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

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from .youtube import YoutubeIE
from ..utils import (
    ExtractorError,
    int_or_none,
    str_or_none,
)


class VVVVIDIE(InfoExtractor):
    _VALID_URL_BASE = r'https?://(?:www\.)?vvvvid\.it/(?:#!)?(?:show|anime|film|series)/'
    _VALID_URL = r'%s(?P<show_id>\d+)/[^/]+/(?P<season_id>\d+)/(?P<id>[0-9]+)' % _VALID_URL_BASE
    _TESTS = [{
        # video_type == 'video/vvvvid'
        'url': 'https://www.vvvvid.it/#!show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048/ping-pong',
        'md5': 'b8d3cecc2e981adc3835adf07f6df91b',
        'info_dict': {
            'id': '489048',
            'ext': 'mp4',
            'title': 'Ping Pong',
            'duration': 239,
            'series': '"Perché dovrei guardarlo?" di Dario Moccia',
            'season_id': '437',
            'episode': 'Ping Pong',
            'episode_number': 1,
            'episode_id': '3334',
            'view_count': int,
            'like_count': int,
            'repost_count': int,
        },
        'params': {
            'skip_download': True,
        },
    }, {
        # video_type == 'video/rcs'
        'url': 'https://www.vvvvid.it/#!show/376/death-note-live-action/377/482493/episodio-01',
        'md5': '33e0edfba720ad73a8782157fdebc648',
        'info_dict': {
            'id': '482493',
            'ext': 'mp4',
            'title': 'Episodio 01',
        },
        'params': {
            'skip_download': True,
        },
    }, {
        # video_type == 'video/youtube'
        'url': 'https://www.vvvvid.it/show/404/one-punch-man/406/486683/trailer',
        'md5': '33e0edfba720ad73a8782157fdebc648',
        'info_dict': {
            'id': 'RzmFKUDOUgw',
            'ext': 'mp4',
            'title': 'Trailer',
            'upload_date': '20150906',
            'description': 'md5:a5e802558d35247fee285875328c0b80',
            'uploader_id': 'BandaiVisual',
            'uploader': 'BANDAI NAMCO Arts Channel',
        },
        'params': {
            'skip_download': True,
        },
    }, {
        'url': 'https://www.vvvvid.it/show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048',
        'only_matching': True
    }]
    _conn_id = None

    def _real_initialize(self):
        self._conn_id = self._download_json(
            'https://www.vvvvid.it/user/login',
            None, headers=self.geo_verification_headers())['data']['conn_id']

    def _download_info(self, show_id, path, video_id, fatal=True, query=None):
        q = {
            'conn_id': self._conn_id,
        }
        if query:
            q.update(query)
        response = self._download_json(
            'https://www.vvvvid.it/vvvvid/ondemand/%s/%s' % (show_id, path),
            video_id, headers=self.geo_verification_headers(), query=q, fatal=fatal)
        if not (response or fatal):
            return
        if response.get('result') == 'error':
            raise ExtractorError('%s said: %s' % (
                self.IE_NAME, response['message']), expected=True)
        return response['data']

    def _extract_common_video_info(self, video_data):
        return {
            'thumbnail': video_data.get('thumbnail'),
            'episode_id': str_or_none(video_data.get('id')),
        }

    def _real_extract(self, url):
        show_id, season_id, video_id = re.match(self._VALID_URL, url).groups()

        response = self._download_info(
            show_id, 'season/%s' % season_id,
            video_id, query={'video_id': video_id})

        vid = int(video_id)
        video_data = list(filter(
            lambda episode: episode.get('video_id') == vid, response))[0]
        title = video_data['title']
        formats = []

        # vvvvid embed_info decryption algorithm is reverse engineered from function $ds(h) at vvvvid.js
        def ds(h):
            g = "MNOPIJKL89+/4567UVWXQRSTEFGHABCDcdefYZabstuvopqr0123wxyzklmnghij"

            def f(m):
                l = []
                o = 0
                b = False
                m_len = len(m)
                while ((not b) and o < m_len):
                    n = m[o] << 2
                    o += 1
                    k = -1
                    j = -1
                    if o < m_len:
                        n += m[o] >> 4
                        o += 1
                        if o < m_len:
                            k = (m[o - 1] << 4) & 255
                            k += m[o] >> 2
                            o += 1
                            if o < m_len:
                                j = (m[o - 1] << 6) & 255
                                j += m[o]
                                o += 1
                            else:
                                b = True
                        else:
                            b = True
                    else:
                        b = True
                    l.append(n)
                    if k != -1:
                        l.append(k)
                    if j != -1:
                        l.append(j)
                return l

            c = []
            for e in h:
                c.append(g.index(e))

            c_len = len(c)
            for e in range(c_len * 2 - 1, -1, -1):
                a = c[e % c_len] ^ c[(e + 1) % c_len]
                c[e % c_len] = a

            c = f(c)
            d = ''
            for e in c:
                d += chr(e)

            return d

        info = {}

        def metadata_from_url(r_url):
            if not info and r_url:
                mobj = re.search(r'_(?:S(\d+))?Ep(\d+)', r_url)
                if mobj:
                    info['episode_number'] = int(mobj.group(2))
                    season_number = mobj.group(1)
                    if season_number:
                        info['season_number'] = int(season_number)

        video_type = video_data.get('video_type')
        is_youtube = False
        for quality in ('', '_sd'):
            embed_code = video_data.get('embed_info' + quality)
            if not embed_code:
                continue
            embed_code = ds(embed_code)
            if video_type == 'video/kenc':
                embed_code = re.sub(r'https?(://[^/]+)/z/', r'https\1/i/', embed_code).replace('/manifest.f4m', '/master.m3u8')
                kenc = self._download_json(
                    'https://www.vvvvid.it/kenc', video_id, query={
                        'action': 'kt',
                        'conn_id': self._conn_id,
                        'url': embed_code,
                    }, fatal=False) or {}
                kenc_message = kenc.get('message')
                if kenc_message:
                    embed_code += '?' + ds(kenc_message)
                formats.extend(self._extract_m3u8_formats(
                    embed_code, video_id, 'mp4', m3u8_id='hls', fatal=False))
            elif video_type == 'video/rcs':
                formats.extend(self._extract_akamai_formats(embed_code, video_id))
            elif video_type == 'video/youtube':
                info.update({
                    '_type': 'url_transparent',
                    'ie_key': YoutubeIE.ie_key(),
                    'url': embed_code,
                })
                is_youtube = True
                break
            else:
                formats.extend(self._extract_wowza_formats(
                    'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
            metadata_from_url(embed_code)

        if not is_youtube:
            self._sort_formats(formats)
            info['formats'] = formats

        metadata_from_url(video_data.get('thumbnail'))
        info.update(self._extract_common_video_info(video_data))
        info.update({
            'id': video_id,
            'title': title,
            'duration': int_or_none(video_data.get('length')),
            'series': video_data.get('show_title'),
            'season_id': season_id,
            'episode': title,
            'view_count': int_or_none(video_data.get('views')),
            'like_count': int_or_none(video_data.get('video_likes')),
            'repost_count': int_or_none(video_data.get('video_shares')),
        })
        return info


class VVVVIDShowIE(VVVVIDIE):
    _VALID_URL = r'(?P<base_url>%s(?P<id>\d+)(?:/(?P<show_title>[^/?&#]+))?)/?(?:[?#&]|$)' % VVVVIDIE._VALID_URL_BASE
    _TESTS = [{
        'url': 'https://www.vvvvid.it/show/156/psyco-pass',
        'info_dict': {
            'id': '156',
            'title': 'Psycho-Pass',
            'description': 'md5:94d572c0bd85894b193b8aebc9a3a806',
        },
        'playlist_count': 46,
    }, {
        'url': 'https://www.vvvvid.it/show/156',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        base_url, show_id, show_title = re.match(self._VALID_URL, url).groups()

        seasons = self._download_info(
            show_id, 'seasons/', show_title)

        show_info = self._download_info(
            show_id, 'info/', show_title, fatal=False)

        if not show_title:
            base_url += "/title"

        entries = []
        for season in (seasons or []):
            episodes = season.get('episodes') or []
            playlist_title = season.get('name') or show_info.get('title')
            for episode in episodes:
                if episode.get('playable') is False:
                    continue
                season_id = str_or_none(episode.get('season_id'))
                video_id = str_or_none(episode.get('video_id'))
                if not (season_id and video_id):
                    continue
                info = self._extract_common_video_info(episode)
                info.update({
                    '_type': 'url_transparent',
                    'ie_key': VVVVIDIE.ie_key(),
                    'url': '/'.join([base_url, season_id, video_id]),
                    'title': episode.get('title'),
                    'description': episode.get('description'),
                    'season_id': season_id,
                    'playlist_title': playlist_title,
                })
                entries.append(info)

        return self.playlist_result(
            entries, show_id, show_info.get('title'), show_info.get('description'))
Commit	Line	Data
7b1e8079 RA	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
	4	import re
	5
	6	from .common import InfoExtractor
2181983a	7	from .youtube import YoutubeIE
7b1e8079 RA	8	from ..utils import (
	9	ExtractorError,
	10	int_or_none,
	11	str_or_none,
	12	)
	13
	14
	15	class VVVVIDIE(InfoExtractor):
29f7c58a	16	_VALID_URL_BASE = r'https?://(?:www\.)?vvvvid\.it/(?:#!)?(?:show\|anime\|film\|series)/'
29f7c58a	17	_VALID_URL = r'%s(?P<show_id>\d+)/[^/]+/(?P<season_id>\d+)/(?P<id>[0-9]+)' % _VALID_URL_BASE
7b1e8079 RA	18	_TESTS = [{
	19	# video_type == 'video/vvvvid'
	20	'url': 'https://www.vvvvid.it/#!show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048/ping-pong',
	21	'md5': 'b8d3cecc2e981adc3835adf07f6df91b',
	22	'info_dict': {
	23	'id': '489048',
	24	'ext': 'mp4',
	25	'title': 'Ping Pong',
29f7c58a	26	'duration': 239,
	27	'series': '"Perché dovrei guardarlo?" di Dario Moccia',
	28	'season_id': '437',
	29	'episode': 'Ping Pong',
	30	'episode_number': 1,
	31	'episode_id': '3334',
	32	'view_count': int,
	33	'like_count': int,
	34	'repost_count': int,
7b1e8079	35	},
fb61b57d RA	36	'params': {
	37	'skip_download': True,
	38	},
7b1e8079 RA	39	}, {
	40	# video_type == 'video/rcs'
	41	'url': 'https://www.vvvvid.it/#!show/376/death-note-live-action/377/482493/episodio-01',
	42	'md5': '33e0edfba720ad73a8782157fdebc648',
	43	'info_dict': {
	44	'id': '482493',
	45	'ext': 'mp4',
	46	'title': 'Episodio 01',
	47	},
fb61b57d RA	48	'params': {
	49	'skip_download': True,
	50	},
2181983a	51	}, {
	52	# video_type == 'video/youtube'
	53	'url': 'https://www.vvvvid.it/show/404/one-punch-man/406/486683/trailer',
	54	'md5': '33e0edfba720ad73a8782157fdebc648',
	55	'info_dict': {
	56	'id': 'RzmFKUDOUgw',
	57	'ext': 'mp4',
	58	'title': 'Trailer',
	59	'upload_date': '20150906',
	60	'description': 'md5:a5e802558d35247fee285875328c0b80',
	61	'uploader_id': 'BandaiVisual',
	62	'uploader': 'BANDAI NAMCO Arts Channel',
	63	},
	64	'params': {
	65	'skip_download': True,
	66	},
29f7c58a	67	}, {
	68	'url': 'https://www.vvvvid.it/show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048',
	69	'only_matching': True
7b1e8079 RA	70	}]
	71	_conn_id = None
	72
	73	def _real_initialize(self):
dc1f3a9f RA	74	self._conn_id = self._download_json(
	75	'https://www.vvvvid.it/user/login',
	76	None, headers=self.geo_verification_headers())['data']['conn_id']
7b1e8079	77
ec5e77c5	78	def _download_info(self, show_id, path, video_id, fatal=True, query=None):
	79	q = {
	80	'conn_id': self._conn_id,
	81	}
	82	if query:
	83	q.update(query)
7b1e8079	84	response = self._download_json(
29f7c58a	85	'https://www.vvvvid.it/vvvvid/ondemand/%s/%s' % (show_id, path),
ec5e77c5	86	video_id, headers=self.geo_verification_headers(), query=q, fatal=fatal)
29f7c58a	87	if not (response or fatal):
	88	return
	89	if response.get('result') == 'error':
7b1e8079 RA	90	raise ExtractorError('%s said: %s' % (
7b1e8079 RA	91	self.IE_NAME, response['message']), expected=True)
29f7c58a	92	return response['data']
	93
	94	def _extract_common_video_info(self, video_data):
	95	return {
	96	'thumbnail': video_data.get('thumbnail'),
	97	'episode_id': str_or_none(video_data.get('id')),
	98	}
	99
	100	def _real_extract(self, url):
	101	show_id, season_id, video_id = re.match(self._VALID_URL, url).groups()
	102
	103	response = self._download_info(
ec5e77c5	104	show_id, 'season/%s' % season_id,
ec5e77c5	105	video_id, query={'video_id': video_id})
7b1e8079 RA	106
	107	vid = int(video_id)
	108	video_data = list(filter(
29f7c58a	109	lambda episode: episode.get('video_id') == vid, response))[0]
29f7c58a	110	title = video_data['title']
7b1e8079 RA	111	formats = []
	112
	113	# vvvvid embed_info decryption algorithm is reverse engineered from function $ds(h) at vvvvid.js
	114	def ds(h):
	115	g = "MNOPIJKL89+/4567UVWXQRSTEFGHABCDcdefYZabstuvopqr0123wxyzklmnghij"
	116
	117	def f(m):
	118	l = []
	119	o = 0
	120	b = False
	121	m_len = len(m)
	122	while ((not b) and o < m_len):
	123	n = m[o] << 2
	124	o += 1
	125	k = -1
	126	j = -1
	127	if o < m_len:
	128	n += m[o] >> 4
	129	o += 1
	130	if o < m_len:
	131	k = (m[o - 1] << 4) & 255
	132	k += m[o] >> 2
	133	o += 1
	134	if o < m_len:
	135	j = (m[o - 1] << 6) & 255
	136	j += m[o]
	137	o += 1
	138	else:
	139	b = True
	140	else:
	141	b = True
	142	else:
	143	b = True
	144	l.append(n)
	145	if k != -1:
	146	l.append(k)
	147	if j != -1:
	148	l.append(j)
	149	return l
	150
	151	c = []
	152	for e in h:
	153	c.append(g.index(e))
	154
	155	c_len = len(c)
	156	for e in range(c_len * 2 - 1, -1, -1):
	157	a = c[e % c_len] ^ c[(e + 1) % c_len]
	158	c[e % c_len] = a
	159
	160	c = f(c)
	161	d = ''
	162	for e in c:
	163	d += chr(e)
	164
	165	return d
	166
29f7c58a	167	info = {}
	168
	169	def metadata_from_url(r_url):
	170	if not info and r_url:
	171	mobj = re.search(r'_(?:S(\d+))?Ep(\d+)', r_url)
	172	if mobj:
	173	info['episode_number'] = int(mobj.group(2))
	174	season_number = mobj.group(1)
	175	if season_number:
	176	info['season_number'] = int(season_number)
	177
2181983a	178	video_type = video_data.get('video_type')
	179	is_youtube = False
	180	for quality in ('', '_sd'):
7b1e8079 RA	181	embed_code = video_data.get('embed_info' + quality)
	182	if not embed_code:
	183	continue
	184	embed_code = ds(embed_code)
421a4595	185	if video_type == 'video/kenc':
	186	embed_code = re.sub(r'https?(://[^/]+)/z/', r'https\1/i/', embed_code).replace('/manifest.f4m', '/master.m3u8')
	187	kenc = self._download_json(
	188	'https://www.vvvvid.it/kenc', video_id, query={
	189	'action': 'kt',
	190	'conn_id': self._conn_id,
	191	'url': embed_code,
	192	}, fatal=False) or {}
	193	kenc_message = kenc.get('message')
	194	if kenc_message:
	195	embed_code += '?' + ds(kenc_message)
	196	formats.extend(self._extract_m3u8_formats(
	197	embed_code, video_id, 'mp4', m3u8_id='hls', fatal=False))
	198	elif video_type == 'video/rcs':
29f7c58a	199	formats.extend(self._extract_akamai_formats(embed_code, video_id))
2181983a	200	elif video_type == 'video/youtube':
	201	info.update({
	202	'_type': 'url_transparent',
	203	'ie_key': YoutubeIE.ie_key(),
	204	'url': embed_code,
	205	})
	206	is_youtube = True
	207	break
7b1e8079 RA	208	else:
	209	formats.extend(self._extract_wowza_formats(
	210	'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
29f7c58a	211	metadata_from_url(embed_code)
29f7c58a	212
2181983a	213	if not is_youtube:
	214	self._sort_formats(formats)
	215	info['formats'] = formats
7b1e8079	216
29f7c58a	217	metadata_from_url(video_data.get('thumbnail'))
	218	info.update(self._extract_common_video_info(video_data))
	219	info.update({
7b1e8079	220	'id': video_id,
29f7c58a	221	'title': title,
7b1e8079 RA	222	'duration': int_or_none(video_data.get('length')),
	223	'series': video_data.get('show_title'),
	224	'season_id': season_id,
29f7c58a	225	'episode': title,
7b1e8079 RA	226	'view_count': int_or_none(video_data.get('views')),
7b1e8079 RA	227	'like_count': int_or_none(video_data.get('video_likes')),
29f7c58a	228	'repost_count': int_or_none(video_data.get('video_shares')),
	229	})
	230	return info
	231
	232
	233	class VVVVIDShowIE(VVVVIDIE):
	234	_VALID_URL = r'(?P<base_url>%s(?P<id>\d+)(?:/(?P<show_title>[^/?&#]+))?)/?(?:[?#&]\|$)' % VVVVIDIE._VALID_URL_BASE
	235	_TESTS = [{
	236	'url': 'https://www.vvvvid.it/show/156/psyco-pass',
	237	'info_dict': {
	238	'id': '156',
	239	'title': 'Psycho-Pass',
	240	'description': 'md5:94d572c0bd85894b193b8aebc9a3a806',
	241	},
	242	'playlist_count': 46,
	243	}, {
	244	'url': 'https://www.vvvvid.it/show/156',
	245	'only_matching': True,
	246	}]
	247
	248	def _real_extract(self, url):
	249	base_url, show_id, show_title = re.match(self._VALID_URL, url).groups()
	250
	251	seasons = self._download_info(
	252	show_id, 'seasons/', show_title)
	253
	254	show_info = self._download_info(
	255	show_id, 'info/', show_title, fatal=False)
	256
ec5e77c5	257	if not show_title:
	258	base_url += "/title"
	259
29f7c58a	260	entries = []
	261	for season in (seasons or []):
	262	episodes = season.get('episodes') or []
ec5e77c5	263	playlist_title = season.get('name') or show_info.get('title')
29f7c58a	264	for episode in episodes:
	265	if episode.get('playable') is False:
	266	continue
	267	season_id = str_or_none(episode.get('season_id'))
	268	video_id = str_or_none(episode.get('video_id'))
	269	if not (season_id and video_id):
	270	continue
	271	info = self._extract_common_video_info(episode)
	272	info.update({
ec5e77c5	273	'_type': 'url_transparent',
29f7c58a	274	'ie_key': VVVVIDIE.ie_key(),
	275	'url': '/'.join([base_url, season_id, video_id]),
	276	'title': episode.get('title'),
	277	'description': episode.get('description'),
	278	'season_id': season_id,
ec5e77c5	279	'playlist_title': playlist_title,
29f7c58a	280	})
	281	entries.append(info)
	282
	283	return self.playlist_result(
	284	entries, show_id, show_info.get('title'), show_info.get('description'))