or json_data['error_code']
+ if json_data.get('error_subcode') == 'CLIENT_GEO':
+ self.raise_geo_restricted(msg=message)
+ elif json_data.get('error_code') == 'INVALID_POLICY_KEY' and not policy_key_extracted:
+ policy_key = None
+ store_pk(None)
+ continue
+ raise ExtractorError(message, expected=True)
+ raise
errors = json_data.get('errors')
if errors and errors[0].get('error_subcode') == 'TVE_AUTH':