1
1
mirror of https://github.com/ytdl-org/youtube-dl synced 2024-11-27 23:56:51 +01:00

[brightcove:embedinpage] Make more robust and extract rtmp streams

This commit is contained in:
Sergey M․ 2015-11-14 05:09:50 +06:00
parent a2973eb597
commit a662489877

View File

@ -23,6 +23,7 @@ from ..utils import (
unescapeHTML, unescapeHTML,
unsmuggle_url, unsmuggle_url,
js_to_json, js_to_json,
float_or_none,
int_or_none, int_or_none,
parse_iso8601, parse_iso8601,
extract_attributes, extract_attributes,
@ -353,7 +354,7 @@ class BrightcoveIE(InfoExtractor):
class BrightcoveInPageEmbedIE(InfoExtractor): class BrightcoveInPageEmbedIE(InfoExtractor):
_VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/([a-z0-9-]+)_([a-z]+)/index.html?.*videoId=(?P<video_id>\d+)' _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[\da-f-]+)_(?P<embed>[a-z]+)/index\.html\?.*videoId=(?P<video_id>\d+)'
_TEST = { _TEST = {
'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001', 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
'md5': 'c8100925723840d4b0d243f7025703be', 'md5': 'c8100925723840d4b0d243f7025703be',
@ -385,59 +386,99 @@ class BrightcoveInPageEmbedIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups() 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) 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( catalog = self._parse_json(
js_to_json( js_to_json(catalog), video_id, fatal=False)
self._search_regex( if catalog:
r'catalog\(({[^}]+})\);', policy_key = catalog.get('policyKey')
webpage,
'catalog' if not policy_key:
) policy_key = self._search_regex(
), r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
video_id webpage, 'policy key', group='pk')
)
policy_key = catalog['policyKey']
req = compat_urllib_request.Request( req = compat_urllib_request.Request(
'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}) headers={'Accept': 'application/json;pk=%s' % policy_key})
json_data = self._download_json(req, video_id) json_data = self._download_json(req, video_id)
title = json_data['name'] title = json_data['name']
description = json_data.get('description')
thumbnail = json_data.get('thumbnail')
timestamp = parse_iso8601(json_data.get('published_at'))
duration = int_or_none(json_data.get('duration'))
formats = [] formats = []
for source in json_data.get('sources'): for source in json_data.get('sources', []):
source_type = source.get('type') source_type = source.get('type')
src = source.get('src')
if source_type == 'application/x-mpegURL': if source_type == 'application/x-mpegURL':
formats.extend(self._extract_m3u8_formats(source.get('src'), video_id)) if not src:
continue
m3u8_formats = self._extract_m3u8_formats(
src, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False)
if m3u8_formats:
formats.extend(m3u8_formats)
else: else:
src = source.get('src') or source.get('streaming_src') streaming_src = source.get('streaming_src')
if src: stream_name, app_name = source.get('stream_name'), source.get('app_name')
formats.append({ if not src and not streaming_src and (not stream_name or not app_name):
'url': src, continue
'tbr': source.get('avg_bitrate'), tbr = float_or_none(source.get('avg_bitrate'), 1000)
height = int_or_none(source.get('height'))
f = {
'tbr': tbr,
'width': int_or_none(source.get('width')), 'width': int_or_none(source.get('width')),
'height': int_or_none(source.get('height')), 'height': height,
'filesize': source.get('size'), 'filesize': int_or_none(source.get('size')),
'container': source.get('container'), 'container': source.get('container'),
'vcodec': source.get('codec'), 'vcodec': source.get('codec'),
'ext': source.get('container').lower(), 'ext': source.get('container').lower(),
}) }
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'),
'preference': 2 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) 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 { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'description': description, 'description': description,
'thumbnail': thumbnail, 'thumbnail': thumbnail,
'timestamp': timestamp,
'duration': duration, 'duration': duration,
'formats': formats, 'timestamp': timestamp,
'uploader_id': account_id, 'uploader_id': account_id,
'formats': formats,
'tags': tags,
} }