[yt-dlp.git] / yt_dlp / extractor / ooyala.py

import base64
import re

from .common import InfoExtractor
from ..compat import (
    compat_b64decode,
    compat_str,
)
from ..utils import (
    determine_ext,
    float_or_none,
    int_or_none,
    smuggle_url,
    try_get,
    unsmuggle_url,
)


class OoyalaBaseIE(InfoExtractor):
    _PLAYER_BASE = 'http://player.ooyala.com/'
    _CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
    _AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s'

    def _extract(self, content_tree_url, video_id, domain=None, supportedformats=None, embed_token=None):
        content_tree = self._download_json(content_tree_url, video_id)['content_tree']
        metadata = content_tree[list(content_tree)[0]]
        embed_code = metadata['embed_code']
        pcode = metadata.get('asset_pcode') or embed_code
        title = metadata['title']

        auth_data = self._download_json(
            self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code),
            video_id, headers=self.geo_verification_headers(), query={
                'domain': domain or 'player.ooyala.com',
                'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
                'embedToken': embed_token,
            })['authorization_data'][embed_code]

        urls = []
        formats = []
        streams = auth_data.get('streams') or [{
            'delivery_type': 'hls',
            'url': {
                'data': base64.b64encode(('http://player.ooyala.com/hls/player/all/%s.m3u8' % embed_code).encode()).decode(),
            }
        }]
        for stream in streams:
            url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
            if not url_data:
                continue
            s_url = compat_b64decode(url_data).decode('utf-8')
            if not s_url or s_url in urls:
                continue
            urls.append(s_url)
            ext = determine_ext(s_url, None)
            delivery_type = stream.get('delivery_type')
            if delivery_type == 'hls' or ext == 'm3u8':
                formats.extend(self._extract_m3u8_formats(
                    re.sub(r'/ip(?:ad|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
                    m3u8_id='hls', fatal=False))
            elif delivery_type == 'hds' or ext == 'f4m':
                formats.extend(self._extract_f4m_formats(
                    s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
            elif delivery_type == 'dash' or ext == 'mpd':
                formats.extend(self._extract_mpd_formats(
                    s_url, embed_code, mpd_id='dash', fatal=False))
            elif delivery_type == 'smooth':
                self._extract_ism_formats(
                    s_url, embed_code, ism_id='mss', fatal=False)
            elif ext == 'smil':
                formats.extend(self._extract_smil_formats(
                    s_url, embed_code, fatal=False))
            else:
                formats.append({
                    'url': s_url,
                    'ext': ext or delivery_type,
                    'vcodec': stream.get('video_codec'),
                    'format_id': delivery_type,
                    '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')),
                    'fps': float_or_none(stream.get('framerate')),
                })
        if not formats and not auth_data.get('authorized'):
            self.raise_no_formats('%s said: %s' % (
                self.IE_NAME, auth_data['message']), expected=True)

        subtitles = {}
        for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
            sub_url = sub.get('url')
            if not sub_url:
                continue
            subtitles[lang] = [{
                'url': sub_url,
            }]

        return {
            'id': embed_code,
            'title': title,
            'description': metadata.get('description'),
            'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
            'duration': float_or_none(metadata.get('duration'), 1000),
            'subtitles': subtitles,
            'formats': formats,
        }


class OoyalaIE(OoyalaBaseIE):
    _VALID_URL = r'(?:ooyala:|https?://.+?\.ooyala\.com/.*?(?:embedCode|ec)=)(?P<id>.+?)(&|$)'

    _TESTS = [
        {
            # From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
            'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
            'info_dict': {
                'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
                'ext': 'mp4',
                'title': 'Explaining Data Recovery from Hard Drives and SSDs',
                'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
                'duration': 853.386,
            },
            # The video in the original webpage now uses PlayWire
            'skip': 'Ooyala said: movie expired',
        }, {
            # Only available for ipad
            'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
            'info_dict': {
                'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
                'ext': 'mp4',
                'title': 'Simulation Overview - Levels of Simulation',
                'duration': 194.948,
            },
        },
        {
            # Information available only through SAS api
            # From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
            'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
            'md5': 'a84001441b35ea492bc03736e59e7935',
            'info_dict': {
                'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
                'ext': 'mp4',
                'title': 'Divide Tool Path.mp4',
                'duration': 204.405,
            }
        },
        {
            # empty stream['url']['data']
            'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
            'only_matching': True,
        }
    ]

    def _extract_from_webpage(self, url, webpage):
        mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage)
                or re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage)
                or re.search(r'OO\.Player\.create\.apply\(\s*OO\.Player\s*,\s*op\(\s*\[\s*[\'"][^\'"]*[\'"]\s*,\s*[\'"](?P<ec>.{32})[\'"]', webpage)
                or re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage)
                or re.search(r'data-ooyala-video-id\s*=\s*[\'"](?P<ec>.{32})[\'"]', webpage))
        if mobj is not None:
            embed_token = self._search_regex(
                r'embedToken[\'"]?\s*:\s*[\'"]([^\'"]+)',
                webpage, 'ooyala embed token', default=None)
            yield self._build_url_result(smuggle_url(
                mobj.group('ec'), {
                    'domain': url,
                    'embed_token': embed_token,
                }))
            return

        # Look for multiple Ooyala embeds on SBN network websites
        mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
        if mobj is not None:
            for v in self._parse_json(mobj.group(1), self._generic_id(url), fatal=False) or []:
                yield self._build_url_result(smuggle_url(v['provider_video_id'], {'domain': url}))

    @staticmethod
    def _url_for_embed_code(embed_code):
        return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code

    @classmethod
    def _build_url_result(cls, embed_code):
        return cls.url_result(cls._url_for_embed_code(embed_code),
                              ie=cls.ie_key())

    def _real_extract(self, url):
        url, smuggled_data = unsmuggle_url(url, {})
        embed_code = self._match_id(url)
        domain = smuggled_data.get('domain')
        supportedformats = smuggled_data.get('supportedformats')
        embed_token = smuggled_data.get('embed_token')
        content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
        return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)


class OoyalaExternalIE(OoyalaBaseIE):
    _VALID_URL = r'''(?x)
                    (?:
                        ooyalaexternal:|
                        https?://.+?\.ooyala\.com/.*?\bexternalId=
                    )
                    (?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',
            'duration': 1302.0,
        },
        'params': {
            # m3u8 download
            'skip_download': True,
        },
    }

    def _real_extract(self, url):
        partner_id, video_id, pcode = self._match_valid_url(url).groups()
        content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
        return self._extract(content_tree_url, video_id)
Commit	Line	Data
1ed2c4b3	1	import base64
09825cb5	2	import re
09825cb5 JMF	3
09825cb5 JMF	4	from .common import InfoExtractor
cf282071 S	5	from ..compat import (
	6	compat_b64decode,
	7	compat_str,
cf282071	8	)
6f600ff5	9	from ..utils import (
4f4dd8d7	10	determine_ext,
4f4dd8d7 S	11	float_or_none,
4f4dd8d7 S	12	int_or_none,
bfd973ec	13	smuggle_url,
4f4dd8d7	14	try_get,
cce9d15d	15	unsmuggle_url,
6f600ff5	16	)
09825cb5	17
e24b5a86	18
1c97b0a7	19	class OoyalaBaseIE(InfoExtractor):
6101f45e	20	_PLAYER_BASE = 'http://player.ooyala.com/'
6101f45e	21	_CONTENT_TREE_BASE = _PLAYER_BASE + 'player_api/v1/content_tree/'
1ed2c4b3	22	_AUTHORIZATION_URL_TEMPLATE = _PLAYER_BASE + 'sas/player_api/v2/authorization/embed_code/%s/%s'
c0d0b01f	23
1ed2c4b3	24	def _extract(self, content_tree_url, video_id, domain=None, supportedformats=None, embed_token=None):
cce9d15d	25	content_tree = self._download_json(content_tree_url, video_id)['content_tree']
90bddb6c	26	metadata = content_tree[list(content_tree)[0]]
	27	embed_code = metadata['embed_code']
	28	pcode = metadata.get('asset_pcode') or embed_code
e8593f34	29	title = metadata['title']
09825cb5	30
a4760d20	31	auth_data = self._download_json(
1ed2c4b3 RA	32	self._AUTHORIZATION_URL_TEMPLATE % (pcode, embed_code),
	33	video_id, headers=self.geo_verification_headers(), query={
	34	'domain': domain or 'player.ooyala.com',
9837cb75 RA	35	'supportedFormats': supportedformats or 'mp4,rtmp,m3u8,hds,dash,smooth',
9837cb75 RA	36	'embedToken': embed_token,
1ed2c4b3	37	})['authorization_data'][embed_code]
aafe2739	38
a4760d20	39	urls = []
a4760d20	40	formats = []
1ed2c4b3 RA	41	streams = auth_data.get('streams') or [{
	42	'delivery_type': 'hls',
	43	'url': {
	44	'data': base64.b64encode(('http://player.ooyala.com/hls/player/all/%s.m3u8' % embed_code).encode()).decode(),
	45	}
	46	}]
	47	for stream in streams:
	48	url_data = try_get(stream, lambda x: x['url']['data'], compat_str)
	49	if not url_data:
	50	continue
	51	s_url = compat_b64decode(url_data).decode('utf-8')
	52	if not s_url or s_url in urls:
	53	continue
	54	urls.append(s_url)
	55	ext = determine_ext(s_url, None)
	56	delivery_type = stream.get('delivery_type')
	57	if delivery_type == 'hls' or ext == 'm3u8':
	58	formats.extend(self._extract_m3u8_formats(
	59	re.sub(r'/ip(?:ad\|hone)/', '/all/', s_url), embed_code, 'mp4', 'm3u8_native',
	60	m3u8_id='hls', fatal=False))
	61	elif delivery_type == 'hds' or ext == 'f4m':
	62	formats.extend(self._extract_f4m_formats(
	63	s_url + '?hdcore=3.7.0', embed_code, f4m_id='hds', fatal=False))
	64	elif delivery_type == 'dash' or ext == 'mpd':
	65	formats.extend(self._extract_mpd_formats(
	66	s_url, embed_code, mpd_id='dash', fatal=False))
	67	elif delivery_type == 'smooth':
	68	self._extract_ism_formats(
	69	s_url, embed_code, ism_id='mss', fatal=False)
	70	elif ext == 'smil':
	71	formats.extend(self._extract_smil_formats(
	72	s_url, embed_code, fatal=False))
	73	else:
	74	formats.append({
	75	'url': s_url,
	76	'ext': ext or delivery_type,
	77	'vcodec': stream.get('video_codec'),
	78	'format_id': delivery_type,
	79	'width': int_or_none(stream.get('width')),
	80	'height': int_or_none(stream.get('height')),
	81	'abr': int_or_none(stream.get('audio_bitrate')),
	82	'vbr': int_or_none(stream.get('video_bitrate')),
	83	'fps': float_or_none(stream.get('framerate')),
	84	})
	85	if not formats and not auth_data.get('authorized'):
b7da73eb	86	self.raise_no_formats('%s said: %s' % (
1ed2c4b3	87	self.IE_NAME, auth_data['message']), expected=True)
90bddb6c	88
e8593f34	89	subtitles = {}
	90	for lang, sub in metadata.get('closed_captions_vtt', {}).get('captions', {}).items():
	91	sub_url = sub.get('url')
	92	if not sub_url:
	93	continue
	94	subtitles[lang] = [{
	95	'url': sub_url,
	96	}]
	97
	98	return {
	99	'id': embed_code,
	100	'title': title,
	101	'description': metadata.get('description'),
	102	'thumbnail': metadata.get('thumbnail_image') or metadata.get('promo_image'),
	103	'duration': float_or_none(metadata.get('duration'), 1000),
	104	'subtitles': subtitles,
	105	'formats': formats,
	106	}
1c97b0a7 S	107
	108
	109	class OoyalaIE(OoyalaBaseIE):
	110	_VALID_URL = r'(?:ooyala:\|https?://.+?\.ooyala\.com/.*?(?:embedCode\|ec)=)(?P<id>.+?)(&\|$)'
	111
	112	_TESTS = [
	113	{
	114	# From http://it.slashdot.org/story/13/04/25/178216/recovering-data-from-broken-hard-drives-and-ssds-video
	115	'url': 'http://player.ooyala.com/player.js?embedCode=pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
	116	'info_dict': {
	117	'id': 'pxczE2YjpfHfn1f3M-ykG_AmJRRn0PD8',
	118	'ext': 'mp4',
	119	'title': 'Explaining Data Recovery from Hard Drives and SSDs',
	120	'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
53e06b25	121	'duration': 853.386,
1c97b0a7	122	},
5819edef YCH	123	# The video in the original webpage now uses PlayWire
5819edef YCH	124	'skip': 'Ooyala said: movie expired',
1c97b0a7 S	125	}, {
	126	# Only available for ipad
	127	'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
	128	'info_dict': {
	129	'id': 'x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',
	130	'ext': 'mp4',
	131	'title': 'Simulation Overview - Levels of Simulation',
53e06b25	132	'duration': 194.948,
1c97b0a7 S	133	},
	134	},
	135	{
	136	# Information available only through SAS api
	137	# From http://community.plm.automation.siemens.com/t5/News-NX-Manufacturing/Tool-Path-Divide/ba-p/4187
	138	'url': 'http://player.ooyala.com/player.js?embedCode=FiOG81ZTrvckcchQxmalf4aQj590qTEx',
	139	'md5': 'a84001441b35ea492bc03736e59e7935',
	140	'info_dict': {
	141	'id': 'FiOG81ZTrvckcchQxmalf4aQj590qTEx',
	142	'ext': 'mp4',
90bddb6c	143	'title': 'Divide Tool Path.mp4',
53e06b25	144	'duration': 204.405,
1c97b0a7	145	}
b5f523ed S	146	},
	147	{
	148	# empty stream['url']['data']
	149	'url': 'http://player.ooyala.com/player.js?embedCode=w2bnZtYjE6axZ_dw1Cd0hQtXd_ige2Is',
	150	'only_matching': True,
1c97b0a7 S	151	}
	152	]
	153
bfd973ec	154	def _extract_from_webpage(self, url, webpage):
	155	mobj = (re.search(r'player\.ooyala\.com/[^"?]+[?#][^"]*?(?:embedCode\|ec)=(?P<ec>[^"&]+)', webpage)
	156	or re.search(r'OO\.Player\.create\([\'"].?[\'"],\s[\'"](?P<ec>.{32})[\'"]', webpage)
	157	or re.search(r'OO\.Player\.create\.apply\(\sOO\.Player\s,\sop\(\s\[\s[\'"][^\'"][\'"]\s,\s[\'"](?P<ec>.{32})[\'"]', webpage)
	158	or re.search(r'SBN\.VideoLinkset\.ooyala\([\'"](?P<ec>.{32})[\'"]\)', webpage)
	159	or re.search(r'data-ooyala-video-id\s=\s[\'"](?P<ec>.{32})[\'"]', webpage))
	160	if mobj is not None:
	161	embed_token = self._search_regex(
	162	r'embedToken[\'"]?\s:\s[\'"]([^\'"]+)',
	163	webpage, 'ooyala embed token', default=None)
	164	yield self._build_url_result(smuggle_url(
	165	mobj.group('ec'), {
	166	'domain': url,
	167	'embed_token': embed_token,
	168	}))
	169	return
	170
	171	# Look for multiple Ooyala embeds on SBN network websites
	172	mobj = re.search(r'SBN\.VideoLinkset\.entryGroup\((\[.*?\])', webpage)
	173	if mobj is not None:
	174	for v in self._parse_json(mobj.group(1), self._generic_id(url), fatal=False) or []:
	175	yield self._build_url_result(smuggle_url(v['provider_video_id'], {'domain': url}))
	176
1c97b0a7 S	177	@staticmethod
	178	def _url_for_embed_code(embed_code):
	179	return 'http://player.ooyala.com/player.js?embedCode=%s' % embed_code
	180
	181	@classmethod
	182	def _build_url_result(cls, embed_code):
	183	return cls.url_result(cls._url_for_embed_code(embed_code),
	184	ie=cls.ie_key())
	185
	186	def _real_extract(self, url):
cce9d15d	187	url, smuggled_data = unsmuggle_url(url, {})
1c97b0a7	188	embed_code = self._match_id(url)
cce9d15d	189	domain = smuggled_data.get('domain')
cb882540	190	supportedformats = smuggled_data.get('supportedformats')
9837cb75	191	embed_token = smuggled_data.get('embed_token')
6101f45e	192	content_tree_url = self._CONTENT_TREE_BASE + 'embed_code/%s/%s' % (embed_code, embed_code)
9837cb75	193	return self._extract(content_tree_url, embed_code, domain, supportedformats, embed_token)
1c97b0a7 S	194
	195
	196	class OoyalaExternalIE(OoyalaBaseIE):
	197	_VALID_URL = r'''(?x)
	198	(?:
	199	ooyalaexternal:\|
	200	https?://.+?\.ooyala\.com/.*?\bexternalId=
	201	)
	202	(?P<partner_id>[^:]+)
	203	:
	204	(?P<id>.+)
	205	(?:
	206	:\|
	207	.*?&pcode=
	208	)
	209	(?P<pcode>.+?)
90bddb6c	210	(?:&\|$)
1c97b0a7 S	211	'''
	212
	213	_TEST = {
	214	'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',
	215	'info_dict': {
	216	'id': 'FkYWtmazr6Ed8xmvILvKLWjd4QvYZpzG',
	217	'ext': 'mp4',
	218	'title': 'dm_140128_30for30Shorts___JudgingJewellv2',
6101f45e	219	'duration': 1302.0,
1c97b0a7 S	220	},
	221	'params': {
	222	# m3u8 download
	223	'skip_download': True,
	224	},
	225	}
	226
	227	def _real_extract(self, url):
5ad28e7f	228	partner_id, video_id, pcode = self._match_valid_url(url).groups()
6101f45e	229	content_tree_url = self._CONTENT_TREE_BASE + 'external_id/%s/%s:%s' % (pcode, partner_id, video_id)
90bddb6c	230	return self._extract(content_tree_url, video_id)