ExtractorError,
GeoRestrictedError,
int_or_none,
+ join_nonempty,
parse_iso8601,
parse_qs,
strip_or_none,
class ArteTVIE(ArteTVBaseIE):
- _VALID_URL = r'''(?x)
+ _VALID_URL = rf'''(?x)
(?:https?://
(?:
- (?:www\.)?arte\.tv/(?P<lang>%(langs)s)/videos|
- api\.arte\.tv/api/player/v\d+/config/(?P<lang_2>%(langs)s)
+ (?:www\.)?arte\.tv/(?P<lang>{ArteTVBaseIE._ARTE_LANGUAGES})/videos|
+ api\.arte\.tv/api/player/v\d+/config/(?P<lang_2>{ArteTVBaseIE._ARTE_LANGUAGES})
)
|arte://program)
- /(?P<id>\d{6}-\d{3}-[AF]|LIVE)
- ''' % {'langs': ArteTVBaseIE._ARTE_LANGUAGES}
+ /(?P<id>\d{{6}}-\d{{3}}-[AF]|LIVE)
+ '''
_TESTS = [{
'url': 'https://www.arte.tv/en/videos/088501-000-A/mexico-stealing-petrol-to-survive/',
'only_matching': True,
}, {
- 'url': 'https://www.arte.tv/pl/videos/100103-000-A/usa-dyskryminacja-na-porodowce/',
- 'info_dict': {
- 'id': '100103-000-A',
- 'title': 'USA: Dyskryminacja na porodówce',
- 'description': 'md5:242017b7cce59ffae340a54baefcafb1',
- 'alt_title': 'ARTE Reportage',
- 'upload_date': '20201103',
- 'duration': 554,
- 'thumbnail': r're:https://api-cdn\.arte\.tv/.+940x530',
- 'timestamp': 1604417980,
- 'ext': 'mp4',
- },
- 'params': {'skip_download': 'm3u8'}
+ 'note': 'No alt_title',
+ 'url': 'https://www.arte.tv/fr/videos/110371-000-A/la-chaleur-supplice-des-arbres-de-rue/',
+ 'only_matching': True,
}, {
'url': 'https://api.arte.tv/api/player/v2/config/de/100605-013-A',
'only_matching': True,
}, {
'url': 'https://api.arte.tv/api/player/v2/config/de/LIVE',
'only_matching': True,
+ }, {
+ 'url': 'https://www.arte.tv/de/videos/110203-006-A/zaz/',
+ 'only_matching': True,
+ }, {
+ 'url': 'https://www.arte.tv/fr/videos/109067-000-A/la-loi-de-teheran/',
+ 'info_dict': {
+ 'id': '109067-000-A',
+ 'ext': 'mp4',
+ 'description': 'md5:d2ca367b8ecee028dddaa8bd1aebc739',
+ 'timestamp': 1713927600,
+ 'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/3rR6PLzfbigSkkeHtkCZNF/940x530',
+ 'duration': 7599,
+ 'title': 'La loi de Téhéran',
+ 'upload_date': '20240424',
+ 'subtitles': {
+ 'fr': 'mincount:1',
+ 'fr-acc': 'mincount:1',
+ 'fr-forced': 'mincount:1',
+ },
+ },
+ }, {
+ 'note': 'age-restricted',
+ 'url': 'https://www.arte.tv/de/videos/006785-000-A/the-element-of-crime/',
+ 'info_dict': {
+ 'id': '006785-000-A',
+ 'description': 'md5:c2f94fdfefc8a280e4dab68ab96ab0ba',
+ 'title': 'The Element of Crime',
+ 'timestamp': 1696111200,
+ 'duration': 5849,
+ 'thumbnail': 'https://api-cdn.arte.tv/img/v2/image/q82dTTfyuCXupPsGxXsd7B/940x530',
+ 'upload_date': '20230930',
+ 'ext': 'mp4',
+ },
+ 'skip': '404 Not Found',
}]
_GEO_BYPASS = True
# all obtained by exhaustive testing
_COUNTRIES_MAP = {
- 'DE_FR': {
+ 'DE_FR': (
'BL', 'DE', 'FR', 'GF', 'GP', 'MF', 'MQ', 'NC',
'PF', 'PM', 'RE', 'WF', 'YT',
- },
+ ),
# with both of the below 'BE' sometimes works, sometimes doesn't
- 'EUR_DE_FR': {
+ 'EUR_DE_FR': (
'AT', 'BL', 'CH', 'DE', 'FR', 'GF', 'GP', 'LI',
'MC', 'MF', 'MQ', 'NC', 'PF', 'PM', 'RE', 'WF',
'YT',
- },
- 'SAT': {
+ ),
+ 'SAT': (
'AD', 'AT', 'AX', 'BG', 'BL', 'CH', 'CY', 'CZ',
'DE', 'DK', 'EE', 'ES', 'FI', 'FR', 'GB', 'GF',
'GR', 'HR', 'HU', 'IE', 'IS', 'IT', 'KN', 'LI',
'LT', 'LU', 'LV', 'MC', 'MF', 'MQ', 'MT', 'NC',
'NL', 'NO', 'PF', 'PL', 'PM', 'PT', 'RE', 'RO',
'SE', 'SI', 'SK', 'SM', 'VA', 'WF', 'YT',
- },
+ ),
}
+ @staticmethod
+ def _fix_accessible_subs_locale(subs):
+ updated_subs = {}
+ for lang, sub_formats in subs.items():
+ for fmt in sub_formats:
+ url = fmt.get('url') or ''
+ suffix = ('acc' if url.endswith('-MAL.m3u8')
+ else 'forced' if '_VO' not in url
+ else None)
+ updated_subs.setdefault(join_nonempty(lang, suffix), []).append(fmt)
+ return updated_subs
+
def _real_extract(self, url):
mobj = self._match_valid_url(url)
video_id = mobj.group('id')
lang = mobj.group('lang') or mobj.group('lang_2')
- langauge_code = self._LANG_MAP.get(lang)
+ language_code = self._LANG_MAP.get(lang)
- config = self._download_json(f'{self._API_BASE}/config/{lang}/{video_id}', video_id)
+ config = self._download_json(f'{self._API_BASE}/config/{lang}/{video_id}', video_id, headers={
+ 'x-validated-age': '18',
+ })
geoblocking = traverse_obj(config, ('data', 'attributes', 'restriction', 'geoblocking')) or {}
if geoblocking.get('restrictedArea'):
'Video is not available in this language edition of Arte or broadcast rights expired', expected=True)
formats, subtitles = [], {}
+ secondary_formats = []
for stream in config['data']['attributes']['streams']:
# official player contains code like `e.get("versions")[0].eStat.ml5`
stream_version = stream['versions'][0]
m = self._VERSION_CODE_RE.match(stream_version_code)
if m:
lang_pref = int(''.join('01'[x] for x in (
- m.group('vlang') == langauge_code, # we prefer voice in the requested language
+ m.group('vlang') == language_code, # we prefer voice in the requested language
not m.group('audio_desc'), # and not the audio description version
bool(m.group('original_voice')), # but if voice is not in the requested language, at least choose the original voice
- m.group('sub_lang') == langauge_code, # if subtitles are present, we prefer them in the requested language
+ m.group('sub_lang') == language_code, # if subtitles are present, we prefer them in the requested language
not m.group('has_sub'), # but we prefer no subtitles otherwise
not m.group('sdh_sub'), # and we prefer not the hard-of-hearing subtitles if there are subtitles
)))
- if stream['protocol'].startswith('HLS'):
+ short_label = traverse_obj(stream_version, 'shortLabel', expected_type=str, default='?')
+ if 'HLS' in stream['protocol']:
fmts, subs = self._extract_m3u8_formats_and_subtitles(
stream['url'], video_id=video_id, ext='mp4', m3u8_id=stream_version_code, fatal=False)
for fmt in fmts:
fmt.update({
- 'format_note': f'{stream_version.get("label", "unknown")} [{stream_version.get("shortLabel", "?")}]',
+ 'format_note': f'{stream_version.get("label", "unknown")} [{short_label}]',
'language_preference': lang_pref,
})
- formats.extend(fmts)
+ if any(map(short_label.startswith, ('cc', 'OGsub'))):
+ secondary_formats.extend(fmts)
+ else:
+ formats.extend(fmts)
+ subs = self._fix_accessible_subs_locale(subs)
self._merge_subtitles(subs, target=subtitles)
elif stream['protocol'] in ('HTTPS', 'RTMP'):
formats.append({
'format_id': f'{stream["protocol"]}-{stream_version_code}',
'url': stream['url'],
- 'format_note': f'{stream_version.get("label", "unknown")} [{stream_version.get("shortLabel", "?")}]',
+ 'format_note': f'{stream_version.get("label", "unknown")} [{short_label}]',
'language_preference': lang_pref,
# 'ext': 'mp4', # XXX: may or may not be necessary, at least for HTTPS
})
else:
self.report_warning(f'Skipping stream with unknown protocol {stream["protocol"]}')
- # TODO: chapters from stream['segments']?
- # The JS also looks for chapters in config['data']['attributes']['chapters'],
- # but I am yet to find a video having those
-
- self._sort_formats(formats)
+ formats.extend(secondary_formats)
+ self._remove_duplicate_formats(formats)
metadata = config['data']['attributes']['metadata']
return {
'id': metadata['providerId'],
'webpage_url': traverse_obj(metadata, ('link', 'url')),
- 'title': metadata.get('subtitle'),
- 'alt_title': metadata.get('title'),
+ 'title': traverse_obj(metadata, 'subtitle', 'title'),
+ 'alt_title': metadata.get('subtitle') and metadata.get('title'),
'description': metadata.get('description'),
'duration': traverse_obj(metadata, ('duration', 'seconds')),
'language': metadata.get('language'),
{'url': image['url'], 'id': image.get('caption')}
for image in metadata.get('images') or [] if url_or_none(image.get('url'))
],
+ # TODO: chapters may also be in stream['segments']?
+ 'chapters': traverse_obj(config, ('data', 'attributes', 'chapters', 'elements', ..., {
+ 'start_time': 'startTime',
+ 'title': 'title',
+ })) or None,
}
class ArteTVEmbedIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?arte\.tv/player/v\d+/index\.php\?.*?\bjson_url=.+'
+ _EMBED_REGEX = [r'<(?:iframe|script)[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?arte\.tv/player/v\d+/index\.php\?.*?\bjson_url=.+?)\1']
_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': {
'description': 'md5:be40b667f45189632b78c1425c7c2ce1',
'upload_date': '20201116',
},
- 'skip': 'No video available'
+ 'skip': 'No video available',
}, {
'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 = parse_qs(url)
json_url = qs['json_url'][0]
class ArteTVPlaylistIE(ArteTVBaseIE):
- _VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>%s)/videos/(?P<id>RC-\d{6})' % ArteTVBaseIE._ARTE_LANGUAGES
+ _VALID_URL = rf'https?://(?:www\.)?arte\.tv/(?P<lang>{ArteTVBaseIE._ARTE_LANGUAGES})/videos/(?P<id>RC-\d{{6}})'
_TESTS = [{
'url': 'https://www.arte.tv/en/videos/RC-016954/earn-a-living/',
'only_matching': True,
class ArteTVCategoryIE(ArteTVBaseIE):
- _VALID_URL = r'https?://(?:www\.)?arte\.tv/(?P<lang>%s)/videos/(?P<id>[\w-]+(?:/[\w-]+)*)/?\s*$' % ArteTVBaseIE._ARTE_LANGUAGES
+ _VALID_URL = rf'https?://(?:www\.)?arte\.tv/(?P<lang>{ArteTVBaseIE._ARTE_LANGUAGES})/videos/(?P<id>[\w-]+(?:/[\w-]+)*)/?\s*$'
_TESTS = [{
'url': 'https://www.arte.tv/en/videos/politics-and-society/',
'info_dict': {
@classmethod
def suitable(cls, url):
return (
- not any(ie.suitable(url) for ie in (ArteTVIE, ArteTVPlaylistIE, ))
+ not any(ie.suitable(url) for ie in (ArteTVIE, ArteTVPlaylistIE))
and super().suitable(url))
def _real_extract(self, url):
items = []
for video in re.finditer(
- r'<a\b[^>]*?href\s*=\s*(?P<q>"|\'|\b)(?P<url>https?://www\.arte\.tv/%s/videos/[\w/-]+)(?P=q)' % lang,
+ rf'<a\b[^>]*?href\s*=\s*(?P<q>"|\'|\b)(?P<url>https?://www\.arte\.tv/{lang}/videos/[\w/-]+)(?P=q)',
webpage):
video = video.group('url')
if video == url:
continue
- if any(ie.suitable(video) for ie in (ArteTVIE, ArteTVPlaylistIE, )):
+ if any(ie.suitable(video) for ie in (ArteTVIE, ArteTVPlaylistIE)):
items.append(video)
- title = (self._og_search_title(webpage, default=None)
- or self._html_search_regex(r'<title\b[^>]*>([^<]+)</title>', default=None))
- title = strip_or_none(title.rsplit('|', 1)[0]) or self._generic_title(url)
+ title = strip_or_none(self._generic_title('', webpage, default='').rsplit('|', 1)[0]) or None
return self.playlist_from_matches(items, playlist_id=playlist_id, playlist_title=title,
description=self._og_search_description(webpage, default=None))
projects / yt-dlp.git / blobdiff