Merge 43655bc01b into 64766459e3

Authored by: Grub4K

test/test_utils.py

@@ -2059,7 +2059,22 @@ Line 1
         assert extract_basic_auth('http://user:pass@foo.bar') == ('http://foo.bar', 'Basic dXNlcjpwYXNz')
 
     @unittest.skipUnless(compat_os_name == 'nt', 'Only relevant on Windows')
-    def test_Popen_windows_escaping(self):
+    def test_windows_escaping(self):
+        tests = [
+            'test"&',
+            '%CMDCMDLINE:~-1%&',
+            'a\nb',
+            '"',
+            '\\',
+            '!',
+            '^!',
+            'a \\ b',
+            'a \\" b',
+            'a \\ b\\',
+            # We replace \r with \n
+            ('a\r\ra', 'a\n\na'),
+        ]
+
         def run_shell(args):
             stdout, stderr, error = Popen.run(
                 args, text=True, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
@@ -2067,15 +2082,18 @@ Line 1
             assert not error
             return stdout
 
-        # Test escaping
-        assert run_shell(['echo', 'test"&']) == '"test""&"\n'
-        assert run_shell(['echo', '%CMDCMDLINE:~-1%&']) == '"%CMDCMDLINE:~-1%&"\n'
-        assert run_shell(['echo', 'a\nb']) == '"a"\n"b"\n'
-        assert run_shell(['echo', '"']) == '""""\n'
-        assert run_shell(['echo', '\\']) == '\\\n'
-        # Test if delayed expansion is disabled
-        assert run_shell(['echo', '^!']) == '"^!"\n'
-        assert run_shell('echo "^!"') == '"^!"\n'
+        for argument in tests:
+            if isinstance(argument, str):
+                expected = argument
+            else:
+                argument, expected = argument
+
+            args = [sys.executable, '-c', 'import sys; print(end=sys.argv[1])', argument, 'end']
+            assert run_shell(args) == expected
+
+            escaped = shell_quote(argument, shell=True)
+            args = f'{sys.executable} -c "import sys; print(end=sys.argv[1])" {escaped} end'
+            assert run_shell(args) == expected
 
 
 if __name__ == '__main__':

yt_dlp/extractor/pornhub.py

@@ -138,12 +138,13 @@ class PornHubIE(PornHubBaseIE):
     _EMBED_REGEX = [r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub(?:premium)?\.(?:com|net|org)/embed/[\da-z]+)']
     _TESTS = [{
         'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015',
-        'md5': 'a6391306d050e4547f62b3f485dd9ba9',
+        'md5': '4d4a4e9178b655776f86cf89ecaf0edf',
         'info_dict': {
             'id': '648719015',
             'ext': 'mp4',
             'title': 'Seductive Indian beauty strips down and fingers her pink pussy',
-            'uploader': 'Babes',
+            'uploader_id': str,
+            'uploader': 'BABES-COM',
             'upload_date': '20130628',
             'timestamp': 1372447216,
             'duration': 361,
@@ -155,6 +156,8 @@ class PornHubIE(PornHubBaseIE):
             'tags': list,
             'categories': list,
             'cast': list,
+            'thumbnail': str,
+            'production': list,
         },
     }, {
         # non-ASCII title
@@ -208,11 +211,23 @@ class PornHubIE(PornHubBaseIE):
         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph601dc30bae19a',
         'info_dict': {
             'id': 'ph601dc30bae19a',
+            'ext': 'mp4',
             'uploader': 'Projekt Melody',
             'uploader_id': 'projekt-melody',
             'upload_date': '20210205',
             'title': '"Welcome to My Pussy Mansion" - CB Stream (02/03/21)',
             'thumbnail': r're:https?://.+',
+            'duration': 8173,
+            'timestamp': 1612564932,
+            'age_limit': 18,
+            'tags': list,
+            'like_count': int,
+            'view_count': int,
+            'categories': list,
+            'dislike_count': int,
+            'production': list,
+            'cast': list,
+            'comment_count': int,
         },
     }, {
         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph557bbb6676d2d',
@@ -263,7 +278,33 @@ class PornHubIE(PornHubBaseIE):
     }, {
         'url': 'http://pornhubvybmsymdol4iibwgwtkpwmeyd6luq2gxajgjzfjvotyt5zhyd.onion/view_video.php?viewkey=ph5a9813bfa7156',
         'only_matching': True,
-    }]
+    }, {
+        # language spoken, model attributes, production
+        'url': 'https://www.pornhub.com/view_video.php?viewkey=65a6ca42725f2',
+        'info_dict': {
+            'id': '65a6ca42725f2',
+            'ext': 'mp4',
+            'title': 'Busty blonde sucks the juice out of me',
+            'uploader_id': 'egon-kowalski',
+            'cast': [],
+            'thumbnail': 'https://ei.phncdn.com/videos/202401/16/446618441/thumbs_14/(m=eaAaGwObaaaa)(mh=dIONqa8IT3Sw1RqL)5.jpg',
+            'uploader': 'Egon Kowalski',
+            'upload_date': '20240116',
+            'timestamp': 1705429963,
+            'duration': 358,
+            'view_count': int,
+            'like_count': int,
+            'dislike_count': int,
+            'comment_count': int,
+            'age_limit': 18,
+            'tags': list,
+            'categories': list,
+            'language_spoken': ['german'],
+            'model_attributes': ['tattoos', 'white', 'no piercing'],
+            'production': ['homemade'],
+        }
+    }
+    ]
 
     def _extract_count(self, pattern, webpage, name):
         return str_to_int(self._search_regex(pattern, webpage, '%s count' % name, default=None))
@@ -488,6 +529,15 @@ class PornHubIE(PornHubBaseIE):
             if div:
                 return [clean_html(x).strip() for x in re.findall(r'(?s)<a[^>]+\bhref=[^>]+>.+?</a>', div)]
 
+        def get_model_attributes():
+            model_attributes = re.findall(
+                r'<a[^>]*\bdata-label=["\']model_attributes["\'][^>]*>(.*?)<\/a>',
+                webpage,
+                re.DOTALL)
+
+            if model_attributes:
+                return [x.strip().lower() for x in model_attributes]
+
         info = self._search_json_ld(webpage, video_id, default={})
         # description provided in JSON-LD is irrelevant
         info['description'] = None
@@ -509,6 +559,9 @@ class PornHubIE(PornHubBaseIE):
             'tags': extract_list('tags'),
             'categories': extract_list('categories'),
             'cast': extract_list('pornstars'),
+            'production': extract_list('production'),
+            'language_spoken': extract_list('langSpoken'),
+            'model_attributes': get_model_attributes(),
             'subtitles': subtitles,
         }, info)
 

yt_dlp/utils/_utils.py

@@ -1638,16 +1638,14 @@ def get_filesystem_encoding():
     return encoding if encoding is not None else 'utf-8'
 
 
-_WINDOWS_QUOTE_TRANS = str.maketrans({'"': '\\"', '\\': '\\\\'})
+_WINDOWS_QUOTE_TRANS = str.maketrans({'"': R'\"'})
 _CMD_QUOTE_TRANS = str.maketrans({
     # Keep quotes balanced by replacing them with `""` instead of `\\"`
     '"': '""',
-    # Requires a variable `=` containing `"^\n\n"` (set in `utils.Popen`)
+    # These require an env-variable `=` containing `"^\n\n"` (set in `utils.Popen`)
     # `=` should be unique since variables containing `=` cannot be set using cmd
     '\n': '%=%',
-    # While we are only required to escape backslashes immediately before quotes,
-    # we instead escape all of 'em anyways to be consistent
-    '\\': '\\\\',
+    '\r': '%=%',
     # Use zero length variable replacement so `%` doesn't get expanded
     # `cd` is always set as long as extensions are enabled (`/E:ON` in `utils.Popen`)
     '%': '%%cd:~,%',
@@ -1656,19 +1654,14 @@ _CMD_QUOTE_TRANS = str.maketrans({
 
 def shell_quote(args, *, shell=False):
     args = list(variadic(args))
-    if any(isinstance(item, bytes) for item in args):
-        deprecation_warning('Passing bytes to utils.shell_quote is deprecated')
-        encoding = get_filesystem_encoding()
-        for index, item in enumerate(args):
-            if isinstance(item, bytes):
-                args[index] = item.decode(encoding)
 
     if compat_os_name != 'nt':
         return shlex.join(args)
 
     trans = _CMD_QUOTE_TRANS if shell else _WINDOWS_QUOTE_TRANS
     return ' '.join(
-        s if re.fullmatch(r'[\w#$*\-+./:?@\\]+', s, re.ASCII) else s.translate(trans).join('""')
+        s if re.fullmatch(r'[\w#$*\-+./:?@\\]+', s, re.ASCII)
+        else re.sub(r'(\\+)("|$)', r'\1\1\2', s).translate(trans).join('""')
         for s in args)