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

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..utils import (
    ExtractorError,
    get_element_by_class,
    int_or_none,
    parse_iso8601,
    smuggle_url,
    str_or_none,
    strip_or_none,
    try_get,
    unsmuggle_url,
    urlencode_postdata,
)


class VidioBaseIE(InfoExtractor):
    _LOGIN_URL = 'https://www.vidio.com/users/login'
    _NETRC_MACHINE = 'vidio'

    def _login(self):
        username, password = self._get_login_info()
        if username is None:
            return

        def is_logged_in():
            res = self._download_json(
                'https://www.vidio.com/interactions.json', None, 'Checking if logged in', fatal=False) or {}
            return bool(res.get('current_user'))

        if is_logged_in():
            return

        login_page = self._download_webpage(
            self._LOGIN_URL, None, 'Downloading log in page')

        login_form = self._form_hidden_inputs("login-form", login_page)
        login_form.update({
            'user[login]': username,
            'user[password]': password,
        })
        login_post, login_post_urlh = self._download_webpage_handle(
            self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(login_form), expected_status=[302, 401])

        if login_post_urlh.status == 401:
            reason = get_element_by_class('onboarding-form__general-error', login_post)
            if reason:
                raise ExtractorError(
                    'Unable to log in: %s' % reason, expected=True)
            raise ExtractorError('Unable to log in')

    def _real_initialize(self):
        self._api_key = self._download_json(
            'https://www.vidio.com/auth', None, data=b'')['api_key']
        self._login()

    def _call_api(self, url, video_id, note=None):
        return self._download_json(url, video_id, note=note, headers={
            'Content-Type': 'application/vnd.api+json',
            'X-API-KEY': self._api_key,
        })


class VidioIE(VidioBaseIE):
    _VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
    _TESTS = [{
        'url': 'http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015',
        'md5': 'cd2801394afc164e9775db6a140b91fe',
        'info_dict': {
            'id': '165683',
            'display_id': 'dj_ambred-booyah-live-2015',
            'ext': 'mp4',
            'title': 'DJ_AMBRED - Booyah (Live 2015)',
            'description': 'md5:27dc15f819b6a78a626490881adbadf8',
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 149,
            'like_count': int,
            'uploader': 'TWELVE Pic',
            'timestamp': 1444902800,
            'upload_date': '20151015',
            'uploader_id': 'twelvepictures',
            'channel': 'Cover Music Video',
            'channel_id': '280236',
            'view_count': int,
            'dislike_count': int,
            'comment_count': int,
            'tags': 'count:4',
        },
    }, {
        'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
        'only_matching': True,
    }, {
        # Premier-exclusive video
        'url': 'https://www.vidio.com/watch/1550718-stand-by-me-doraemon',
        'only_matching': True
    }]

    def _real_extract(self, url):
        match = re.match(self._VALID_URL, url).groupdict()
        video_id, display_id = match.get('id'), match.get('display_id')
        data = self._call_api('https://api.vidio.com/videos/' + video_id, display_id)
        video = data['videos'][0]
        title = video['title'].strip()
        is_premium = video.get('is_premium')

        if is_premium:
            sources = self._download_json(
                'https://www.vidio.com/interactions_stream.json?video_id=%s&type=videos' % video_id,
                display_id, note='Downloading premier API JSON')
            if not (sources.get('source') or sources.get('source_dash')):
                self.raise_login_required('This video is only available for registered users with the appropriate subscription')

            formats, subs = [], {}
            if sources.get('source'):
                hls_formats, hls_subs = self._extract_m3u8_formats_and_subtitles(
                    sources['source'], display_id, 'mp4', 'm3u8_native')
                formats.extend(hls_formats)
                subs.update(hls_subs)
            if sources.get('source_dash'):  # TODO: Find video example with source_dash
                dash_formats, dash_subs = self._extract_mpd_formats_and_subtitles(
                    sources['source_dash'], display_id, 'dash')
                formats.extend(dash_formats)
                subs.update(dash_subs)
        else:
            hls_url = data['clips'][0]['hls_url']
            formats, subs = self._extract_m3u8_formats_and_subtitles(
                hls_url, display_id, 'mp4', 'm3u8_native')

        self._sort_formats(formats)

        get_first = lambda x: try_get(data, lambda y: y[x + 's'][0], dict) or {}
        channel = get_first('channel')
        user = get_first('user')
        username = user.get('username')
        get_count = lambda x: int_or_none(video.get('total_' + x))

        return {
            'id': video_id,
            'display_id': display_id,
            'title': title,
            'description': strip_or_none(video.get('description')),
            'thumbnail': video.get('image_url_medium'),
            'duration': int_or_none(video.get('duration')),
            'like_count': get_count('likes'),
            'formats': formats,
            'subtitles': subs,
            'uploader': user.get('name'),
            'timestamp': parse_iso8601(video.get('created_at')),
            'uploader_id': username,
            'uploader_url': 'https://www.vidio.com/@' + username if username else None,
            'channel': channel.get('name'),
            'channel_id': str_or_none(channel.get('id')),
            'view_count': get_count('view_count'),
            'dislike_count': get_count('dislikes'),
            'comment_count': get_count('comments'),
            'tags': video.get('tag_list'),
        }


class VidioPremierIE(VidioBaseIE):
    _VALID_URL = r'https?://(?:www\.)?vidio\.com/premier/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://www.vidio.com/premier/2885/badai-pasti-berlalu',
        'playlist_mincount': 14,
    }, {
        # Series with both free and premier-exclusive videos
        'url': 'https://www.vidio.com/premier/2567/sosmed',
        'only_matching': True,
    }]

    def _playlist_entries(self, playlist_url, display_id):
        index = 1
        while playlist_url:
            playlist_json = self._call_api(playlist_url, display_id, 'Downloading API JSON page %s' % index)
            for video_json in playlist_json.get('data', []):
                link = video_json['links']['watchpage']
                yield self.url_result(link, 'Vidio', video_json['id'])
            playlist_url = try_get(playlist_json, lambda x: x['links']['next'])
            index += 1

    def _real_extract(self, url):
        url, idata = unsmuggle_url(url, {})
        playlist_id, display_id = re.match(self._VALID_URL, url).groups()

        playlist_url = idata.get('url')
        if playlist_url:  # Smuggled data contains an API URL. Download only that playlist
            playlist_id = idata['id']
            return self.playlist_result(
                self._playlist_entries(playlist_url, playlist_id),
                playlist_id=playlist_id, playlist_title=idata.get('title'))

        playlist_data = self._call_api('https://api.vidio.com/content_profiles/%s/playlists' % playlist_id, display_id)

        return self.playlist_from_matches(
            playlist_data.get('data', []), playlist_id=playlist_id, ie=self.ie_key(),
            getter=lambda data: smuggle_url(url, {
                'url': data['relationships']['videos']['links']['related'],
                'id': data['id'],
                'title': try_get(data, lambda x: x['attributes']['name'])
            }))


class VidioLiveIE(VidioBaseIE):
    _VALID_URL = r'https?://(?:www\.)?vidio\.com/live/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
    _TESTS = [{
        'url': 'https://www.vidio.com/live/204-sctv',
        'info_dict': {
            'id': '204',
            'title': 'SCTV',
            'uploader': 'SCTV',
            'uploader_id': 'sctv',
            'thumbnail': r're:^https?://.*\.jpg$',
        },
    }, {
        # Premier-exclusive livestream
        'url': 'https://www.vidio.com/live/6362-tvn',
        'only_matching': True,
    }, {
        # DRM premier-exclusive livestream
        'url': 'https://www.vidio.com/live/6299-bein-1',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        video_id, display_id = re.match(self._VALID_URL, url).groups()
        stream_data = self._call_api(
            'https://www.vidio.com/api/livestreamings/%s/detail' % video_id, display_id)
        stream_meta = stream_data['livestreamings'][0]
        user = stream_data.get('users', [{}])[0]

        title = stream_meta.get('title')
        username = user.get('username')

        formats = []
        if stream_meta.get('is_drm'):
            if not self.get_param('allow_unplayable_formats'):
                self.raise_no_formats(
                    'This video is DRM protected.', expected=True)
        if stream_meta.get('is_premium'):
            sources = self._download_json(
                'https://www.vidio.com/interactions_stream.json?video_id=%s&type=livestreamings' % video_id,
                display_id, note='Downloading premier API JSON')
            if not (sources.get('source') or sources.get('source_dash')):
                self.raise_login_required('This video is only available for registered users with the appropriate subscription')

            if str_or_none(sources.get('source')):
                token_json = self._download_json(
                    'https://www.vidio.com/live/%s/tokens' % video_id,
                    display_id, note='Downloading HLS token JSON', data=b'')
                formats.extend(self._extract_m3u8_formats(
                    sources['source'] + '?' + token_json.get('token', ''), display_id, 'mp4', 'm3u8_native'))
            if str_or_none(sources.get('source_dash')):
                pass
        else:
            if stream_meta.get('stream_token_url'):
                token_json = self._download_json(
                    'https://www.vidio.com/live/%s/tokens' % video_id,
                    display_id, note='Downloading HLS token JSON', data=b'')
                formats.extend(self._extract_m3u8_formats(
                    stream_meta['stream_token_url'] + '?' + token_json.get('token', ''),
                    display_id, 'mp4', 'm3u8_native'))
            if stream_meta.get('stream_dash_url'):
                pass
            if stream_meta.get('stream_url'):
                formats.extend(self._extract_m3u8_formats(
                    stream_meta['stream_url'], display_id, 'mp4', 'm3u8_native'))
        self._sort_formats(formats)

        return {
            'id': video_id,
            'display_id': display_id,
            'title': title,
            'is_live': True,
            'description': strip_or_none(stream_meta.get('description')),
            'thumbnail': stream_meta.get('image'),
            'like_count': int_or_none(stream_meta.get('like')),
            'dislike_count': int_or_none(stream_meta.get('dislike')),
            'formats': formats,
            'uploader': user.get('name'),
            'timestamp': parse_iso8601(stream_meta.get('start_time')),
            'uploader_id': username,
            'uploader_url': 'https://www.vidio.com/@' + username if username else None,
        }
Commit	Line	Data
7def3571 T	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
	4	import re
7def3571	5
0fc832e1	6	from .common import InfoExtractor
2181983a	7	from ..utils import (
10bb7e51 M	8	ExtractorError,
10bb7e51 M	9	get_element_by_class,
2181983a	10	int_or_none,
2181983a	11	parse_iso8601,
f2cd7060	12	smuggle_url,
2181983a	13	str_or_none,
	14	strip_or_none,
	15	try_get,
f2cd7060	16	unsmuggle_url,
10bb7e51	17	urlencode_postdata,
2181983a	18	)
7def3571 T	19
7def3571 T	20
f2cd7060	21	class VidioBaseIE(InfoExtractor):
10bb7e51 M	22	_LOGIN_URL = 'https://www.vidio.com/users/login'
	23	_NETRC_MACHINE = 'vidio'
	24
	25	def _login(self):
	26	username, password = self._get_login_info()
	27	if username is None:
	28	return
	29
	30	def is_logged_in():
	31	res = self._download_json(
	32	'https://www.vidio.com/interactions.json', None, 'Checking if logged in', fatal=False) or {}
	33	return bool(res.get('current_user'))
	34
	35	if is_logged_in():
	36	return
	37
	38	login_page = self._download_webpage(
	39	self._LOGIN_URL, None, 'Downloading log in page')
	40
	41	login_form = self._form_hidden_inputs("login-form", login_page)
	42	login_form.update({
	43	'user[login]': username,
	44	'user[password]': password,
	45	})
	46	login_post, login_post_urlh = self._download_webpage_handle(
	47	self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(login_form), expected_status=[302, 401])
	48
	49	if login_post_urlh.status == 401:
	50	reason = get_element_by_class('onboarding-form__general-error', login_post)
	51	if reason:
	52	raise ExtractorError(
	53	'Unable to log in: %s' % reason, expected=True)
	54	raise ExtractorError('Unable to log in')
7def3571	55
2181983a	56	def _real_initialize(self):
	57	self._api_key = self._download_json(
	58	'https://www.vidio.com/auth', None, data=b'')['api_key']
10bb7e51	59	self._login()
0fc832e1	60
f2cd7060 M	61	def _call_api(self, url, video_id, note=None):
	62	return self._download_json(url, video_id, note=note, headers={
	63	'Content-Type': 'application/vnd.api+json',
	64	'X-API-KEY': self._api_key,
	65	})
	66
	67
	68	class VidioIE(VidioBaseIE):
	69	_VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
	70	_TESTS = [{
	71	'url': 'http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015',
	72	'md5': 'cd2801394afc164e9775db6a140b91fe',
	73	'info_dict': {
	74	'id': '165683',
	75	'display_id': 'dj_ambred-booyah-live-2015',
	76	'ext': 'mp4',
	77	'title': 'DJ_AMBRED - Booyah (Live 2015)',
	78	'description': 'md5:27dc15f819b6a78a626490881adbadf8',
	79	'thumbnail': r're:^https?://.*\.jpg$',
	80	'duration': 149,
	81	'like_count': int,
	82	'uploader': 'TWELVE Pic',
	83	'timestamp': 1444902800,
	84	'upload_date': '20151015',
	85	'uploader_id': 'twelvepictures',
	86	'channel': 'Cover Music Video',
	87	'channel_id': '280236',
	88	'view_count': int,
	89	'dislike_count': int,
	90	'comment_count': int,
	91	'tags': 'count:4',
	92	},
	93	}, {
	94	'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
	95	'only_matching': True,
	96	}, {
	97	# Premier-exclusive video
	98	'url': 'https://www.vidio.com/watch/1550718-stand-by-me-doraemon',
	99	'only_matching': True
	100	}]
	101
2181983a	102	def _real_extract(self, url):
f2cd7060 M	103	match = re.match(self._VALID_URL, url).groupdict()
	104	video_id, display_id = match.get('id'), match.get('display_id')
	105	data = self._call_api('https://api.vidio.com/videos/' + video_id, display_id)
2181983a	106	video = data['videos'][0]
2181983a	107	title = video['title'].strip()
46c43ffc	108	is_premium = video.get('is_premium')
f2cd7060	109
46c43ffc M	110	if is_premium:
	111	sources = self._download_json(
	112	'https://www.vidio.com/interactions_stream.json?video_id=%s&type=videos' % video_id,
	113	display_id, note='Downloading premier API JSON')
	114	if not (sources.get('source') or sources.get('source_dash')):
f2cd7060	115	self.raise_login_required('This video is only available for registered users with the appropriate subscription')
46c43ffc M	116
	117	formats, subs = [], {}
	118	if sources.get('source'):
	119	hls_formats, hls_subs = self._extract_m3u8_formats_and_subtitles(
	120	sources['source'], display_id, 'mp4', 'm3u8_native')
	121	formats.extend(hls_formats)
	122	subs.update(hls_subs)
	123	if sources.get('source_dash'): # TODO: Find video example with source_dash
	124	dash_formats, dash_subs = self._extract_mpd_formats_and_subtitles(
	125	sources['source_dash'], display_id, 'dash')
	126	formats.extend(dash_formats)
	127	subs.update(dash_subs)
	128	else:
	129	hls_url = data['clips'][0]['hls_url']
	130	formats, subs = self._extract_m3u8_formats_and_subtitles(
	131	hls_url, display_id, 'mp4', 'm3u8_native')
0fc832e1	132
07ad0cf3	133	self._sort_formats(formats)
0fc832e1	134
2181983a	135	get_first = lambda x: try_get(data, lambda y: y[x + 's'][0], dict) or {}
	136	channel = get_first('channel')
	137	user = get_first('user')
	138	username = user.get('username')
	139	get_count = lambda x: int_or_none(video.get('total_' + x))
7def3571 T	140
	141	return {
	142	'id': video_id,
0fc832e1 S	143	'display_id': display_id,
0fc832e1 S	144	'title': title,
2181983a	145	'description': strip_or_none(video.get('description')),
	146	'thumbnail': video.get('image_url_medium'),
	147	'duration': int_or_none(video.get('duration')),
	148	'like_count': get_count('likes'),
0fc832e1	149	'formats': formats,
46c43ffc	150	'subtitles': subs,
2181983a	151	'uploader': user.get('name'),
	152	'timestamp': parse_iso8601(video.get('created_at')),
	153	'uploader_id': username,
	154	'uploader_url': 'https://www.vidio.com/@' + username if username else None,
	155	'channel': channel.get('name'),
	156	'channel_id': str_or_none(channel.get('id')),
	157	'view_count': get_count('view_count'),
	158	'dislike_count': get_count('dislikes'),
	159	'comment_count': get_count('comments'),
	160	'tags': video.get('tag_list'),
7def3571	161	}
f2cd7060 M	162
	163
	164	class VidioPremierIE(VidioBaseIE):
	165	_VALID_URL = r'https?://(?:www\.)?vidio\.com/premier/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
	166	_TESTS = [{
	167	'url': 'https://www.vidio.com/premier/2885/badai-pasti-berlalu',
	168	'playlist_mincount': 14,
	169	}, {
	170	# Series with both free and premier-exclusive videos
	171	'url': 'https://www.vidio.com/premier/2567/sosmed',
	172	'only_matching': True,
	173	}]
	174
	175	def _playlist_entries(self, playlist_url, display_id):
	176	index = 1
	177	while playlist_url:
	178	playlist_json = self._call_api(playlist_url, display_id, 'Downloading API JSON page %s' % index)
	179	for video_json in playlist_json.get('data', []):
	180	link = video_json['links']['watchpage']
	181	yield self.url_result(link, 'Vidio', video_json['id'])
	182	playlist_url = try_get(playlist_json, lambda x: x['links']['next'])
	183	index += 1
	184
	185	def _real_extract(self, url):
	186	url, idata = unsmuggle_url(url, {})
	187	playlist_id, display_id = re.match(self._VALID_URL, url).groups()
	188
	189	playlist_url = idata.get('url')
	190	if playlist_url: # Smuggled data contains an API URL. Download only that playlist
	191	playlist_id = idata['id']
	192	return self.playlist_result(
	193	self._playlist_entries(playlist_url, playlist_id),
	194	playlist_id=playlist_id, playlist_title=idata.get('title'))
	195
	196	playlist_data = self._call_api('https://api.vidio.com/content_profiles/%s/playlists' % playlist_id, display_id)
	197
	198	return self.playlist_from_matches(
	199	playlist_data.get('data', []), playlist_id=playlist_id, ie=self.ie_key(),
	200	getter=lambda data: smuggle_url(url, {
	201	'url': data['relationships']['videos']['links']['related'],
	202	'id': data['id'],
	203	'title': try_get(data, lambda x: x['attributes']['name'])
	204	}))
	205
	206
	207	class VidioLiveIE(VidioBaseIE):
	208	_VALID_URL = r'https?://(?:www\.)?vidio\.com/live/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
	209	_TESTS = [{
	210	'url': 'https://www.vidio.com/live/204-sctv',
	211	'info_dict': {
	212	'id': '204',
	213	'title': 'SCTV',
	214	'uploader': 'SCTV',
	215	'uploader_id': 'sctv',
	216	'thumbnail': r're:^https?://.*\.jpg$',
	217	},
	218	}, {
	219	# Premier-exclusive livestream
	220	'url': 'https://www.vidio.com/live/6362-tvn',
	221	'only_matching': True,
	222	}, {
	223	# DRM premier-exclusive livestream
	224	'url': 'https://www.vidio.com/live/6299-bein-1',
	225	'only_matching': True,
226	}]
227
228	def _real_extract(self, url):
229	video_id, display_id = re.match(self._VALID_URL, url).groups()
230	stream_data = self._call_api(
231	'https://www.vidio.com/api/livestreamings/%s/detail' % video_id, display_id)
232	stream_meta = stream_data['livestreamings'][0]
233	user = stream_data.get('users', [{}])[0]
234
235	title = stream_meta.get('title')
236	username = user.get('username')
237
238	formats = []
239	if stream_meta.get('is_drm'):
240	if not self.get_param('allow_unplayable_formats'):
241	self.raise_no_formats(
242	'This video is DRM protected.', expected=True)
243	if stream_meta.get('is_premium'):
244	sources = self._download_json(
245	'https://www.vidio.com/interactions_stream.json?video_id=%s&type=livestreamings' % video_id,
246	display_id, note='Downloading premier API JSON')
247	if not (sources.get('source') or sources.get('source_dash')):
248	self.raise_login_required('This video is only available for registered users with the appropriate subscription')
249
250	if str_or_none(sources.get('source')):
251	token_json = self._download_json(
252	'https://www.vidio.com/live/%s/tokens' % video_id,
253	display_id, note='Downloading HLS token JSON', data=b'')
254	formats.extend(self._extract_m3u8_formats(
255	sources['source'] + '?' + token_json.get('token', ''), display_id, 'mp4', 'm3u8_native'))
256	if str_or_none(sources.get('source_dash')):
257	pass
258	else:
259	if stream_meta.get('stream_token_url'):
260	token_json = self._download_json(
261	'https://www.vidio.com/live/%s/tokens' % video_id,
262	display_id, note='Downloading HLS token JSON', data=b'')
263	formats.extend(self._extract_m3u8_formats(
264	stream_meta['stream_token_url'] + '?' + token_json.get('token', ''),
265	display_id, 'mp4', 'm3u8_native'))
266	if stream_meta.get('stream_dash_url'):
267	pass
268	if stream_meta.get('stream_url'):
269	formats.extend(self._extract_m3u8_formats(
270	stream_meta['stream_url'], display_id, 'mp4', 'm3u8_native'))
271	self._sort_formats(formats)
272
273	return {
274	'id': video_id,
275	'display_id': display_id,
276	'title': title,
277	'is_live': True,
278	'description': strip_or_none(stream_meta.get('description')),
279	'thumbnail': stream_meta.get('image'),
280	'like_count': int_or_none(stream_meta.get('like')),
281	'dislike_count': int_or_none(stream_meta.get('dislike')),
282	'formats': formats,
283	'uploader': user.get('name'),
284	'timestamp': parse_iso8601(stream_meta.get('start_time')),
285	'uploader_id': username,
286	'uploader_url': 'https://www.vidio.com/@' + username if username else None,
287	}