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

# coding: utf-8
from __future__ import unicode_literals

import re

from .common import InfoExtractor
from ..compat import (
    compat_str,
    compat_urlparse,
)
from ..utils import (
    ExtractorError,
    int_or_none,
    qualities,
    try_get,
    unified_strdate,
    url_or_none,
)


class ArteTVBaseIE(InfoExtractor):
    _ARTE_LANGUAGES = 'fr|de|en|es|it|pl'
    _API_BASE = 'https://api.arte.tv/api/player/v1'


class ArteTVIE(ArteTVBaseIE):
    _VALID_URL = r'''(?x)
                    https?://
                        (?:
                            (?:www\.)?arte\.tv/(?P<lang>%(langs)s)/videos|
                            api\.arte\.tv/api/player/v\d+/config/(?P<lang_2>%(langs)s)
                        )
                        /(?P<id>\d{6}-\d{3}-[AF])
                    ''' % {'langs': ArteTVBaseIE._ARTE_LANGUAGES}
    _TESTS = [{
        'url': 'https://www.arte.tv/en/videos/088501-000-A/mexico-stealing-petrol-to-survive/',
        'info_dict': {
            'id': '088501-000-A',
            'ext': 'mp4',
            'title': 'Mexico: Stealing Petrol to Survive',
            'upload_date': '20190628',
        },
    }, {
        'url': 'https://www.arte.tv/pl/videos/100103-000-A/usa-dyskryminacja-na-porodowce/',
        'only_matching': True,
    }, {
        'url': 'https://api.arte.tv/api/player/v2/config/de/100605-013-A',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        mobj = re.match(self._VALID_URL, url)
        video_id = mobj.group('id')
        lang = mobj.group('lang') or mobj.group('lang_2')

        info = self._download_json(
            '%s/config/%s/%s' % (self._API_BASE, lang, video_id), video_id)
        player_info = info['videoJsonPlayer']

        vsr = try_get(player_info, lambda x: x['VSR'], dict)
        if not vsr:
            error = None
            if try_get(player_info, lambda x: x['custom_msg']['type']) == 'error':
                error = try_get(
                    player_info, lambda x: x['custom_msg']['msg'], compat_str)
            if not error:
                error = 'Video %s is not available' % player_info.get('VID') or video_id
            raise ExtractorError(error, expected=True)

        upload_date_str = player_info.get('shootingDate')
        if not upload_date_str:
            upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]

        title = (player_info.get('VTI') or player_info['VID']).strip()
        subtitle = player_info.get('VSU', '').strip()
        if subtitle:
            title += ' - %s' % subtitle

        qfunc = qualities(['MQ', 'HQ', 'EQ', 'SQ'])

        LANGS = {
            'fr': 'F',
            'de': 'A',
            'en': 'E[ANG]',
            'es': 'E[ESP]',
            'it': 'E[ITA]',
            'pl': 'E[POL]',
        }

        langcode = LANGS.get(lang, lang)

        formats = []
        for format_id, format_dict in vsr.items():
            f = dict(format_dict)
            format_url = url_or_none(f.get('url'))
            streamer = f.get('streamer')
            if not format_url and not streamer:
                continue
            versionCode = f.get('versionCode')
            l = re.escape(langcode)

            # Language preference from most to least priority
            # Reference: section 6.8 of
            # https://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-07-1.pdf
            PREFERENCES = (
                # original version in requested language, without subtitles
                r'VO{0}$'.format(l),
                # original version in requested language, with partial subtitles in requested language
                r'VO{0}-ST{0}$'.format(l),
                # original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
                r'VO{0}-STM{0}$'.format(l),
                # non-original (dubbed) version in requested language, without subtitles
                r'V{0}$'.format(l),
                # non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language
                r'V{0}-ST{0}$'.format(l),
                # non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
                r'V{0}-STM{0}$'.format(l),
                # original version in requested language, with partial subtitles in different language
                r'VO{0}-ST(?!{0}).+?$'.format(l),
                # original version in requested language, with subtitles for the deaf and hard-of-hearing in different language
                r'VO{0}-STM(?!{0}).+?$'.format(l),
                # original version in different language, with partial subtitles in requested language
                r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l),
                # original version in different language, with subtitles for the deaf and hard-of-hearing in requested language
                r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l),
                # original version in different language, without subtitles
                r'VO(?:(?!{0}))?$'.format(l),
                # original version in different language, with partial subtitles in different language
                r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l),
                # original version in different language, with subtitles for the deaf and hard-of-hearing in different language
                r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l),
            )

            for pref, p in enumerate(PREFERENCES):
                if re.match(p, versionCode):
                    lang_pref = len(PREFERENCES) - pref
                    break
            else:
                lang_pref = -1

            media_type = f.get('mediaType')
            if media_type == 'hls':
                m3u8_formats = self._extract_m3u8_formats(
                    format_url, video_id, 'mp4', entry_protocol='m3u8_native',
                    m3u8_id=format_id, fatal=False)
                for m3u8_format in m3u8_formats:
                    m3u8_format['language_preference'] = lang_pref
                formats.extend(m3u8_formats)
                continue

            format = {
                'format_id': format_id,
                'language_preference': lang_pref,
                'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
                'width': int_or_none(f.get('width')),
                'height': int_or_none(f.get('height')),
                'tbr': int_or_none(f.get('bitrate')),
                'quality': qfunc(f.get('quality')),
            }

            if media_type == 'rtmp':
                format['url'] = f['streamer']
                format['play_path'] = 'mp4:' + f['url']
                format['ext'] = 'flv'
            else:
                format['url'] = f['url']

            formats.append(format)

        # For this extractor, quality only represents the relative quality
        # with respect to other formats with the same resolution
        self._sort_formats(formats, ('res', 'quality'))

        return {
            'id': player_info.get('VID') or video_id,
            'title': title,
            'description': player_info.get('VDE'),
            'upload_date': unified_strdate(upload_date_str),
            'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
            'formats': formats,
        }


class ArteTVEmbedIE(InfoExtractor):
    _VALID_URL = r'https?://(?:www\.)?arte\.tv/player/v\d+/index\.php\?.*?\bjson_url=.+'
    _TESTS = [{
        'url': 'https://www.arte.tv/player/v5/index.php?json_url=https%3A%2F%2Fapi.arte.tv%2Fapi%2Fplayer%2Fv2%2Fconfig%2Fde%2F100605-013-A&lang=de&autoplay=true&mute=0100605-013-A',
        'info_dict': {
            'id': '100605-013-A',
            'ext': 'mp4',
            'title': 'United we Stream November Lockdown Edition #13',
            'description': 'md5:be40b667f45189632b78c1425c7c2ce1',
            'upload_date': '20201116',
        },
    }, {
        'url': 'https://www.arte.tv/player/v3/index.php?json_url=https://api.arte.tv/api/player/v2/config/de/100605-013-A',
        'only_matching': True,
    }]

    @staticmethod
    def _extract_urls(webpage):
        return [url for _, url in re.findall(
            r'<(?:iframe|script)[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?arte\.tv/player/v\d+/index\.php\?.*?\bjson_url=.+?)\1',
            webpage)]

    def _real_extract(self, url):
        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
        json_url = qs['json_url'][0]
        video_id = ArteTVIE._match_id(json_url)
        return self.url_result(
            json_url, ie=ArteTVIE.ie_key(), video_id=video_id)


class ArteTVPlaylistIE(ArteTVBaseIE):
    _VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>%s)/videos/(?P<id>RC-\d{6})' % ArteTVBaseIE._ARTE_LANGUAGES
    _TESTS = [{
        'url': 'https://www.arte.tv/en/videos/RC-016954/earn-a-living/',
        'info_dict': {
            'id': 'RC-016954',
            'title': 'Earn a Living',
            'description': 'md5:d322c55011514b3a7241f7fb80d494c2',
        },
        'playlist_mincount': 6,
    }, {
        'url': 'https://www.arte.tv/pl/videos/RC-014123/arte-reportage/',
        'only_matching': True,
    }]

    def _real_extract(self, url):
        lang, playlist_id = re.match(self._VALID_URL, url).groups()
        collection = self._download_json(
            '%s/collectionData/%s/%s?source=videos'
            % (self._API_BASE, lang, playlist_id), playlist_id)
        entries = []
        for video in collection['videos']:
            if not isinstance(video, dict):
                continue
            video_url = url_or_none(video.get('url')) or url_or_none(video.get('jsonUrl'))
            if not video_url:
                continue
            video_id = video.get('programId')
            entries.append({
                '_type': 'url_transparent',
                'url': video_url,
                'id': video_id,
                'title': video.get('title'),
                'alt_title': video.get('subtitle'),
                'thumbnail': url_or_none(try_get(video, lambda x: x['mainImage']['url'], compat_str)),
                'duration': int_or_none(video.get('durationSeconds')),
                'view_count': int_or_none(video.get('views')),
                'ie_key': ArteTVIE.ie_key(),
            })
        title = collection.get('title')
        description = collection.get('shortDescription') or collection.get('teaserText')
        return self.playlist_result(entries, playlist_id, title, description)
Commit	Line	Data
dcdb292f	1	# coding: utf-8
3798eadc PH	2	from __future__ import unicode_literals
3798eadc PH	3
d5822b96	4	import re
d5822b96 PH	5
d5822b96 PH	6	from .common import InfoExtractor
8bdd16b4	7	from ..compat import (
	8	compat_str,
	9	compat_urlparse,
	10	)
d5822b96	11	from ..utils import (
c0892b2b	12	ExtractorError,
d24a2b20	13	int_or_none,
aff2f4f4	14	qualities,
8cc1840c	15	try_get,
c0892b2b	16	unified_strdate,
8bdd16b4	17	url_or_none,
d5822b96 PH	18	)
d5822b96 PH	19
d5822b96	20
6e6b9f60	21	class ArteTVBaseIE(InfoExtractor):
8bdd16b4	22	_ARTE_LANGUAGES = 'fr\|de\|en\|es\|it\|pl'
	23	_API_BASE = 'https://api.arte.tv/api/player/v1'
	24
	25
	26	class ArteTVIE(ArteTVBaseIE):
	27	_VALID_URL = r'''(?x)
	28	https?://
	29	(?:
	30	(?:www\.)?arte\.tv/(?P<lang>%(langs)s)/videos\|
	31	api\.arte\.tv/api/player/v\d+/config/(?P<lang_2>%(langs)s)
	32	)
	33	/(?P<id>\d{6}-\d{3}-[AF])
	34	''' % {'langs': ArteTVBaseIE._ARTE_LANGUAGES}
	35	_TESTS = [{
	36	'url': 'https://www.arte.tv/en/videos/088501-000-A/mexico-stealing-petrol-to-survive/',
	37	'info_dict': {
	38	'id': '088501-000-A',
	39	'ext': 'mp4',
	40	'title': 'Mexico: Stealing Petrol to Survive',
	41	'upload_date': '20190628',
	42	},
	43	}, {
	44	'url': 'https://www.arte.tv/pl/videos/100103-000-A/usa-dyskryminacja-na-porodowce/',
	45	'only_matching': True,
	46	}, {
	47	'url': 'https://api.arte.tv/api/player/v2/config/de/100605-013-A',
	48	'only_matching': True,
	49	}]
	50
	51	def _real_extract(self, url):
	52	mobj = re.match(self._VALID_URL, url)
	53	video_id = mobj.group('id')
	54	lang = mobj.group('lang') or mobj.group('lang_2')
	55
	56	info = self._download_json(
	57	'%s/config/%s/%s' % (self._API_BASE, lang, video_id), video_id)
c40f5cf4 JMF	58	player_info = info['videoJsonPlayer']
c40f5cf4 JMF	59
8cc1840c	60	vsr = try_get(player_info, lambda x: x['VSR'], dict)
6348671c	61	if not vsr:
8cc1840c S	62	error = None
	63	if try_get(player_info, lambda x: x['custom_msg']['type']) == 'error':
	64	error = try_get(
	65	player_info, lambda x: x['custom_msg']['msg'], compat_str)
	66	if not error:
	67	error = 'Video %s is not available' % player_info.get('VID') or video_id
	68	raise ExtractorError(error, expected=True)
c0892b2b	69
99b67fec PH	70	upload_date_str = player_info.get('shootingDate')
99b67fec PH	71	if not upload_date_str:
8bbd3d14	72	upload_date_str = (player_info.get('VRA') or player_info.get('VDA') or '').split(' ')[0]
99b67fec	73
8bdd16b4	74	title = (player_info.get('VTI') or player_info['VID']).strip()
74214d35 S	75	subtitle = player_info.get('VSU', '').strip()
	76	if subtitle:
	77	title += ' - %s' % subtitle
	78
ff0f4cfe	79	qfunc = qualities(['MQ', 'HQ', 'EQ', 'SQ'])
c40f5cf4	80
08d65046 S	81	LANGS = {
	82	'fr': 'F',
	83	'de': 'A',
	84	'en': 'E[ANG]',
	85	'es': 'E[ESP]',
ff0f4cfe RA	86	'it': 'E[ITA]',
ff0f4cfe RA	87	'pl': 'E[POL]',
08d65046 S	88	}
08d65046 S	89
9c072d38 S	90	langcode = LANGS.get(lang, lang)
9c072d38 S	91
aff2f4f4	92	formats = []
c0892b2b	93	for format_id, format_dict in vsr.items():
aff2f4f4	94	f = dict(format_dict)
8bdd16b4	95	format_url = url_or_none(f.get('url'))
	96	streamer = f.get('streamer')
	97	if not format_url and not streamer:
	98	continue
aff2f4f4	99	versionCode = f.get('versionCode')
9c072d38 S	100	l = re.escape(langcode)
	101
	102	# Language preference from most to least priority
ff0f4cfe RA	103	# Reference: section 6.8 of
ff0f4cfe RA	104	# https://www.arte.tv/sites/en/corporate/files/complete-technical-guidelines-arte-geie-v1-07-1.pdf
9c072d38 S	105	PREFERENCES = (
	106	# original version in requested language, without subtitles
	107	r'VO{0}$'.format(l),
	108	# original version in requested language, with partial subtitles in requested language
	109	r'VO{0}-ST{0}$'.format(l),
	110	# original version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
	111	r'VO{0}-STM{0}$'.format(l),
	112	# non-original (dubbed) version in requested language, without subtitles
	113	r'V{0}$'.format(l),
	114	# non-original (dubbed) version in requested language, with subtitles partial subtitles in requested language
	115	r'V{0}-ST{0}$'.format(l),
	116	# non-original (dubbed) version in requested language, with subtitles for the deaf and hard-of-hearing in requested language
	117	r'V{0}-STM{0}$'.format(l),
	118	# original version in requested language, with partial subtitles in different language
	119	r'VO{0}-ST(?!{0}).+?$'.format(l),
	120	# original version in requested language, with subtitles for the deaf and hard-of-hearing in different language
	121	r'VO{0}-STM(?!{0}).+?$'.format(l),
	122	# original version in different language, with partial subtitles in requested language
	123	r'VO(?:(?!{0}).+?)?-ST{0}$'.format(l),
	124	# original version in different language, with subtitles for the deaf and hard-of-hearing in requested language
	125	r'VO(?:(?!{0}).+?)?-STM{0}$'.format(l),
	126	# original version in different language, without subtitles
	127	r'VO(?:(?!{0}))?$'.format(l),
	128	# original version in different language, with partial subtitles in different language
	129	r'VO(?:(?!{0}).+?)?-ST(?!{0}).+?$'.format(l),
	130	# original version in different language, with subtitles for the deaf and hard-of-hearing in different language
	131	r'VO(?:(?!{0}).+?)?-STM(?!{0}).+?$'.format(l),
	132	)
	133
	134	for pref, p in enumerate(PREFERENCES):
	135	if re.match(p, versionCode):
	136	lang_pref = len(PREFERENCES) - pref
	137	break
	138	else:
	139	lang_pref = -1
	140
8bdd16b4	141	media_type = f.get('mediaType')
	142	if media_type == 'hls':
	143	m3u8_formats = self._extract_m3u8_formats(
	144	format_url, video_id, 'mp4', entry_protocol='m3u8_native',
	145	m3u8_id=format_id, fatal=False)
	146	for m3u8_format in m3u8_formats:
	147	m3u8_format['language_preference'] = lang_pref
	148	formats.extend(m3u8_formats)
	149	continue
	150
aff2f4f4 PH	151	format = {
aff2f4f4 PH	152	'format_id': format_id,
aff2f4f4 PH	153	'language_preference': lang_pref,
	154	'format_note': '%s, %s' % (f.get('versionCode'), f.get('versionLibelle')),
	155	'width': int_or_none(f.get('width')),
	156	'height': int_or_none(f.get('height')),
	157	'tbr': int_or_none(f.get('bitrate')),
1b7b1d6e	158	'quality': qfunc(f.get('quality')),
c40f5cf4	159	}
aff2f4f4	160
8bdd16b4	161	if media_type == 'rtmp':
aff2f4f4 PH	162	format['url'] = f['streamer']
	163	format['play_path'] = 'mp4:' + f['url']
	164	format['ext'] = 'flv'
c40f5cf4	165	else:
aff2f4f4 PH	166	format['url'] = f['url']
	167
	168	formats.append(format)
	169
dca3ff4a	170	# For this extractor, quality only represents the relative quality
	171	# with respect to other formats with the same resolution
	172	self._sort_formats(formats, ('res', 'quality'))
c40f5cf4	173
8bdd16b4	174	return {
	175	'id': player_info.get('VID') or video_id,
	176	'title': title,
	177	'description': player_info.get('VDE'),
	178	'upload_date': unified_strdate(upload_date_str),
	179	'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
	180	'formats': formats,
	181	}
c40f5cf4	182
24114fee	183
8bdd16b4	184	class ArteTVEmbedIE(InfoExtractor):
8bdd16b4	185	_VALID_URL = r'https?://(?:www\.)?arte\.tv/player/v\d+/index\.php\?.*?\bjson_url=.+'
9c54ae33	186	_TESTS = [{
8bdd16b4	187	'url': 'https://www.arte.tv/player/v5/index.php?json_url=https%3A%2F%2Fapi.arte.tv%2Fapi%2Fplayer%2Fv2%2Fconfig%2Fde%2F100605-013-A&lang=de&autoplay=true&mute=0100605-013-A',
9c54ae33	188	'info_dict': {
8bdd16b4	189	'id': '100605-013-A',
9c54ae33	190	'ext': 'mp4',
8bdd16b4	191	'title': 'United we Stream November Lockdown Edition #13',
	192	'description': 'md5:be40b667f45189632b78c1425c7c2ce1',
	193	'upload_date': '20201116',
69a0c470	194	},
8bdd16b4	195	}, {
	196	'url': 'https://www.arte.tv/player/v3/index.php?json_url=https://api.arte.tv/api/player/v2/config/de/100605-013-A',
	197	'only_matching': True,
9c54ae33	198	}]
56a8ab7d	199
8bdd16b4	200	@staticmethod
	201	def _extract_urls(webpage):
	202	return [url for _, url in re.findall(
	203	r'<(?:iframe\|script)[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?arte\.tv/player/v\d+/index\.php\?.*?\bjson_url=.+?)\1',
	204	webpage)]
6e6b9f60	205
893f8832	206	def _real_extract(self, url):
8bdd16b4	207	qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
	208	json_url = qs['json_url'][0]
	209	video_id = ArteTVIE._match_id(json_url)
	210	return self.url_result(
	211	json_url, ie=ArteTVIE.ie_key(), video_id=video_id)
4b492e35 S	212
4b492e35 S	213
6e6b9f60	214	class ArteTVPlaylistIE(ArteTVBaseIE):
8bdd16b4	215	_VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>%s)/videos/(?P<id>RC-\d{6})' % ArteTVBaseIE._ARTE_LANGUAGES
6e6b9f60	216	_TESTS = [{
ff0f4cfe	217	'url': 'https://www.arte.tv/en/videos/RC-016954/earn-a-living/',
6e6b9f60	218	'info_dict': {
ff0f4cfe RA	219	'id': 'RC-016954',
	220	'title': 'Earn a Living',
	221	'description': 'md5:d322c55011514b3a7241f7fb80d494c2',
6e6b9f60 S	222	},
6e6b9f60 S	223	'playlist_mincount': 6,
8bdd16b4	224	}, {
	225	'url': 'https://www.arte.tv/pl/videos/RC-014123/arte-reportage/',
	226	'only_matching': True,
6e6b9f60 S	227	}]
	228
	229	def _real_extract(self, url):
ff0f4cfe	230	lang, playlist_id = re.match(self._VALID_URL, url).groups()
6e6b9f60	231	collection = self._download_json(
8bdd16b4	232	'%s/collectionData/%s/%s?source=videos'
	233	% (self._API_BASE, lang, playlist_id), playlist_id)
	234	entries = []
	235	for video in collection['videos']:
	236	if not isinstance(video, dict):
	237	continue
	238	video_url = url_or_none(video.get('url')) or url_or_none(video.get('jsonUrl'))
	239	if not video_url:
	240	continue
	241	video_id = video.get('programId')
	242	entries.append({
	243	'_type': 'url_transparent',
	244	'url': video_url,
	245	'id': video_id,
	246	'title': video.get('title'),
	247	'alt_title': video.get('subtitle'),
	248	'thumbnail': url_or_none(try_get(video, lambda x: x['mainImage']['url'], compat_str)),
	249	'duration': int_or_none(video.get('durationSeconds')),
	250	'view_count': int_or_none(video.get('views')),
	251	'ie_key': ArteTVIE.ie_key(),
	252	})
6e6b9f60 S	253	title = collection.get('title')
6e6b9f60 S	254	description = collection.get('shortDescription') or collection.get('teaserText')
6e6b9f60	255	return self.playlist_result(entries, playlist_id, title, description)