[youtube] Add Invidious list for playlists/channels (#1567)

Authored by: rhendric

yt_dlp/extractor/youtube.py

@@ -259,6 +259,70 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
     # If True it will raise an error if no login info is provided
     _LOGIN_REQUIRED = False
 
+    _INVIDIOUS_SITES = (
+        # invidious-redirect websites
+        r'(?:www\.)?redirect\.invidious\.io',
+        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\.zee\.li',
+        r'(?:www\.)?invidious\.ethibox\.fr',
+        r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
+        # youtube-dl invidious instances list
+        r'(?:(?:www|no)\.)?invidiou\.sh',
+        r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
+        r'(?:www\.)?invidious\.kabi\.tk',
+        r'(?:www\.)?invidious\.mastodon\.host',
+        r'(?:www\.)?invidious\.zapashcanon\.fr',
+        r'(?:www\.)?(?:invidious(?:-us)?|piped)\.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',
+        r'(?:www\.)?invidious\.namazso\.eu',
+        r'(?:www\.)?invidious\.silkky\.cloud',
+        r'(?:www\.)?invidious\.exonip\.de',
+        r'(?:www\.)?invidious\.riverside\.rocks',
+        r'(?:www\.)?invidious\.blamefran\.net',
+        r'(?:www\.)?invidious\.moomoo\.de',
+        r'(?:www\.)?ytb\.trom\.tf',
+        r'(?:www\.)?yt\.cyberhost\.uk',
+        r'(?:www\.)?kgg2m7yk5aybusll\.onion',
+        r'(?:www\.)?qklhadlycap4cnod\.onion',
+        r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
+        r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
+        r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
+        r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
+        r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
+        r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
+        r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
+        r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
+        r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
+        r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
+    )
+
     def _login(self):
         """
         Attempt to log in to YouTube.
@@ -697,69 +761,6 @@ def _extract_video(self, renderer):
 
 class YoutubeIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube'
-    _INVIDIOUS_SITES = (
-        # invidious-redirect websites
-        r'(?:www\.)?redirect\.invidious\.io',
-        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\.zee\.li',
-        r'(?:www\.)?invidious\.ethibox\.fr',
-        r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion',
-        # youtube-dl invidious instances list
-        r'(?:(?:www|no)\.)?invidiou\.sh',
-        r'(?:(?:www|fi)\.)?invidious\.snopyta\.org',
-        r'(?:www\.)?invidious\.kabi\.tk',
-        r'(?:www\.)?invidious\.mastodon\.host',
-        r'(?:www\.)?invidious\.zapashcanon\.fr',
-        r'(?:www\.)?(?:invidious(?:-us)?|piped)\.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',
-        r'(?:www\.)?invidious\.namazso\.eu',
-        r'(?:www\.)?invidious\.silkky\.cloud',
-        r'(?:www\.)?invidious\.exonip\.de',
-        r'(?:www\.)?invidious\.riverside\.rocks',
-        r'(?:www\.)?invidious\.blamefran\.net',
-        r'(?:www\.)?invidious\.moomoo\.de',
-        r'(?:www\.)?ytb\.trom\.tf',
-        r'(?:www\.)?yt\.cyberhost\.uk',
-        r'(?:www\.)?kgg2m7yk5aybusll\.onion',
-        r'(?:www\.)?qklhadlycap4cnod\.onion',
-        r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion',
-        r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion',
-        r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion',
-        r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion',
-        r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p',
-        r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion',
-        r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion',
-        r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion',
-        r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion',
-        r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion',
-    )
     _VALID_URL = r"""(?x)^
                      (
                          (?:https?://|//)                                    # http(s):// or protocol-independent URL
@@ -793,7 +794,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                      (?P<id>[0-9A-Za-z_-]{11})                                # here is it! the YouTube video ID
                      (?(1).+)?                                                # if we found the ID, everything can follow
                      (?:\#|$)""" % {
-        'invidious': '|'.join(_INVIDIOUS_SITES),
+        'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
     }
     _PLAYER_INFO_RE = (
         r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
@@ -3064,7 +3065,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                         (?:\w+\.)?
                         (?:
                             youtube(?:kids)?\.com|
-                            invidio\.us
+                            %(invidious)s
                         )/
                         (?:
                             (?P<channel_type>channel|c|user|browse)/|
@@ -3072,10 +3073,13 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                                 feed/|hashtag/|
                                 (?:playlist|watch)\?.*?\blist=
                             )|
-                            (?!(?:%s)\b)  # Direct URLs
+                            (?!(?:%(reserved_names)s)\b)  # Direct URLs
                         )
                         (?P<id>[^/?\#&]+)
-                    ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES
+                    ''' % {
+        'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES,
+        'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
+    }
     IE_NAME = 'youtube:tab'
 
     _TESTS = [{
@@ -4293,12 +4297,15 @@ class YoutubePlaylistIE(InfoExtractor):
                         (?:
                             (?:
                                 youtube(?:kids)?\.com|
-                                invidio\.us
+                                %(invidious)s
                             )
                             /.*?\?.*?\blist=
                         )?
                         (?P<id>%(playlist_id)s)
-                     )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
+                     )''' % {
+        'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE,
+        'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES),
+    }
     IE_NAME = 'youtube:playlist'
     _TESTS = [{
         'note': 'issue #673',