Merge f51d16b307 into 64766459e3

Authored by: Grub4K

.github/workflows/core.yml

@@ -53,7 +53,7 @@ jobs:
       with:
         python-version: ${{ matrix.python-version }}
     - name: Install test requirements
-      run: python3 ./devscripts/install_deps.py --include dev --include curl_cffi
+      run: python3 ./devscripts/install_deps.py --include dev --include curl-cffi
     - name: Run tests
       continue-on-error: False
       run: |

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/orf.py

@@ -569,19 +569,20 @@ class ORFFM4StoryIE(InfoExtractor):
 
 class ORFONIE(InfoExtractor):
     IE_NAME = 'orf:on'
-    _VALID_URL = r'https?://on\.orf\.at/video/(?P<id>\d{8})/(?P<slug>[\w-]+)'
+    _VALID_URL = r'https?://on\.orf\.at/video/(?P<id>\d+)(/(?P<slug>[\w-]+))?'
     _TESTS = [{
-        'url': 'https://on.orf.at/video/14210000/school-of-champions-48',
+        'url': 'https://on.orf.at/video/3220355',
+        'md5': 'f94d98e667cf9a3851317efb4e136662',
         'info_dict': {
-            'id': '14210000',
+            'id': '3220355',
             'ext': 'mp4',
-            'duration': 2651.08,
-            'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0167/98/thumb_16697671_segments_highlight_teaser.jpeg',
-            'title': 'School of Champions (4/8)',
-            'description': 'md5:d09ad279fc2e8502611e7648484b6afd',
+            'duration': 445.04,
+            'thumbnail': 'https://api-tvthek.orf.at/assets/segments/0002/60/thumb_159573_segments_highlight_teaser.png',
+            'title': '50 Jahre Burgenland: Der Festumzug',
+            'description': 'md5:1560bf855119544ee8c4fa5376a2a6b0',
             'media_type': 'episode',
-            'timestamp': 1706472362,
-            'upload_date': '20240128',
+            'timestamp': 52916400,
+            'upload_date': '19710905',
         }
     }]
 
@@ -590,6 +591,9 @@ class ORFONIE(InfoExtractor):
         api_json = self._download_json(
             f'https://api-tvthek.orf.at/api/v4.3/public/episode/encrypted/{encrypted_id}', display_id)
 
+        if traverse_obj(api_json, 'is_drm_protected'):
+            self.report_drm(video_id)
+
         formats, subtitles = [], {}
         for manifest_type in traverse_obj(api_json, ('sources', {dict.keys}, ...)):
             for manifest_url in traverse_obj(api_json, ('sources', manifest_type, ..., 'src', {url_or_none})):
@@ -604,10 +608,33 @@ class ORFONIE(InfoExtractor):
                 formats.extend(fmts)
                 self._merge_subtitles(subs, target=subtitles)
 
+        for subtitle_type in ['vtt']:  # not working formats 'xml', 'srt', 'sami', 'ttml', 'stl'
+            subtitle_url = traverse_obj(api_json, ('_embedded', 'subtitle', f'{subtitle_type}_url'), {str})
+            if subtitle_url is None:
+                continue
+            self._merge_subtitles({
+                'de': [
+                    {
+                        'url': subtitle_url,
+                        'ext': f'{subtitle_type}',
+                    }
+                ],
+            }, target=subtitles)
+
+        age_classification = traverse_obj(api_json, ('age_classification'), {str})
+        age_limit = None
+        if isinstance(age_classification, str) and len(age_classification) != 0:
+            # age_classification is in the format `<age:int>+`
+            # example: "6+" or "18+"
+            age_limit_str = age_classification[:-1]
+            if age_limit_str.isdigit():
+                age_limit = int(age_limit_str)
+
         return {
             'id': video_id,
             'formats': formats,
             'subtitles': subtitles,
+            'age_limit': age_limit,
             **traverse_obj(api_json, {
                 'duration': ('duration_second', {float_or_none}),
                 'title': (('title', 'headline'), {str}),
@@ -618,6 +645,8 @@ class ORFONIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id, display_id = self._match_valid_url(url).group('id', 'slug')
+        if display_id is None:
+            display_id = video_id
         webpage = self._download_webpage(url, display_id)
 
         return {

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)