[ooyala:external] Add extractor

This commit is contained in:
Sergey M․ 2015-05-16 20:00:40 +06:00
parent 2cda13213d
commit 1c97b0a777
2 changed files with 137 additions and 89 deletions

View file

@ -376,7 +376,10 @@ from .nytimes import (
from .nuvid import NuvidIE from .nuvid import NuvidIE
from .odnoklassniki import OdnoklassnikiIE from .odnoklassniki import OdnoklassnikiIE
from .oktoberfesttv import OktoberfestTVIE from .oktoberfesttv import OktoberfestTVIE
from .ooyala import OoyalaIE from .ooyala import (
OoyalaIE,
OoyalaExternalIE,
)
from .openfilm import OpenFilmIE from .openfilm import OpenFilmIE
from .orf import ( from .orf import (
ORFTVthekIE, ORFTVthekIE,

View file

@ -12,7 +12,100 @@ from ..utils import (
) )
class OoyalaIE(InfoExtractor): class OoyalaBaseIE(InfoExtractor):
def _extract_result(self, info, more_info):
embedCode = info['embedCode']
video_url = info.get('ipad_url') or info['url']
if determine_ext(video_url) == 'm3u8':
formats = self._extract_m3u8_formats(video_url, embedCode, ext='mp4')
else:
formats = [{
'url': video_url,
'ext': 'mp4',
}]
return {
'id': embedCode,
'title': unescapeHTML(info['title']),
'formats': formats,
'description': unescapeHTML(more_info['description']),
'thumbnail': more_info['promo'],
}
def _extract(self, player_url, video_id):
player = self._download_webpage(player_url, video_id)
mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="',
player, 'mobile player url')
# Looks like some videos are only available for particular devices
# (e.g. http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0
# is only available for ipad)
# Working around with fetching URLs for all the devices found starting with 'unknown'
# until we succeed or eventually fail for each device.
devices = re.findall(r'device\s*=\s*"([^"]+)";', player)
devices.remove('unknown')
devices.insert(0, 'unknown')
for device in devices:
mobile_player = self._download_webpage(
'%s&device=%s' % (mobile_url, device), video_id,
'Downloading mobile player JS for %s device' % device)
videos_info = self._search_regex(
r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
mobile_player, 'info', fatal=False, default=None)
if videos_info:
break
if not videos_info:
formats = []
auth_data = self._download_json(
'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?domain=www.example.org&supportedFormats=mp4,webm' % (embedCode, embedCode),
video_id)
cur_auth_data = auth_data['authorization_data'][video_id]
for stream in cur_auth_data['streams']:
formats.append({
'url': base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8'),
'ext': stream.get('delivery_type'),
'format': stream.get('video_codec'),
'format_id': stream.get('profile'),
'width': int_or_none(stream.get('width')),
'height': int_or_none(stream.get('height')),
'abr': int_or_none(stream.get('audio_bitrate')),
'vbr': int_or_none(stream.get('video_bitrate')),
})
if formats:
return {
'id': video_id,
'formats': formats,
'title': 'Ooyala video',
}
if not cur_auth_data['authorized']:
raise ExtractorError(cur_auth_data['message'], expected=True)
if not videos_info:
raise ExtractorError('Unable to extract info')
videos_info = videos_info.replace('\\"', '"')
videos_more_info = self._search_regex(
r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"', '"')
videos_info = json.loads(videos_info)
videos_more_info = json.loads(videos_more_info)
if videos_more_info.get('lineup'):
videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]
return {
'_type': 'playlist',
'id': video_id,
'title': unescapeHTML(videos_more_info['title']),
'entries': videos,
}
else:
return self._extract_result(videos_info[0], videos_more_info)
class OoyalaIE(OoyalaBaseIE):
_VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)' _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'
_TESTS = [ _TESTS = [
@ -57,95 +150,47 @@ class OoyalaIE(InfoExtractor):
return cls.url_result(cls._url_for_embed_code(embed_code), return cls.url_result(cls._url_for_embed_code(embed_code),
ie=cls.ie_key()) ie=cls.ie_key())
def _extract_result(self, info, more_info): def _real_extract(self, url):
embedCode = info['embedCode'] embed_code = self._match_id(url)
video_url = info.get('ipad_url') or info['url'] player_url = 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
return self._extract(player_url, embed_code)
if determine_ext(video_url) == 'm3u8':
formats = self._extract_m3u8_formats(video_url, embedCode, ext='mp4')
else:
formats = [{
'url': video_url,
'ext': 'mp4',
}]
return { class OoyalaExternalIE(OoyalaBaseIE):
'id': embedCode, _VALID_URL = r'''(?x)
'title': unescapeHTML(info['title']), (?:
'formats': formats, ooyalaexternal:|
'description': unescapeHTML(more_info['description']), https?://.+?\.ooyala\.com/.*?\bexternalId=
'thumbnail': more_info['promo'], )
} (?P<partner_id>[^:]+)
:
(?P<id>.+)
(?:
:|
.*?&pcode=
)
(?P<pcode>.+?)
(&|$)
'''
_TEST = {
'url': 'https://player.ooyala.com/player.js?externalId=espn:10365079&pcode=1kNG061cgaoolOncv54OAO1ceO-I&adSetCode=91cDU6NuXTGKz3OdjOxFdAgJVtQcKJnI&callback=handleEvents&hasModuleParams=1&height=968&playerBrandingId=7af3bd04449c444c964f347f11873075&targetReplaceId=videoPlayer&width=1656&wmode=opaque&allowScriptAccess=always',
'info_dict': {
'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
'ext': 'mp4',
'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
'description': '',
},
'params': {
# m3u8 download
'skip_download': True,
},
}
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
embedCode = mobj.group('id') partner_id = mobj.group('partner_id')
player_url = 'http://player.ooyala.com/player.js?embedCode=%s' % embedCode video_id = mobj.group('id')
player = self._download_webpage(player_url, embedCode) pcode = mobj.group('pcode')
mobile_url = self._search_regex(r'mobile_player_url="(.+?)&device="', player_url = 'http://player.ooyala.com/player.js?externalId=%s:%s&pcode=%s' % (partner_id, video_id, pcode)
player, 'mobile player url') return self._extract(player_url, video_id)
# Looks like some videos are only available for particular devices
# (e.g. http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0
# is only available for ipad)
# Working around with fetching URLs for all the devices found starting with 'unknown'
# until we succeed or eventually fail for each device.
devices = re.findall(r'device\s*=\s*"([^"]+)";', player)
devices.remove('unknown')
devices.insert(0, 'unknown')
for device in devices:
mobile_player = self._download_webpage(
'%s&device=%s' % (mobile_url, device), embedCode,
'Downloading mobile player JS for %s device' % device)
videos_info = self._search_regex(
r'var streams=window.oo_testEnv\?\[\]:eval\("\((\[{.*?}\])\)"\);',
mobile_player, 'info', fatal=False, default=None)
if videos_info:
break
if not videos_info:
formats = []
auth_data = self._download_json(
'http://player.ooyala.com/sas/player_api/v1/authorization/embed_code/%s/%s?domain=www.example.org&supportedFormats=mp4,webm' % (embedCode, embedCode),
embedCode)
cur_auth_data = auth_data['authorization_data'][embedCode]
for stream in cur_auth_data['streams']:
formats.append({
'url': base64.b64decode(stream['url']['data'].encode('ascii')).decode('utf-8'),
'ext': stream.get('delivery_type'),
'format': stream.get('video_codec'),
'format_id': stream.get('profile'),
'width': int_or_none(stream.get('width')),
'height': int_or_none(stream.get('height')),
'abr': int_or_none(stream.get('audio_bitrate')),
'vbr': int_or_none(stream.get('video_bitrate')),
})
if formats:
return {
'id': embedCode,
'formats': formats,
'title': 'Ooyala video',
}
if not cur_auth_data['authorized']:
raise ExtractorError(cur_auth_data['message'], expected=True)
if not videos_info:
raise ExtractorError('Unable to extract info')
videos_info = videos_info.replace('\\"', '"')
videos_more_info = self._search_regex(
r'eval\("\(({.*?\\"promo\\".*?})\)"', mobile_player, 'more info').replace('\\"', '"')
videos_info = json.loads(videos_info)
videos_more_info = json.loads(videos_more_info)
if videos_more_info.get('lineup'):
videos = [self._extract_result(info, more_info) for (info, more_info) in zip(videos_info, videos_more_info['lineup'])]
return {
'_type': 'playlist',
'id': embedCode,
'title': unescapeHTML(videos_more_info['title']),
'entries': videos,
}
else:
return self._extract_result(videos_info[0], videos_more_info)