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

import base64
import itertools
import re

from .common import InfoExtractor
from ..dependencies import websockets
from ..utils import (
    ExtractorError,
    UserNotLive,
    clean_html,
    float_or_none,
    get_element_by_class,
    get_element_by_id,
    parse_duration,
    qualities,
    str_to_int,
    traverse_obj,
    try_get,
    unified_timestamp,
    urlencode_postdata,
    urljoin,
)


class TwitCastingIE(InfoExtractor):
    _VALID_URL = r'https?://(?:[^/?#]+\.)?twitcasting\.tv/(?P<uploader_id>[^/?#]+)/(?:movie|twplayer)/(?P<id>\d+)'
    _M3U8_HEADERS = {
        'Origin': 'https://twitcasting.tv',
        'Referer': 'https://twitcasting.tv/',
    }
    _TESTS = [{
        'url': 'https://twitcasting.tv/ivetesangalo/movie/2357609',
        'md5': '745243cad58c4681dc752490f7540d7f',
        'info_dict': {
            'id': '2357609',
            'ext': 'mp4',
            'title': 'Live #2357609',
            'uploader_id': 'ivetesangalo',
            'description': 'Twitter Oficial da cantora brasileira Ivete Sangalo.',
            'thumbnail': r're:^https?://.*\.jpg$',
            'upload_date': '20110822',
            'timestamp': 1313978424,
            'duration': 32,
            'view_count': int,
        },
        'params': {
            'skip_download': True,
        },
    }, {
        'url': 'https://twitcasting.tv/mttbernardini/movie/3689740',
        'info_dict': {
            'id': '3689740',
            'ext': 'mp4',
            'title': 'Live playing something #3689740',
            'uploader_id': 'mttbernardini',
            'description': 'md5:1dc7efa2f1ab932fcd119265cebeec69',
            'thumbnail': r're:^https?://.*\.jpg$',
            'upload_date': '20120211',
            'timestamp': 1328995624,
            'duration': 681,
            'view_count': int,
        },
        'params': {
            'skip_download': True,
            'videopassword': 'abc',
        },
    }, {
        'url': 'https://twitcasting.tv/loft_heaven/movie/685979292',
        'info_dict': {
            'id': '685979292',
            'ext': 'mp4',
            'title': '【無料配信】南波一海のhear/here “ナタリー望月哲さんに聞く編集と「渋谷系狂騒曲」”',
            'uploader_id': 'loft_heaven',
            'description': 'md5:3a0c7b53019df987ce545c935538bacf',
            'upload_date': '20210604',
            'timestamp': 1622802114,
            'thumbnail': r're:^https?://.*\.jpg$',
            'duration': 6964,
            'view_count': int,
        },
        'params': {
            'skip_download': True,
        },
    }]

    def _parse_data_movie_playlist(self, dmp, video_id):
        # attempt 1: parse as JSON directly
        try:
            return self._parse_json(dmp, video_id)
        except ExtractorError:
            pass
        # attempt 2: decode reversed base64
        decoded = base64.b64decode(dmp[::-1])
        return self._parse_json(decoded, video_id)

    def _real_extract(self, url):
        uploader_id, video_id = self._match_valid_url(url).groups()

        webpage, urlh = self._download_webpage_handle(url, video_id)
        video_password = self.get_param('videopassword')
        request_data = None
        if video_password:
            request_data = urlencode_postdata({
                'password': video_password,
                **self._hidden_inputs(webpage),
            }, encoding='utf-8')
            webpage, urlh = self._download_webpage_handle(
                url, video_id, data=request_data,
                headers={'Origin': 'https://twitcasting.tv'},
                note='Trying video password')
        if urlh.url != url and request_data:
            webpage = self._download_webpage(
                urlh.url, video_id, data=request_data,
                headers={'Origin': 'https://twitcasting.tv'},
                note='Retrying authentication')
        # has to check here as the first request can contain password input form even if the password is correct
        if re.search(r'<form\s+method="POST">\s*<input\s+[^>]+?name="password"', webpage):
            raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)

        title = (clean_html(get_element_by_id('movietitle', webpage))
                 or self._html_search_meta(['og:title', 'twitter:title'], webpage, fatal=True))

        video_js_data = try_get(
            webpage,
            lambda x: self._parse_data_movie_playlist(self._search_regex(
                r'data-movie-playlist=\'([^\']+?)\'',
                x, 'movie playlist', default=None), video_id)['2'], list)

        thumbnail = traverse_obj(video_js_data, (0, 'thumbnailUrl')) or self._og_search_thumbnail(webpage)
        description = clean_html(get_element_by_id(
            'authorcomment', webpage)) or self._html_search_meta(
            ['description', 'og:description', 'twitter:description'], webpage)
        duration = (try_get(video_js_data, lambda x: sum(float_or_none(y.get('duration')) for y in x) / 1000)
                    or parse_duration(clean_html(get_element_by_class('tw-player-duration-time', webpage))))
        view_count = str_to_int(self._search_regex(
            (r'Total\s*:\s*Views\s*([\d,]+)', r'総視聴者\s*:\s*([\d,]+)\s*</'), webpage, 'views', None))
        timestamp = unified_timestamp(self._search_regex(
            r'data-toggle="true"[^>]+datetime="([^"]+)"',
            webpage, 'datetime', None))

        stream_server_data = self._download_json(
            f'https://twitcasting.tv/streamserver.php?target={uploader_id}&mode=client', video_id,
            'Downloading live info', fatal=False)

        is_live = any(f'data-{x}' in webpage for x in ['is-onlive="true"', 'live-type="live"', 'status="online"'])
        if not traverse_obj(stream_server_data, 'llfmp4') and is_live:
            self.raise_login_required(method='cookies')

        base_dict = {
            'title': title,
            'description': description,
            'thumbnail': thumbnail,
            'timestamp': timestamp,
            'uploader_id': uploader_id,
            'duration': duration,
            'view_count': view_count,
            'is_live': is_live,
        }

        def find_dmu(x):
            data_movie_url = self._search_regex(
                r'data-movie-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
                x, 'm3u8 url', group='url', default=None)
            if data_movie_url:
                return [data_movie_url]

        m3u8_urls = (try_get(webpage, find_dmu, list)
                     or traverse_obj(video_js_data, (..., 'source', 'url'))
                     or ([f'https://twitcasting.tv/{uploader_id}/metastream.m3u8'] if is_live else None))
        if not m3u8_urls:
            raise ExtractorError('Failed to get m3u8 playlist')

        if is_live:
            m3u8_url = m3u8_urls[0]
            formats = self._extract_m3u8_formats(
                m3u8_url, video_id, ext='mp4', m3u8_id='hls',
                live=True, headers=self._M3U8_HEADERS)

            if traverse_obj(stream_server_data, ('hls', 'source')):
                formats.extend(self._extract_m3u8_formats(
                    m3u8_url, video_id, ext='mp4', m3u8_id='source',
                    live=True, query={'mode': 'source'},
                    note='Downloading source quality m3u8',
                    headers=self._M3U8_HEADERS, fatal=False))

            if websockets:
                qq = qualities(['base', 'mobilesource', 'main'])
                streams = traverse_obj(stream_server_data, ('llfmp4', 'streams')) or {}
                for mode, ws_url in streams.items():
                    formats.append({
                        'url': ws_url,
                        'format_id': f'ws-{mode}',
                        'ext': 'mp4',
                        'quality': qq(mode),
                        'source_preference': -10,
                        # TwitCasting simply sends moof atom directly over WS
                        'protocol': 'websocket_frag',
                    })

            infodict = {
                'formats': formats,
                '_format_sort_fields': ('source', ),
            }
        elif len(m3u8_urls) == 1:
            formats = self._extract_m3u8_formats(
                m3u8_urls[0], video_id, 'mp4', headers=self._M3U8_HEADERS)
            infodict = {
                # No problem here since there's only one manifest
                'formats': formats,
                'http_headers': self._M3U8_HEADERS,
            }
        else:
            infodict = {
                '_type': 'multi_video',
                'entries': [{
                    'id': f'{video_id}-{num}',
                    'url': m3u8_url,
                    'ext': 'mp4',
                    # Requesting the manifests here will cause download to fail.
                    # So use ffmpeg instead. See: https://github.com/yt-dlp/yt-dlp/issues/382
                    'protocol': 'm3u8',
                    'http_headers': self._M3U8_HEADERS,
                    **base_dict,
                } for (num, m3u8_url) in enumerate(m3u8_urls)],
            }

        return {
            'id': video_id,
            **base_dict,
            **infodict,
        }


class TwitCastingLiveIE(InfoExtractor):
    _VALID_URL = r'https?://(?:[^/?#]+\.)?twitcasting\.tv/(?P<id>[^/?#]+)/?(?:[#?]|$)'
    _TESTS = [{
        'url': 'https://twitcasting.tv/ivetesangalo',
        'only_matching': True,
    }, {
        'url': 'https://twitcasting.tv/c:unusedlive',
        'expected_exception': 'UserNotLive',
    }]

    def _real_extract(self, url):
        uploader_id = self._match_id(url)
        self.to_screen(
            f'Downloading live video of user {uploader_id}. '
            f'Pass "https://twitcasting.tv/{uploader_id}/show" to download the history')

        is_live = traverse_obj(self._download_json(
            f'https://frontendapi.twitcasting.tv/watch/user/{uploader_id}',
            uploader_id, 'Checking live status', data=b'', fatal=False), ('is_live', {bool}))
        if is_live is False:  # only raise here if API response was as expected
            raise UserNotLive(video_id=uploader_id)

        # Use /show/ page so that password-protected and members-only livestreams can be found
        webpage = self._download_webpage(
            f'https://twitcasting.tv/{uploader_id}/show/', uploader_id, 'Downloading live history')
        is_live = is_live or self._search_regex(
            r'(?s)(<span\s*class="tw-movie-thumbnail2-badge"\s*data-status="live">\s*LIVE)',
            webpage, 'is live?', default=False)
        # Current live is always the first match
        current_live = self._search_regex(
            r'(?s)<a\s+class="tw-movie-thumbnail2"\s+href="/[^/"]+/movie/(?P<video_id>\d+)"',
            webpage, 'current live ID', default=None, group='video_id')
        if not is_live or not current_live:
            raise UserNotLive(video_id=uploader_id)

        return self.url_result(f'https://twitcasting.tv/{uploader_id}/movie/{current_live}', TwitCastingIE)


class TwitCastingUserIE(InfoExtractor):
    _VALID_URL = r'https?://(?:[^/?#]+\.)?twitcasting\.tv/(?P<id>[^/?#]+)/(:?show|archive)/?(?:[#?]|$)'
    _TESTS = [{
        'url': 'https://twitcasting.tv/natsuiromatsuri/archive/',
        'info_dict': {
            'id': 'natsuiromatsuri',
            'title': 'natsuiromatsuri - Live History',
        },
        'playlist_mincount': 235,
    }, {
        'url': 'https://twitcasting.tv/noriyukicas/show',
        'only_matching': True,
    }]

    def _entries(self, uploader_id):
        base_url = next_url = f'https://twitcasting.tv/{uploader_id}/show'
        for page_num in itertools.count(1):
            webpage = self._download_webpage(
                next_url, uploader_id, query={'filter': 'watchable'}, note=f'Downloading page {page_num}')
            matches = re.finditer(
                r'(?s)<a\s+class="tw-movie-thumbnail2"\s+href="(?P<url>/[^/"]+/movie/\d+)"', webpage)
            for mobj in matches:
                yield self.url_result(urljoin(base_url, mobj.group('url')))

            next_url = self._search_regex(
                r'<a href="(/%s/show/%d-\d+)[?"]' % (re.escape(uploader_id), page_num),
                webpage, 'next url', default=None)
            next_url = urljoin(base_url, next_url)
            if not next_url:
                return

    def _real_extract(self, url):
        uploader_id = self._match_id(url)
        return self.playlist_result(
            self._entries(uploader_id), uploader_id, f'{uploader_id} - Live History')
Commit	Line	Data
da9a60ca	1	import base64
e85a3971	2	import itertools
036f9051	3	import re
036f9051	4
29f7c58a	5	from .common import InfoExtractor
9b8ee23b	6	from ..dependencies import websockets
29f7c58a	7	from ..utils import (
a4a42602	8	ExtractorError,
c1d71d0d AW	9	UserNotLive,
c1d71d0d AW	10	clean_html,
29f7c58a	11	float_or_none,
	12	get_element_by_class,
	13	get_element_by_id,
	14	parse_duration,
e6779b94	15	qualities,
29f7c58a	16	str_to_int,
fabb27fc	17	traverse_obj,
e85a3971	18	try_get,
29f7c58a	19	unified_timestamp,
29f7c58a	20	urlencode_postdata,
e85a3971	21	urljoin,
29f7c58a	22	)
29f7c58a	23
036f9051	24
cf0db4d9	25	class TwitCastingIE(InfoExtractor):
cebbd33b	26	_VALID_URL = r'https?://(?:[^/?#]+\.)?twitcasting\.tv/(?P<uploader_id>[^/?#]+)/(?:movie\|twplayer)/(?P<id>\d+)'
a4a42602 LTHD	27	_M3U8_HEADERS = {
	28	'Origin': 'https://twitcasting.tv',
	29	'Referer': 'https://twitcasting.tv/',
	30	}
88b54749	31	_TESTS = [{
036f9051	32	'url': 'https://twitcasting.tv/ivetesangalo/movie/2357609',
	33	'md5': '745243cad58c4681dc752490f7540d7f',
	34	'info_dict': {
	35	'id': '2357609',
	36	'ext': 'mp4',
00a9a25c	37	'title': 'Live #2357609',
036f9051	38	'uploader_id': 'ivetesangalo',
29f7c58a	39	'description': 'Twitter Oficial da cantora brasileira Ivete Sangalo.',
036f9051	40	'thumbnail': r're:^https?://.*\.jpg$',
29f7c58a	41	'upload_date': '20110822',
9a9006ba	42	'timestamp': 1313978424,
29f7c58a	43	'duration': 32,
29f7c58a	44	'view_count': int,
cf0db4d9 S	45	},
	46	'params': {
	47	'skip_download': True,
	48	},
88b54749 MZ	49	}, {
	50	'url': 'https://twitcasting.tv/mttbernardini/movie/3689740',
	51	'info_dict': {
	52	'id': '3689740',
	53	'ext': 'mp4',
	54	'title': 'Live playing something #3689740',
	55	'uploader_id': 'mttbernardini',
9a9006ba	56	'description': 'md5:1dc7efa2f1ab932fcd119265cebeec69',
88b54749	57	'thumbnail': r're:^https?://.*\.jpg$',
9a9006ba S	58	'upload_date': '20120211',
9a9006ba S	59	'timestamp': 1328995624,
29f7c58a	60	'duration': 681,
29f7c58a	61	'view_count': int,
88b54749 MZ	62	},
	63	'params': {
	64	'skip_download': True,
	65	'videopassword': 'abc',
	66	},
a4a42602	67	}, {
a4a42602 LTHD	68	'url': 'https://twitcasting.tv/loft_heaven/movie/685979292',
	69	'info_dict': {
	70	'id': '685979292',
	71	'ext': 'mp4',
9a9006ba S	72	'title': '【無料配信】南波一海のhear/here “ナタリー望月哲さんに聞く編集と「渋谷系狂騒曲」”',
	73	'uploader_id': 'loft_heaven',
	74	'description': 'md5:3a0c7b53019df987ce545c935538bacf',
	75	'upload_date': '20210604',
	76	'timestamp': 1622802114,
	77	'thumbnail': r're:^https?://.*\.jpg$',
	78	'duration': 6964,
	79	'view_count': int,
	80	},
	81	'params': {
	82	'skip_download': True,
a4a42602	83	},
88b54749	84	}]
036f9051	85
da9a60ca L	86	def _parse_data_movie_playlist(self, dmp, video_id):
	87	# attempt 1: parse as JSON directly
	88	try:
	89	return self._parse_json(dmp, video_id)
	90	except ExtractorError:
	91	pass
	92	# attempt 2: decode reversed base64
	93	decoded = base64.b64decode(dmp[::-1])
	94	return self._parse_json(decoded, video_id)
	95
036f9051	96	def _real_extract(self, url):
5ad28e7f	97	uploader_id, video_id = self._match_valid_url(url).groups()
036f9051	98
9a9006ba	99	webpage, urlh = self._download_webpage_handle(url, video_id)
a06916d9	100	video_password = self.get_param('videopassword')
88b54749 MZ	101	request_data = None
	102	if video_password:
	103	request_data = urlencode_postdata({
	104	'password': video_password,
9a9006ba	105	**self._hidden_inputs(webpage),
5a13fdd2	106	}, encoding='utf-8')
9a9006ba S	107	webpage, urlh = self._download_webpage_handle(
	108	url, video_id, data=request_data,
	109	headers={'Origin': 'https://twitcasting.tv'},
	110	note='Trying video password')
3d2623a8	111	if urlh.url != url and request_data:
df635a09	112	webpage = self._download_webpage(
3d2623a8	113	urlh.url, video_id, data=request_data,
df635a09 LNO	114	headers={'Origin': 'https://twitcasting.tv'},
df635a09 LNO	115	note='Retrying authentication')
f099df14 LNO	116	# has to check here as the first request can contain password input form even if the password is correct
	117	if re.search(r'<form\s+method="POST">\s*<input\s+[^>]+?name="password"', webpage):
	118	raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True)
036f9051	119
e85a3971	120	title = (clean_html(get_element_by_id('movietitle', webpage))
e85a3971	121	or self._html_search_meta(['og:title', 'twitter:title'], webpage, fatal=True))
cf0db4d9	122
a4a42602 LTHD	123	video_js_data = try_get(
a4a42602 LTHD	124	webpage,
da9a60ca	125	lambda x: self._parse_data_movie_playlist(self._search_regex(
a4a42602 LTHD	126	r'data-movie-playlist=\'([^\']+?)\'',
a4a42602 LTHD	127	x, 'movie playlist', default=None), video_id)['2'], list)
fabb27fc	128
a4a42602	129	thumbnail = traverse_obj(video_js_data, (0, 'thumbnailUrl')) or self._og_search_thumbnail(webpage)
29f7c58a	130	description = clean_html(get_element_by_id(
	131	'authorcomment', webpage)) or self._html_search_meta(
	132	['description', 'og:description', 'twitter:description'], webpage)
a4a42602 LTHD	133	duration = (try_get(video_js_data, lambda x: sum(float_or_none(y.get('duration')) for y in x) / 1000)
a4a42602 LTHD	134	or parse_duration(clean_html(get_element_by_class('tw-player-duration-time', webpage))))
29f7c58a	135	view_count = str_to_int(self._search_regex(
9a9006ba	136	(r'Total\s:\sViews\s([\d,]+)', r'総視聴者\s:\s([\d,]+)\s</'), webpage, 'views', None))
29f7c58a	137	timestamp = unified_timestamp(self._search_regex(
	138	r'data-toggle="true"[^>]+datetime="([^"]+)"',
	139	webpage, 'datetime', None))
cf0db4d9	140
a4a42602	141	stream_server_data = self._download_json(
add96eb9	142	f'https://twitcasting.tv/streamserver.php?target={uploader_id}&mode=client', video_id,
a4a42602	143	'Downloading live info', fatal=False)
e85a3971	144
7b8b1cf5	145	is_live = any(f'data-{x}' in webpage for x in ['is-onlive="true"', 'live-type="live"', 'status="online"'])
a4a42602 LTHD	146	if not traverse_obj(stream_server_data, 'llfmp4') and is_live:
	147	self.raise_login_required(method='cookies')
	148
	149	base_dict = {
036f9051	150	'title': title,
	151	'description': description,
	152	'thumbnail': thumbnail,
29f7c58a	153	'timestamp': timestamp,
036f9051	154	'uploader_id': uploader_id,
29f7c58a	155	'duration': duration,
29f7c58a	156	'view_count': view_count,
e85a3971	157	'is_live': is_live,
036f9051	158	}
e85a3971	159
a4a42602 LTHD	160	def find_dmu(x):
	161	data_movie_url = self._search_regex(
	162	r'data-movie-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
	163	x, 'm3u8 url', group='url', default=None)
	164	if data_movie_url:
	165	return [data_movie_url]
	166
	167	m3u8_urls = (try_get(webpage, find_dmu, list)
	168	or traverse_obj(video_js_data, (..., 'source', 'url'))
	169	or ([f'https://twitcasting.tv/{uploader_id}/metastream.m3u8'] if is_live else None))
	170	if not m3u8_urls:
	171	raise ExtractorError('Failed to get m3u8 playlist')
	172
	173	if is_live:
	174	m3u8_url = m3u8_urls[0]
	175	formats = self._extract_m3u8_formats(
	176	m3u8_url, video_id, ext='mp4', m3u8_id='hls',
	177	live=True, headers=self._M3U8_HEADERS)
	178
df635a09 LNO	179	if traverse_obj(stream_server_data, ('hls', 'source')):
	180	formats.extend(self._extract_m3u8_formats(
	181	m3u8_url, video_id, ext='mp4', m3u8_id='source',
	182	live=True, query={'mode': 'source'},
	183	note='Downloading source quality m3u8',
	184	headers=self._M3U8_HEADERS, fatal=False))
a4a42602	185
9b8ee23b	186	if websockets:
a4a42602 LTHD	187	qq = qualities(['base', 'mobilesource', 'main'])
	188	streams = traverse_obj(stream_server_data, ('llfmp4', 'streams')) or {}
	189	for mode, ws_url in streams.items():
	190	formats.append({
	191	'url': ws_url,
add96eb9	192	'format_id': f'ws-{mode}',
a4a42602 LTHD	193	'ext': 'mp4',
a4a42602 LTHD	194	'quality': qq(mode),
df635a09	195	'source_preference': -10,
a4a42602 LTHD	196	# TwitCasting simply sends moof atom directly over WS
	197	'protocol': 'websocket_frag',
	198	})
	199
a4a42602	200	infodict = {
9f14daf2	201	'formats': formats,
9f14daf2	202	'_format_sort_fields': ('source', ),
a4a42602	203	}
06b1628d LNO	204	elif len(m3u8_urls) == 1:
	205	formats = self._extract_m3u8_formats(
	206	m3u8_urls[0], video_id, 'mp4', headers=self._M3U8_HEADERS)
06b1628d LNO	207	infodict = {
	208	# No problem here since there's only one manifest
	209	'formats': formats,
997378f9	210	'http_headers': self._M3U8_HEADERS,
06b1628d	211	}
a4a42602 LTHD	212	else:
	213	infodict = {
	214	'_type': 'multi_video',
	215	'entries': [{
	216	'id': f'{video_id}-{num}',
	217	'url': m3u8_url,
	218	'ext': 'mp4',
	219	# Requesting the manifests here will cause download to fail.
	220	# So use ffmpeg instead. See: https://github.com/yt-dlp/yt-dlp/issues/382
	221	'protocol': 'm3u8',
	222	'http_headers': self._M3U8_HEADERS,
	223	**base_dict,
	224	} for (num, m3u8_url) in enumerate(m3u8_urls)],
	225	}
	226
	227	return {
	228	'id': video_id,
	229	**base_dict,
	230	**infodict,
	231	}
	232
e85a3971	233
e85a3971	234	class TwitCastingLiveIE(InfoExtractor):
cebbd33b	235	_VALID_URL = r'https?://(?:[^/?#]+\.)?twitcasting\.tv/(?P<id>[^/?#]+)/?(?:[#?]\|$)'
e85a3971	236	_TESTS = [{
	237	'url': 'https://twitcasting.tv/ivetesangalo',
	238	'only_matching': True,
c1d71d0d AW	239	}, {
	240	'url': 'https://twitcasting.tv/c:unusedlive',
	241	'expected_exception': 'UserNotLive',
e85a3971	242	}]
	243
	244	def _real_extract(self, url):
	245	uploader_id = self._match_id(url)
	246	self.to_screen(
add96eb9	247	f'Downloading live video of user {uploader_id}. '
add96eb9	248	f'Pass "https://twitcasting.tv/{uploader_id}/show" to download the history')
e85a3971	249
585d0ed9 JC	250	is_live = traverse_obj(self._download_json(
	251	f'https://frontendapi.twitcasting.tv/watch/user/{uploader_id}',
	252	uploader_id, 'Checking live status', data=b'', fatal=False), ('is_live', {bool}))
	253	if is_live is False: # only raise here if API response was as expected
c1d71d0d	254	raise UserNotLive(video_id=uploader_id)
585d0ed9 JC	255
	256	# Use /show/ page so that password-protected and members-only livestreams can be found
	257	webpage = self._download_webpage(
	258	f'https://twitcasting.tv/{uploader_id}/show/', uploader_id, 'Downloading live history')
	259	is_live = is_live or self._search_regex(
	260	r'(?s)(<span\sclass="tw-movie-thumbnail2-badge"\sdata-status="live">\s*LIVE)',
	261	webpage, 'is live?', default=False)
	262	# Current live is always the first match
	263	current_live = self._search_regex(
	264	r'(?s)<a\s+class="tw-movie-thumbnail2"\s+href="/[^/"]+/movie/(?P<video_id>\d+)"',
	265	webpage, 'current live ID', default=None, group='video_id')
	266	if not is_live or not current_live:
	267	raise UserNotLive(video_id=uploader_id)
	268
2325d03a	269	return self.url_result(f'https://twitcasting.tv/{uploader_id}/movie/{current_live}', TwitCastingIE)
e85a3971	270
	271
	272	class TwitCastingUserIE(InfoExtractor):
cebbd33b	273	_VALID_URL = r'https?://(?:[^/?#]+\.)?twitcasting\.tv/(?P<id>[^/?#]+)/(:?show\|archive)/?(?:[#?]\|$)'
e85a3971	274	_TESTS = [{
cebbd33b	275	'url': 'https://twitcasting.tv/natsuiromatsuri/archive/',
	276	'info_dict': {
	277	'id': 'natsuiromatsuri',
	278	'title': 'natsuiromatsuri - Live History',
	279	},
	280	'playlist_mincount': 235,
	281	}, {
e85a3971	282	'url': 'https://twitcasting.tv/noriyukicas/show',
	283	'only_matching': True,
	284	}]
	285
	286	def _entries(self, uploader_id):
add96eb9	287	base_url = next_url = f'https://twitcasting.tv/{uploader_id}/show'
e85a3971	288	for page_num in itertools.count(1):
e85a3971	289	webpage = self._download_webpage(
add96eb9	290	next_url, uploader_id, query={'filter': 'watchable'}, note=f'Downloading page {page_num}')
e85a3971	291	matches = re.finditer(
ff2fde1b	292	r'(?s)<a\s+class="tw-movie-thumbnail2"\s+href="(?P<url>/[^/"]+/movie/\d+)"', webpage)
e85a3971	293	for mobj in matches:
	294	yield self.url_result(urljoin(base_url, mobj.group('url')))
	295
	296	next_url = self._search_regex(
	297	r'<a href="(/%s/show/%d-\d+)[?"]' % (re.escape(uploader_id), page_num),
	298	webpage, 'next url', default=None)
	299	next_url = urljoin(base_url, next_url)
	300	if not next_url:
	301	return
	302
	303	def _real_extract(self, url):
	304	uploader_id = self._match_id(url)
	305	return self.playlist_result(
add96eb9	306	self._entries(uploader_id), uploader_id, f'{uploader_id} - Live History')