[ie/patreon] Support new URL format (#16926)

Closes #16923
Authored by: 0xvd

test/test_utils.py

@@ -327,6 +327,12 @@ class TestUtil(unittest.TestCase):
             with self.assertRaises(_UnsafeExtensionError):
                 prepend_extension('abc.unexpected_ext', ext, 'ext')
 
+        # Test allow-unsafe-ext compat option
+        _UnsafeExtensionError._enabled = False
+        self.assertEqual(prepend_extension('abc.ext', 'un/safe'), 'abc.un/safe.ext')
+        # Re-enable sanitization for other tests
+        _UnsafeExtensionError._enabled = True
+
     def test_replace_extension(self):
         self.assertEqual(replace_extension('abc.ext', 'temp'), 'abc.temp')
         self.assertEqual(replace_extension('abc.ext', 'temp', 'ext'), 'abc.temp')
@@ -345,6 +351,12 @@ class TestUtil(unittest.TestCase):
             with self.assertRaises(_UnsafeExtensionError):
                 replace_extension('abc.unexpected_ext', ext, 'ext')
 
+        # Test allow-unsafe-ext compat option
+        _UnsafeExtensionError._enabled = False
+        self.assertEqual(replace_extension('abc.ext', 'bin'), 'abc.bin')
+        # Re-enable sanitization for other tests
+        _UnsafeExtensionError._enabled = True
+
     def test_subtitles_filename(self):
         self.assertEqual(subtitles_filename('abc.ext', 'en', 'vtt'), 'abc.en.vtt')
         self.assertEqual(subtitles_filename('abc.ext', 'en', 'vtt', 'ext'), 'abc.en.vtt')
@@ -2160,6 +2172,10 @@ Line 1
         headers6 = HTTPHeaderDict(a=1, b=2)
         self.assertEqual(pickle.loads(pickle.dumps(headers6)), headers6)
 
+        headers7 = HTTPHeaderDict()
+        headers7 |= {'X-dlp': 'data'}
+        self.assertEqual(headers7.sensitive(), {'X-dlp': 'data'})
+
     def test_extract_basic_auth(self):
         assert extract_basic_auth('http://:foo.bar') == ('http://:foo.bar', None)
         assert extract_basic_auth('http://foo.bar') == ('http://foo.bar', None)

yt_dlp/__init__.py

@@ -619,7 +619,7 @@ def validate_options(opts):
         warnings.append(
             'Using allow-unsafe-ext opens you up to potential attacks. '
             'Use with great care!')
-        _UnsafeExtensionError.sanitize_extension = lambda x, prepend=False: x
+        _UnsafeExtensionError._enabled = False
 
     return warnings, deprecation_warnings
 

yt_dlp/downloader/external.py

@@ -209,7 +209,7 @@ class CurlFD(ExternalFD):
             return False
 
         cls.exe = path
-        cls._curl_version = version_tuple(parts[1])
+        cls._curl_version = version_tuple(parts[1], lenient=True)
         return path
 
     def _make_cmd(self, tmpfilename, info_dict):

yt_dlp/extractor/bandcamp.py

@@ -420,10 +420,10 @@ class BandcampWeeklyIE(BandcampIE):  # XXX: Do not subclass from concrete IE
         'info_dict': {
             'id': '224',
             'ext': 'mp3',
-            'title': 'Bandcamp Weekly, 2017-04-04',
+            'title': 'Magic Moments, 2017-04-04',
             'description': 'md5:5d48150916e8e02d030623a48512c874',
             'thumbnail': 'https://f4.bcbits.com/img/9982549_0.jpg',
-            'series': 'Bandcamp Weekly',
+            'series': 'Magic Moments',
             'episode_id': '224',
             'release_timestamp': 1491264000,
             'release_date': '20170404',
@@ -440,10 +440,10 @@ class BandcampWeeklyIE(BandcampIE):  # XXX: Do not subclass from concrete IE
     def _real_extract(self, url):
         show_id = self._match_id(url)
         show_data = self._download_json(
-            'https://bandcamp.com/api/bcradio_api/1/get_show',
+            'https://bandcamp.com/api/player/2/player_data_web',
             show_id, 'Downloading radio show JSON',
-            data=json.dumps({'id': show_id}).encode(),
+            data=json.dumps({'item_id': int(show_id), 'item_type': 'radio'}).encode(),
-            headers={'Content-Type': 'application/json'})
+            headers={'Content-Type': 'application/json'})['tracklist']
         audio_data = show_data['compiledTrack']
 
         stream_url = audio_data['streamUrl']

yt_dlp/extractor/patreon.py

@@ -65,8 +65,9 @@ class PatreonBaseIE(InfoExtractor):
 
 class PatreonIE(PatreonBaseIE):
     IE_NAME = 'patreon'
-    _VALID_URL = r'https?://(?:www\.)?patreon\.com/(?:creation\?hid=|posts/(?:[\w-]+-)?)(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?patreon\.com/(?:creation\?hid=|(?:[^/?#]+/)?posts/(?:[\w-]+-)?)(?P<id>\d+)'
     _TESTS = [{
+        # FIXME: Fails due to no description extracted
         'url': 'http://www.patreon.com/creation?hid=743933',
         'md5': 'e25505eec1053a6e6813b8ed369875cc',
         'info_dict': {
@@ -107,17 +108,17 @@ class PatreonIE(PatreonBaseIE):
             'id': 'SU4fj_aEMVw',
             'ext': 'mp4',
             'title': 'I\'m on Patreon!',
-            'uploader': 'TraciJHines',
+            'uploader': 'Traci Oden',
             'thumbnail': 're:^https?://.*$',
             'upload_date': '20150211',
             'description': 'md5:8af6425f50bd46fbf29f3db0fc3a8364',
-            'uploader_id': '@TraciHinesMusic',
+            'uploader_id': '@TraciOden',
             'categories': ['Entertainment'],
             'duration': 282,
             'view_count': int,
             'tags': 'count:39',
             'age_limit': 0,
-            'channel': 'TraciJHines',
+            'channel': 'Traci Oden',
             'channel_url': 'https://www.youtube.com/channel/UCGLim4T2loE5rwCMdpCIPVg',
             'live_status': 'not_live',
             'like_count': int,
@@ -125,7 +126,7 @@ class PatreonIE(PatreonBaseIE):
             'availability': 'public',
             'channel_follower_count': int,
             'playable_in_embed': True,
-            'uploader_url': 'https://www.youtube.com/@TraciHinesMusic',
+            'uploader_url': 'https://www.youtube.com/@TraciOden',
             'comment_count': int,
             'channel_is_verified': True,
             'chapters': 'count:4',
@@ -157,6 +158,7 @@ class PatreonIE(PatreonBaseIE):
         },
         'skip': 'Patron-only content',
     }, {
+        # FIXME: Fails due to no description extracted
         # m3u8 video (https://github.com/yt-dlp/yt-dlp/issues/2277)
         'url': 'https://www.patreon.com/posts/video-sketchbook-32452882',
         'info_dict': {
@@ -220,6 +222,7 @@ class PatreonIE(PatreonBaseIE):
             'channel_id': '2147162',
             'uploader_url': 'https://www.patreon.com/yaboyroshi',
         },
+        'skip': 'HTTP Error 401 for m3u8 request; site now requires login to play the video',
     }, {
         # NSFW vimeo embed URL
         'url': 'https://www.patreon.com/posts/4k-spiderman-4k-96414599',
@@ -242,6 +245,7 @@ class PatreonIE(PatreonBaseIE):
         },
         'params': {'skip_download': 'm3u8'},
         'expected_warnings': ['Failed to parse XML: not well-formed'],
+        'skip': 'Video removed',
     }, {
         # multiple attachments/embeds
         'url': 'https://www.patreon.com/posts/holy-wars-solos-100601977',
@@ -285,6 +289,7 @@ class PatreonIE(PatreonBaseIE):
         },
         'params': {'getcomments': True},
     }, {
+        # FIXME: Error: No supported media found in this post
         # Inlined media in post; uses _extract_from_media_api
         'url': 'https://www.patreon.com/posts/scottfalco-146966245',
         'info_dict': {
@@ -304,6 +309,26 @@ class PatreonIE(PatreonBaseIE):
             'timestamp': 1767061800,
             'upload_date': '20251230',
         },
+    }, {
+        # FIXME: need to extract description
+        'url': 'https://www.patreon.com/Insanimate/posts/meatcanyon-in-142663524',
+        'md5': '132332e3bb345f75d8b471242346dee6',
+        'info_dict': {
+            'id': '142663524',
+            'ext': 'mp4',
+            'title': 'Meatcanyon in Playground',
+            'uploader': 'Insanimate',
+            'uploader_id': '2828146',
+            'uploader_url': 'https://www.patreon.com/Insanimate',
+            'channel_id': '6260877',
+            'channel_url': 'https://www.patreon.com/Insanimate',
+            'channel_follower_count': int,
+            'comment_count': int,
+            'like_count': int,
+            'thumbnail': 're:^https?://.*$',
+            'timestamp': 1762101034,
+            'upload_date': '20251102',
+        },
     }]
     _RETURN_TYPE = 'video'
     _HTTP_HEADERS = {
@@ -357,7 +382,7 @@ class PatreonIE(PatreonBaseIE):
         post = self._call_api(
             f'posts/{video_id}', video_id, query={
                 'fields[media]': 'download_url,mimetype,size_bytes,file_name',
-                'fields[post]': 'comment_count,content,embed,image,like_count,post_file,published_at,title,current_user_can_view',
+                'fields[post]': 'comment_count,content,content_teaser_text,cleaned_teaser_text,embed,image,like_count,post_file,published_at,title,current_user_can_view',
                 'fields[user]': 'full_name,url',
                 'fields[post_tag]': 'value',
                 'fields[campaign]': 'url,name,patron_count',
@@ -367,7 +392,7 @@ class PatreonIE(PatreonBaseIE):
         attributes = post['data']['attributes']
         info = traverse_obj(attributes, {
             'title': ('title', {str.strip}),
-            'description': ('content', {clean_html}),
+            'description': (('content', 'content_teaser_text', 'cleaned_teaser_text'), {clean_html}, any),
             'thumbnail': ('image', ('large_url', 'url'), {url_or_none}, any),
             'timestamp': ('published_at', {parse_iso8601}),
             'like_count': ('like_count', {int_or_none}),

yt_dlp/utils/_utils.py

@@ -5218,12 +5218,17 @@ class _UnsafeExtensionError(Exception):
         'sbv',
     ])
 
+    _enabled = True
+
     def __init__(self, extension, /):
         super().__init__(f'unsafe file extension: {extension!r}')
         self.extension = extension
 
     @classmethod
     def sanitize_extension(cls, extension, /, *, prepend=False, _allowed_exts=()):
+        if not cls._enabled:
+            return extension
+
         if extension is None:
             return None
 

yt_dlp/utils/networking.py

@@ -64,7 +64,7 @@ class HTTPHeaderDict(dict):
             other = other.sensitive()
         if isinstance(other, dict):
             self.update(other)
-            return
+            return self
         return NotImplemented
 
     def __or__(self, other, /) -> typing.Self: