diff --git a/yt_dlp/extractor/cbsnews.py b/yt_dlp/extractor/cbsnews.py index 345debcf0..1285ed65e 100644 --- a/yt_dlp/extractor/cbsnews.py +++ b/yt_dlp/extractor/cbsnews.py @@ -26,7 +26,7 @@ class CBSNewsEmbedIE(CBSIE): def _real_extract(self, url): item = self._parse_json(zlib.decompress(compat_b64decode( compat_urllib_parse_unquote(self._match_id(url))), - -zlib.MAX_WBITS), None)['video']['items'][0] + -zlib.MAX_WBITS).decode('utf-8'), None)['video']['items'][0] return self._extract_video_info(item['mpxRefId'], 'cbsnews') diff --git a/yt_dlp/extractor/lbry.py b/yt_dlp/extractor/lbry.py index 865cda761..cdfbefcd4 100644 --- a/yt_dlp/extractor/lbry.py +++ b/yt_dlp/extractor/lbry.py @@ -122,6 +122,26 @@ class LBRYIE(LBRYBaseIE): 'channel_url': 'https://lbry.tv/@LBRYFoundation:0ed629d2b9c601300cacf7eabe9da0be79010212', 'vcodec': 'none', } + }, { + # HLS + 'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e', + 'md5': 'fc82f45ea54915b1495dd7cb5cc1289f', + 'info_dict': { + 'id': 'e51671357333fe22ae88aad320bde2f6f96b1410', + 'ext': 'mp4', + 'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁', + 'description': 'md5:9c539c6a03fb843956de61a4d5288d5e', + 'timestamp': 1618254123, + 'upload_date': '20210412', + 'release_timestamp': 1618254002, + 'release_date': '20210412', + 'tags': list, + 'duration': 554, + 'channel': 'Gardening In Canada', + 'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc', + 'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc', + 'formats': 'mincount:3', + } }, { 'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e', 'only_matching': True, @@ -168,10 +188,18 @@ def _real_extract(self, url): streaming_url = self._call_api_proxy( 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url'] info = self._parse_stream(result, url) + urlh = self._request_webpage( + streaming_url, display_id, note='Downloading streaming redirect url info') + if determine_ext(urlh.geturl()) == 'm3u8': + info['formats'] = self._extract_m3u8_formats( + urlh.geturl(), display_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls') + self._sort_formats(info['formats']) + else: + info['url'] = streaming_url info.update({ 'id': claim_id, 'title': title, - 'url': streaming_url, }) return info diff --git a/yt_dlp/extractor/pluralsight.py b/yt_dlp/extractor/pluralsight.py index abd08bc28..2d63855df 100644 --- a/yt_dlp/extractor/pluralsight.py +++ b/yt_dlp/extractor/pluralsight.py @@ -393,7 +393,7 @@ def guess_allowed_qualities(): # To somewhat reduce the probability of these consequences # we will sleep random amount of time before each call to ViewClip. self._sleep( - random.randint(2, 5), display_id, + random.randint(5, 10), display_id, '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling') if not viewclip: diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py index 4c9da101f..940f3def2 100644 --- a/yt_dlp/extractor/youtube.py +++ b/yt_dlp/extractor/youtube.py @@ -53,6 +53,10 @@ ) +def parse_qs(url): + return compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + + class YoutubeBaseInfoExtractor(InfoExtractor): """Provide base functions for Youtube extractors""" _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' @@ -438,14 +442,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'(?:(?:www|dev)\.)?invidio\.us', # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md r'(?:www\.)?invidious\.pussthecat\.org', - r'(?:www\.)?invidious\.048596\.xyz', r'(?:www\.)?invidious\.zee\.li', - r'(?:www\.)?vid\.puffyan\.us', r'(?:(?:www|au)\.)?ytprivate\.com', r'(?:www\.)?invidious\.namazso\.eu', r'(?:www\.)?invidious\.ethibox\.fr', - r'(?:www\.)?inv\.skyn3t\.in', - r'(?:www\.)?invidious\.himiko\.cloud', r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion', r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion', r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion', @@ -454,25 +454,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'(?:(?:www|no)\.)?invidiou\.sh', r'(?:(?:www|fi)\.)?invidious\.snopyta\.org', r'(?:www\.)?invidious\.kabi\.tk', - r'(?:www\.)?invidious\.13ad\.de', r'(?:www\.)?invidious\.mastodon\.host', r'(?:www\.)?invidious\.zapashcanon\.fr', r'(?:www\.)?invidious\.kavin\.rocks', + r'(?:www\.)?invidious\.tinfoil-hat\.net', + r'(?:www\.)?invidious\.himiko\.cloud', + r'(?:www\.)?invidious\.reallyancient\.tech', r'(?:www\.)?invidious\.tube', r'(?:www\.)?invidiou\.site', r'(?:www\.)?invidious\.site', r'(?:www\.)?invidious\.xyz', r'(?:www\.)?invidious\.nixnet\.xyz', + r'(?:www\.)?invidious\.048596\.xyz', r'(?:www\.)?invidious\.drycat\.fr', + r'(?:www\.)?inv\.skyn3t\.in', r'(?:www\.)?tube\.poal\.co', r'(?:www\.)?tube\.connect\.cafe', r'(?:www\.)?vid\.wxzm\.sx', r'(?:www\.)?vid\.mint\.lgbt', + r'(?:www\.)?vid\.puffyan\.us', r'(?:www\.)?yewtu\.be', r'(?:www\.)?yt\.elukerio\.org', r'(?:www\.)?yt\.lelux\.fi', r'(?:www\.)?invidious\.ggc-project\.de', r'(?:www\.)?yt\.maisputain\.ovh', + r'(?:www\.)?ytprivate\.com', + r'(?:www\.)?invidious\.13ad\.de', r'(?:www\.)?invidious\.toot\.koeln', r'(?:www\.)?invidious\.fdn\.fr', r'(?:www\.)?watch\.nettohikari\.com', @@ -515,16 +522,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= ) )? # all until now is optional -> you can pass the naked ID - (?P[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID - (?!.*?\blist= - (?: - %(playlist_id)s| # combined list/video URLs are handled by the playlist IE - WL # WL are handled by the watch later IE - ) - ) + (?P[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID (?(1).+)? # if we found the ID, everything can follow $""" % { - 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE, 'invidious': '|'.join(_INVIDIOUS_SITES), } _PLAYER_INFO_RE = ( @@ -1009,6 +1009,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, 'skip': 'This video does not exist.', }, + { + # Video with incomplete 'yt:stretch=16:' + 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI', + 'only_matching': True, + }, { # Video licensed under Creative Commons 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA', @@ -1304,6 +1309,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, ] + @classmethod + def suitable(cls, url): + qs = parse_qs(url) + if qs.get('list', [None])[0]: + return False + return super(YoutubeIE, cls).suitable(url) + def __init__(self, *args, **kwargs): super(YoutubeIE, self).__init__(*args, **kwargs) self._code_cache = {} @@ -2079,15 +2091,16 @@ def feed_entry(name): for m in re.finditer(self._meta_regex('og:video:tag'), webpage)] for keyword in keywords: if keyword.startswith('yt:stretch='): - stretch_ratio = map( - lambda x: int_or_none(x, default=0), - keyword.split('=')[1].split(':')) - w, h = (list(stretch_ratio) + [0])[:2] - if w > 0 and h > 0: - ratio = w / h - for f in formats: - if f.get('vcodec') != 'none': - f['stretched_ratio'] = ratio + mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword) + if mobj: + # NB: float is intentional for forcing float division + w, h = (float(v) for v in mobj.groups()) + if w > 0 and h > 0: + ratio = w / h + for f in formats: + if f.get('vcodec') != 'none': + f['stretched_ratio'] = ratio + break thumbnails = [] for container in (video_details, microformat): @@ -2484,6 +2497,15 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg', 'uploader': 'Игорь Клейнер', }, + }, { + # playlists, series + 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3', + 'playlist_mincount': 5, + 'info_dict': { + 'id': 'UCYO_jab_esuFRV4b17AJtAw', + 'title': '3Blue1Brown - Playlists', + 'description': 'md5:e1384e8a133307dd10edee76e875d62f', + }, }, { # playlists, singlepage 'url': 'https://www.youtube.com/user/ThirstForScience/playlists', @@ -2790,6 +2812,9 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'title': '#cctv9', }, 'playlist_mincount': 350, + }, { + 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU', + 'only_matching': True, }] @classmethod @@ -2813,14 +2838,16 @@ def _extract_channel_id(self, webpage): @staticmethod def _extract_basic_item_renderer(item): # Modified from _extract_grid_item_renderer - known_renderers = ( - 'playlistRenderer', 'videoRenderer', 'channelRenderer', - 'gridPlaylistRenderer', 'gridVideoRenderer', 'gridChannelRenderer' + known_basic_renderers = ( + 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer' ) for key, renderer in item.items(): - if key not in known_renderers: + if not isinstance(renderer, dict): continue - return renderer + elif key in known_basic_renderers: + return renderer + elif key.startswith('grid') and key.endswith('Renderer'): + return renderer def _grid_entries(self, grid_renderer): for item in grid_renderer['items']: @@ -2830,7 +2857,8 @@ def _grid_entries(self, grid_renderer): if not isinstance(renderer, dict): continue title = try_get( - renderer, lambda x: x['title']['runs'][0]['text'], compat_str) + renderer, (lambda x: x['title']['runs'][0]['text'], + lambda x: x['title']['simpleText']), compat_str) # playlist playlist_id = renderer.get('playlistId') if playlist_id: @@ -2838,10 +2866,12 @@ def _grid_entries(self, grid_renderer): 'https://www.youtube.com/playlist?list=%s' % playlist_id, ie=YoutubeTabIE.ie_key(), video_id=playlist_id, video_title=title) + continue # video video_id = renderer.get('videoId') if video_id: yield self._extract_video(renderer) + continue # channel channel_id = renderer.get('channelId') if channel_id: @@ -2850,6 +2880,17 @@ def _grid_entries(self, grid_renderer): yield self.url_result( 'https://www.youtube.com/channel/%s' % channel_id, ie=YoutubeTabIE.ie_key(), video_title=title) + continue + # generic endpoint URL support + ep_url = urljoin('https://www.youtube.com/', try_get( + renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'], + compat_str)) + if ep_url: + for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE): + if ie.suitable(ep_url): + yield self.url_result( + ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title) + break def _shelf_entries_from_content(self, shelf_renderer): content = shelf_renderer.get('content') @@ -3444,7 +3485,7 @@ def _real_extract(self, url): url = '%s/videos%s' % (mobj.get('pre'), mobj.get('post') or '') # Handle both video/playlist URLs - qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + qs = parse_qs(url) video_id = qs.get('v', [None])[0] playlist_id = qs.get('list', [None])[0] @@ -3550,12 +3591,16 @@ class YoutubePlaylistIE(InfoExtractor): @classmethod def suitable(cls, url): - return False if YoutubeTabIE.suitable(url) else super( - YoutubePlaylistIE, cls).suitable(url) + if YoutubeTabIE.suitable(url): + return False + qs = parse_qs(url) + if qs.get('v', [None])[0]: + return False + return super(YoutubePlaylistIE, cls).suitable(url) def _real_extract(self, url): playlist_id = self._match_id(url) - qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + qs = parse_qs(url) if not qs: qs = {'list': playlist_id} return self.url_result( diff --git a/yt_dlp/utils.py b/yt_dlp/utils.py index 3ba2a1ec8..3e566285f 100644 --- a/yt_dlp/utils.py +++ b/yt_dlp/utils.py @@ -40,6 +40,7 @@ from .compat import ( compat_HTMLParseError, compat_HTMLParser, + compat_HTTPError, compat_basestring, compat_chr, compat_cookiejar, @@ -2925,12 +2926,60 @@ def http_response(self, request, response): class YoutubeDLRedirectHandler(compat_urllib_request.HTTPRedirectHandler): - if sys.version_info[0] < 3: - def redirect_request(self, req, fp, code, msg, headers, newurl): - # On python 2 urlh.geturl() may sometimes return redirect URL - # as byte string instead of unicode. This workaround allows - # to force it always return unicode. - return compat_urllib_request.HTTPRedirectHandler.redirect_request(self, req, fp, code, msg, headers, compat_str(newurl)) + """YoutubeDL redirect handler + + The code is based on HTTPRedirectHandler implementation from CPython [1]. + + This redirect handler solves two issues: + - ensures redirect URL is always unicode under python 2 + - introduces support for experimental HTTP response status code + 308 Permanent Redirect [2] used by some sites [3] + + 1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py + 2. https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/308 + 3. https://github.com/ytdl-org/youtube-dl/issues/28768 + """ + + http_error_301 = http_error_303 = http_error_307 = http_error_308 = compat_urllib_request.HTTPRedirectHandler.http_error_302 + + def redirect_request(self, req, fp, code, msg, headers, newurl): + """Return a Request or None in response to a redirect. + + This is called by the http_error_30x methods when a + redirection response is received. If a redirection should + take place, return a new Request to allow http_error_30x to + perform the redirect. Otherwise, raise HTTPError if no-one + else should try to handle this url. Return None if you can't + but another Handler might. + """ + m = req.get_method() + if (not (code in (301, 302, 303, 307, 308) and m in ("GET", "HEAD") + or code in (301, 302, 303) and m == "POST")): + raise compat_HTTPError(req.full_url, code, msg, headers, fp) + # Strictly (according to RFC 2616), 301 or 302 in response to + # a POST MUST NOT cause a redirection without confirmation + # from the user (of urllib.request, in this case). In practice, + # essentially all clients do redirect in this case, so we do + # the same. + + # On python 2 urlh.geturl() may sometimes return redirect URL + # as byte string instead of unicode. This workaround allows + # to force it always return unicode. + if sys.version_info[0] < 3: + newurl = compat_str(newurl) + + # Be conciliant with URIs containing a space. This is mainly + # redundant with the more complete encoding done in http_error_302(), + # but it is kept for compatibility with other callers. + newurl = newurl.replace(' ', '%20') + + CONTENT_HEADERS = ("content-length", "content-type") + # NB: don't use dict comprehension for python 2.6 compatibility + newheaders = dict((k, v) for k, v in req.headers.items() + if k.lower() not in CONTENT_HEADERS) + return compat_urllib_request.Request( + newurl, headers=newheaders, origin_req_host=req.origin_req_host, + unverifiable=True) def extract_timezone(date_str):