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

import itertools
from datetime import datetime

from .common import InfoExtractor
from ..utils import (
    determine_ext,
    ExtractorError,
    float_or_none,
    format_field,
    int_or_none,
    str_or_none,
    traverse_obj,
    unified_timestamp,
    url_or_none,
)


_API_BASE_URL = 'https://prod-api-v2.production.rokfin.com/api/v2/public/'


class RokfinIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?rokfin\.com/(?P<id>(?P<type>post|stream)/\d+)'
    _TESTS = [{
        'url': 'https://www.rokfin.com/post/57548/Mitt-Romneys-Crazy-Solution-To-Climate-Change',
        'info_dict': {
            'id': 'post/57548',
            'ext': 'mp4',
            'title': 'Mitt Romney\'s Crazy Solution To Climate Change',
            'thumbnail': r're:https://img\.production\.rokfin\.com/.+',
            'upload_date': '20211023',
            'timestamp': 1634998029,
            'channel': 'Jimmy Dore',
            'channel_id': 65429,
            'channel_url': 'https://rokfin.com/TheJimmyDoreShow',
            'duration': 213.0,
            'availability': 'public',
            'live_status': 'not_live',
            'dislike_count': int,
            'like_count': int,
        }
    }, {
        'url': 'https://rokfin.com/post/223/Julian-Assange-Arrested-Streaming-In-Real-Time',
        'info_dict': {
            'id': 'post/223',
            'ext': 'mp4',
            'title': 'Julian Assange Arrested: Streaming In Real Time',
            'thumbnail': r're:https://img\.production\.rokfin\.com/.+',
            'upload_date': '20190412',
            'timestamp': 1555052644,
            'channel': 'Ron Placone',
            'channel_id': 10,
            'channel_url': 'https://rokfin.com/RonPlacone',
            'availability': 'public',
            'live_status': 'not_live',
            'dislike_count': int,
            'like_count': int,
            'tags': ['FreeThinkingMedia^', 'RealProgressives^'],
        }
    }, {
        'url': 'https://www.rokfin.com/stream/10543/Its-A-Crazy-Mess-Regional-Director-Blows-Whistle-On-Pfizers-Vaccine-Trial-Data',
        'info_dict': {
            'id': 'stream/10543',
            'ext': 'mp4',
            'title': '"It\'s A Crazy Mess" Regional Director Blows Whistle On Pfizer\'s Vaccine Trial Data',
            'thumbnail': r're:https://img\.production\.rokfin\.com/.+',
            'description': 'md5:324ce2d3e3b62e659506409e458b9d8e',
            'channel': 'Ryan Cristián',
            'channel_id': 53856,
            'channel_url': 'https://rokfin.com/TLAVagabond',
            'availability': 'public',
            'is_live': False,
            'was_live': True,
            'live_status': 'was_live',
            'timestamp': 1635874720,
            'release_timestamp': 1635874720,
            'release_date': '20211102',
            'upload_date': '20211102',
            'dislike_count': int,
            'like_count': int,
            'tags': ['FreeThinkingMedia^'],
        }
    }]

    def _real_extract(self, url):
        video_id, video_type = self._match_valid_url(url).group('id', 'type')

        metadata = self._download_json(f'{_API_BASE_URL}{video_id}', video_id)

        scheduled = unified_timestamp(metadata.get('scheduledAt'))
        live_status = ('was_live' if metadata.get('stoppedAt')
                       else 'is_upcoming' if scheduled
                       else 'is_live' if video_type == 'stream'
                       else 'not_live')

        video_url = traverse_obj(metadata, 'url', ('content', 'contentUrl'), expected_type=url_or_none)
        formats, subtitles = [{'url': video_url}] if video_url else [], {}
        if determine_ext(video_url) == 'm3u8':
            formats, subtitles = self._extract_m3u8_formats_and_subtitles(
                video_url, video_id, fatal=False, live=live_status == 'is_live')

        if not formats:
            if traverse_obj(metadata, 'premiumPlan', 'premium'):
                self.raise_login_required('This video is only available to premium users', True, method='cookies')
            elif scheduled:
                self.raise_no_formats(
                    f'Stream is offline; sheduled for {datetime.fromtimestamp(scheduled).strftime("%Y-%m-%d %H:%M:%S")}',
                    video_id=video_id, expected=True)
        self._sort_formats(formats)

        uploader = traverse_obj(metadata, ('createdBy', 'username'), ('creator', 'username'))
        timestamp = (scheduled or float_or_none(metadata.get('postedAtMilli'), 1000)
                     or unified_timestamp(metadata.get('creationDateTime')))
        return {
            'id': video_id,
            'formats': formats,
            'subtitles': subtitles,
            'title': str_or_none(traverse_obj(metadata, 'title', ('content', 'contentTitle'))),
            'duration': float_or_none(traverse_obj(metadata, ('content', 'duration'))),
            'thumbnail': url_or_none(traverse_obj(metadata, 'thumbnail', ('content', 'thumbnailUrl1'))),
            'description': str_or_none(traverse_obj(metadata, 'description', ('content', 'contentDescription'))),
            'like_count': int_or_none(metadata.get('likeCount')),
            'dislike_count': int_or_none(metadata.get('dislikeCount')),
            'channel': str_or_none(traverse_obj(metadata, ('createdBy', 'name'), ('creator', 'name'))),
            'channel_id': traverse_obj(metadata, ('createdBy', 'id'), ('creator', 'id')),
            'channel_url': url_or_none(f'https://rokfin.com/{uploader}') if uploader else None,
            'timestamp': timestamp,
            'release_timestamp': timestamp if live_status != 'not_live' else None,
            'tags': traverse_obj(metadata, ('tags', ..., 'title'), expected_type=str_or_none),
            'live_status': live_status,
            'availability': self._availability(
                needs_premium=bool(traverse_obj(metadata, 'premiumPlan', 'premium')),
                is_private=False, needs_subscription=False, needs_auth=False, is_unlisted=False),
            # 'comment_count': metadata.get('numComments'), # Data provided by website is wrong
            '__post_extractor': self.extract_comments(video_id) if video_type == 'post' else None,
        }

    def _get_comments(self, video_id):
        pages_total = None
        for page_n in itertools.count():
            raw_comments = self._download_json(
                f'{_API_BASE_URL}comment?postId={video_id[5:]}&page={page_n}&size=50',
                video_id, note=f'Downloading viewer comments page {page_n + 1}{format_field(pages_total, template=" of %s")}',
                fatal=False) or {}

            for comment in raw_comments.get('content') or []:
                yield {
                    'text': str_or_none(comment.get('comment')),
                    'author': str_or_none(comment.get('name')),
                    'id': comment.get('commentId'),
                    'author_id': comment.get('userId'),
                    'parent': 'root',
                    'like_count': int_or_none(comment.get('numLikes')),
                    'dislike_count': int_or_none(comment.get('numDislikes')),
                    'timestamp': unified_timestamp(comment.get('postedAt'))
                }

            pages_total = int_or_none(raw_comments.get('totalPages')) or None
            is_last = raw_comments.get('last')
            if not raw_comments.get('content') or is_last or (page_n > pages_total if pages_total else is_last is not False):
                return


class RokfinPlaylistBaseIE(InfoExtractor):
    _TYPES = {
        'video': 'post',
        'audio': 'post',
        'stream': 'stream',
        'dead_stream': 'stream',
        'stack': 'stack',
    }

    def _get_video_data(self, metadata):
        for content in metadata.get('content') or []:
            media_type = self._TYPES.get(content.get('mediaType'))
            video_id = content.get('id') if media_type == 'post' else content.get('mediaId')
            if not media_type or not video_id:
                continue

            yield self.url_result(f'https://rokfin.com/{media_type}/{video_id}', video_id=f'{media_type}/{video_id}',
                                  video_title=str_or_none(traverse_obj(content, ('content', 'contentTitle'))))


class RokfinStackIE(RokfinPlaylistBaseIE):
    IE_NAME = 'rokfin:stack'
    _VALID_URL = r'https?://(?:www\.)?rokfin\.com/stack/(?P<id>[^/]+)'
    _TESTS = [{
        'url': 'https://www.rokfin.com/stack/271/Tulsi-Gabbard-Portsmouth-Townhall-FULL--Feb-9-2020',
        'playlist_count': 8,
        'info_dict': {
            'id': '271',
        },
    }]

    def _real_extract(self, url):
        list_id = self._match_id(url)
        return self.playlist_result(self._get_video_data(
            self._download_json(f'{_API_BASE_URL}stack/{list_id}', list_id)), list_id)


class RokfinChannelIE(RokfinPlaylistBaseIE):
    IE_NAME = 'rokfin:channel'
    _VALID_URL = r'https?://(?:www\.)?rokfin\.com/(?!((feed/?)|(discover/?)|(channels/?))$)(?P<id>[^/]+)/?$'
    _TESTS = [{
        'url': 'https://rokfin.com/TheConvoCouch',
        'playlist_mincount': 100,
        'info_dict': {
            'id': '12071-new',
            'title': 'TheConvoCouch - New',
            'description': 'md5:bb622b1bca100209b91cd685f7847f06',
        },
    }]

    _TABS = {
        'new': 'posts',
        'top': 'top',
        'videos': 'video',
        'podcasts': 'audio',
        'streams': 'stream',
        'stacks': 'stack',
    }

    def _real_initialize(self):
        self._validate_extractor_args()

    def _validate_extractor_args(self):
        requested_tabs = self._configuration_arg('tab', None)
        if requested_tabs is not None and (len(requested_tabs) > 1 or requested_tabs[0] not in self._TABS):
            raise ExtractorError(f'Invalid extractor-arg "tab". Must be one of {", ".join(self._TABS)}', expected=True)

    def _entries(self, channel_id, channel_name, tab):
        pages_total = None
        for page_n in itertools.count(0):
            if tab in ('posts', 'top'):
                data_url = f'{_API_BASE_URL}user/{channel_name}/{tab}?page={page_n}&size=50'
            else:
                data_url = f'{_API_BASE_URL}post/search/{tab}?page={page_n}&size=50&creator={channel_id}'
            metadata = self._download_json(
                data_url, channel_name,
                note=f'Downloading video metadata page {page_n + 1}{format_field(pages_total, template=" of %s")}')

            yield from self._get_video_data(metadata)
            pages_total = int_or_none(metadata.get('totalPages')) or None
            is_last = metadata.get('last')
            if is_last or (page_n > pages_total if pages_total else is_last is not False):
                return

    def _real_extract(self, url):
        channel_name = self._match_id(url)
        channel_info = self._download_json(f'{_API_BASE_URL}user/{channel_name}', channel_name)
        channel_id = channel_info['id']
        tab = self._configuration_arg('tab', default=['new'])[0]

        return self.playlist_result(
            self._entries(channel_id, channel_name, self._TABS[tab]),
            f'{channel_id}-{tab}', f'{channel_name} - {tab.title()}', str_or_none(channel_info.get('description')))
Commit	Line	Data
334b1c48	1	import itertools
	2	from datetime import datetime
	3
	4	from .common import InfoExtractor
	5	from ..utils import (
	6	determine_ext,
7eaf7f9a	7	ExtractorError,
334b1c48	8	float_or_none,
	9	format_field,
	10	int_or_none,
	11	str_or_none,
	12	traverse_obj,
	13	unified_timestamp,
	14	url_or_none,
	15	)
	16
	17
7eaf7f9a	18	_API_BASE_URL = 'https://prod-api-v2.production.rokfin.com/api/v2/public/'
	19
	20
334b1c48	21	class RokfinIE(InfoExtractor):
	22	_VALID_URL = r'https?://(?:www\.)?rokfin\.com/(?P<id>(?P<type>post\|stream)/\d+)'
	23	_TESTS = [{
	24	'url': 'https://www.rokfin.com/post/57548/Mitt-Romneys-Crazy-Solution-To-Climate-Change',
	25	'info_dict': {
	26	'id': 'post/57548',
	27	'ext': 'mp4',
	28	'title': 'Mitt Romney\'s Crazy Solution To Climate Change',
	29	'thumbnail': r're:https://img\.production\.rokfin\.com/.+',
	30	'upload_date': '20211023',
	31	'timestamp': 1634998029,
	32	'channel': 'Jimmy Dore',
	33	'channel_id': 65429,
	34	'channel_url': 'https://rokfin.com/TheJimmyDoreShow',
	35	'duration': 213.0,
	36	'availability': 'public',
	37	'live_status': 'not_live',
	38	'dislike_count': int,
	39	'like_count': int,
	40	}
	41	}, {
	42	'url': 'https://rokfin.com/post/223/Julian-Assange-Arrested-Streaming-In-Real-Time',
	43	'info_dict': {
	44	'id': 'post/223',
	45	'ext': 'mp4',
	46	'title': 'Julian Assange Arrested: Streaming In Real Time',
	47	'thumbnail': r're:https://img\.production\.rokfin\.com/.+',
	48	'upload_date': '20190412',
	49	'timestamp': 1555052644,
	50	'channel': 'Ron Placone',
	51	'channel_id': 10,
	52	'channel_url': 'https://rokfin.com/RonPlacone',
	53	'availability': 'public',
	54	'live_status': 'not_live',
	55	'dislike_count': int,
	56	'like_count': int,
	57	'tags': ['FreeThinkingMedia^', 'RealProgressives^'],
	58	}
	59	}, {
	60	'url': 'https://www.rokfin.com/stream/10543/Its-A-Crazy-Mess-Regional-Director-Blows-Whistle-On-Pfizers-Vaccine-Trial-Data',
	61	'info_dict': {
	62	'id': 'stream/10543',
	63	'ext': 'mp4',
	64	'title': '"It\'s A Crazy Mess" Regional Director Blows Whistle On Pfizer\'s Vaccine Trial Data',
	65	'thumbnail': r're:https://img\.production\.rokfin\.com/.+',
	66	'description': 'md5:324ce2d3e3b62e659506409e458b9d8e',
	67	'channel': 'Ryan Cristián',
	68	'channel_id': 53856,
	69	'channel_url': 'https://rokfin.com/TLAVagabond',
	70	'availability': 'public',
	71	'is_live': False,
	72	'was_live': True,
	73	'live_status': 'was_live',
	74	'timestamp': 1635874720,
	75	'release_timestamp': 1635874720,
	76	'release_date': '20211102',
	77	'upload_date': '20211102',
	78	'dislike_count': int,
	79	'like_count': int,
	80	'tags': ['FreeThinkingMedia^'],
	81	}
	82	}]
	83
	84	def _real_extract(self, url):
85	video_id, video_type = self._match_valid_url(url).group('id', 'type')
86
7eaf7f9a	87	metadata = self._download_json(f'{_API_BASE_URL}{video_id}', video_id)
334b1c48	88
	89	scheduled = unified_timestamp(metadata.get('scheduledAt'))
	90	live_status = ('was_live' if metadata.get('stoppedAt')
	91	else 'is_upcoming' if scheduled
	92	else 'is_live' if video_type == 'stream'
	93	else 'not_live')
	94
	95	video_url = traverse_obj(metadata, 'url', ('content', 'contentUrl'), expected_type=url_or_none)
	96	formats, subtitles = [{'url': video_url}] if video_url else [], {}
	97	if determine_ext(video_url) == 'm3u8':
	98	formats, subtitles = self._extract_m3u8_formats_and_subtitles(
	99	video_url, video_id, fatal=False, live=live_status == 'is_live')
	100
	101	if not formats:
9461cb58	102	if traverse_obj(metadata, 'premiumPlan', 'premium'):
334b1c48	103	self.raise_login_required('This video is only available to premium users', True, method='cookies')
	104	elif scheduled:
	105	self.raise_no_formats(
	106	f'Stream is offline; sheduled for {datetime.fromtimestamp(scheduled).strftime("%Y-%m-%d %H:%M:%S")}',
	107	video_id=video_id, expected=True)
	108	self._sort_formats(formats)
	109
	110	uploader = traverse_obj(metadata, ('createdBy', 'username'), ('creator', 'username'))
	111	timestamp = (scheduled or float_or_none(metadata.get('postedAtMilli'), 1000)
	112	or unified_timestamp(metadata.get('creationDateTime')))
	113	return {
	114	'id': video_id,
	115	'formats': formats,
	116	'subtitles': subtitles,
	117	'title': str_or_none(traverse_obj(metadata, 'title', ('content', 'contentTitle'))),
	118	'duration': float_or_none(traverse_obj(metadata, ('content', 'duration'))),
	119	'thumbnail': url_or_none(traverse_obj(metadata, 'thumbnail', ('content', 'thumbnailUrl1'))),
	120	'description': str_or_none(traverse_obj(metadata, 'description', ('content', 'contentDescription'))),
	121	'like_count': int_or_none(metadata.get('likeCount')),
	122	'dislike_count': int_or_none(metadata.get('dislikeCount')),
	123	'channel': str_or_none(traverse_obj(metadata, ('createdBy', 'name'), ('creator', 'name'))),
	124	'channel_id': traverse_obj(metadata, ('createdBy', 'id'), ('creator', 'id')),
	125	'channel_url': url_or_none(f'https://rokfin.com/{uploader}') if uploader else None,
	126	'timestamp': timestamp,
	127	'release_timestamp': timestamp if live_status != 'not_live' else None,
	128	'tags': traverse_obj(metadata, ('tags', ..., 'title'), expected_type=str_or_none),
	129	'live_status': live_status,
	130	'availability': self._availability(
9461cb58	131	needs_premium=bool(traverse_obj(metadata, 'premiumPlan', 'premium')),
334b1c48	132	is_private=False, needs_subscription=False, needs_auth=False, is_unlisted=False),
	133	# 'comment_count': metadata.get('numComments'), # Data provided by website is wrong
	134	'__post_extractor': self.extract_comments(video_id) if video_type == 'post' else None,
	135	}
	136
	137	def _get_comments(self, video_id):
	138	pages_total = None
	139	for page_n in itertools.count():
	140	raw_comments = self._download_json(
7eaf7f9a	141	f'{_API_BASE_URL}comment?postId={video_id[5:]}&page={page_n}&size=50',
334b1c48	142	video_id, note=f'Downloading viewer comments page {page_n + 1}{format_field(pages_total, template=" of %s")}',
	143	fatal=False) or {}
	144
	145	for comment in raw_comments.get('content') or []:
	146	yield {
	147	'text': str_or_none(comment.get('comment')),
	148	'author': str_or_none(comment.get('name')),
	149	'id': comment.get('commentId'),
	150	'author_id': comment.get('userId'),
	151	'parent': 'root',
	152	'like_count': int_or_none(comment.get('numLikes')),
	153	'dislike_count': int_or_none(comment.get('numDislikes')),
	154	'timestamp': unified_timestamp(comment.get('postedAt'))
	155	}
	156
7eaf7f9a	157	pages_total = int_or_none(raw_comments.get('totalPages')) or None
	158	is_last = raw_comments.get('last')
	159	if not raw_comments.get('content') or is_last or (page_n > pages_total if pages_total else is_last is not False):
334b1c48	160	return
7eaf7f9a	161
	162
	163	class RokfinPlaylistBaseIE(InfoExtractor):
	164	_TYPES = {
	165	'video': 'post',
	166	'audio': 'post',
	167	'stream': 'stream',
	168	'dead_stream': 'stream',
	169	'stack': 'stack',
	170	}
	171
	172	def _get_video_data(self, metadata):
	173	for content in metadata.get('content') or []:
	174	media_type = self._TYPES.get(content.get('mediaType'))
	175	video_id = content.get('id') if media_type == 'post' else content.get('mediaId')
	176	if not media_type or not video_id:
	177	continue
	178
	179	yield self.url_result(f'https://rokfin.com/{media_type}/{video_id}', video_id=f'{media_type}/{video_id}',
	180	video_title=str_or_none(traverse_obj(content, ('content', 'contentTitle'))))
	181
	182
	183	class RokfinStackIE(RokfinPlaylistBaseIE):
	184	IE_NAME = 'rokfin:stack'
	185	_VALID_URL = r'https?://(?:www\.)?rokfin\.com/stack/(?P<id>[^/]+)'
	186	_TESTS = [{
	187	'url': 'https://www.rokfin.com/stack/271/Tulsi-Gabbard-Portsmouth-Townhall-FULL--Feb-9-2020',
	188	'playlist_count': 8,
	189	'info_dict': {
	190	'id': '271',
	191	},
	192	}]
	193
	194	def _real_extract(self, url):
	195	list_id = self._match_id(url)
	196	return self.playlist_result(self._get_video_data(
	197	self._download_json(f'{_API_BASE_URL}stack/{list_id}', list_id)), list_id)
	198
	199
	200	class RokfinChannelIE(RokfinPlaylistBaseIE):
	201	IE_NAME = 'rokfin:channel'
	202	_VALID_URL = r'https?://(?:www\.)?rokfin\.com/(?!((feed/?)\|(discover/?)\|(channels/?))$)(?P<id>[^/]+)/?$'
	203	_TESTS = [{
	204	'url': 'https://rokfin.com/TheConvoCouch',
	205	'playlist_mincount': 100,
	206	'info_dict': {
	207	'id': '12071-new',
	208	'title': 'TheConvoCouch - New',
	209	'description': 'md5:bb622b1bca100209b91cd685f7847f06',
	210	},
	211	}]
	212
	213	_TABS = {
	214	'new': 'posts',
	215	'top': 'top',
	216	'videos': 'video',
	217	'podcasts': 'audio',
	218	'streams': 'stream',
	219	'stacks': 'stack',
	220	}
	221
	222	def _real_initialize(self):
	223	self._validate_extractor_args()
	224
225	def _validate_extractor_args(self):
226	requested_tabs = self._configuration_arg('tab', None)
227	if requested_tabs is not None and (len(requested_tabs) > 1 or requested_tabs[0] not in self._TABS):
228	raise ExtractorError(f'Invalid extractor-arg "tab". Must be one of {", ".join(self._TABS)}', expected=True)
229
230	def _entries(self, channel_id, channel_name, tab):
231	pages_total = None
232	for page_n in itertools.count(0):
233	if tab in ('posts', 'top'):
234	data_url = f'{_API_BASE_URL}user/{channel_name}/{tab}?page={page_n}&size=50'
235	else:
236	data_url = f'{_API_BASE_URL}post/search/{tab}?page={page_n}&size=50&creator={channel_id}'
237	metadata = self._download_json(
238	data_url, channel_name,
239	note=f'Downloading video metadata page {page_n + 1}{format_field(pages_total, template=" of %s")}')
240
241	yield from self._get_video_data(metadata)
242	pages_total = int_or_none(metadata.get('totalPages')) or None
243	is_last = metadata.get('last')
244	if is_last or (page_n > pages_total if pages_total else is_last is not False):
245	return
246
247	def _real_extract(self, url):
248	channel_name = self._match_id(url)
249	channel_info = self._download_json(f'{_API_BASE_URL}user/{channel_name}', channel_name)
250	channel_id = channel_info['id']
251	tab = self._configuration_arg('tab', default=['new'])[0]
252
253	return self.playlist_result(
254	self._entries(channel_id, channel_name, self._TABS[tab]),
255	f'{channel_id}-{tab}', f'{channel_name} - {tab.title()}', str_or_none(channel_info.get('description')))