Merge d07eb72f6f 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/fox.py

@@ -10,17 +10,19 @@ from ..networking.exceptions import HTTPError
 from ..utils import (
     ExtractorError,
     int_or_none,
+    join_nonempty,
     parse_age_limit,
     parse_duration,
     traverse_obj,
     try_get,
     unified_timestamp,
     url_or_none,
+    urljoin,
 )
 
 
 class FOXIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?fox(?:sports)?\.com/(?:watch|replay)/(?P<id>[\da-fA-F]+)'
+    _VALID_URL = r'https?://(?:www\.)?fox(?:sports)?\.com/(?:watch|replay)/(?!play-)(?P<id>[\w-]+)'
     _TESTS = [{
         # clip
         'url': 'https://www.fox.com/watch/4b765a60490325103ea69888fb2bd4e8/',
@@ -33,8 +35,9 @@ class FOXIE(InfoExtractor):
             'duration': 102,
             'timestamp': 1504291893,
             'upload_date': '20170901',
-            'creator': 'FOX',
-            'series': 'Gotham',
+            'creators': ['FOX'],
+            # actual series name 'Gotham' is no longer returned by the API
+            'series': 'Aftermath: Bruce Wayne Develops Into The Dark Knight',
             'age_limit': 14,
             'episode': 'Aftermath: Bruce Wayne Develops Into The Dark Knight',
             'thumbnail': r're:^https?://.*\.jpg$',
@@ -42,6 +45,24 @@ class FOXIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # XML endpoint
+        'url': 'https://www.foxsports.com/watch/fmc-m2du80v5ewz11pbw',
+        'md5': '5451a633a5ca87b582a4d025df6852e6',
+        'info_dict': {
+            'id': 'fmc-m2du80v5ewz11pbw',
+            'ext': 'mp4',
+            'title': 'WWE FRIDAY NIGHT SMACKDOWN',
+            'description': 'From Fiserv Forum in Milwaukee, WI',
+            'duration': 5367,
+            'timestamp': 1698176671,
+            'upload_date': '20231024',
+            'creators': ['fox-digital'],
+            'series': 'WWE FRIDAY NIGHT SMACKDOWN',
+            'age_limit': 0,
+            'episode': 'WWE FRIDAY NIGHT SMACKDOWN',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
     }, {
         # episode, geo-restricted
         'url': 'https://www.fox.com/watch/087036ca7f33c8eb79b08152b4dd75c1/',
@@ -57,9 +78,13 @@ class FOXIE(InfoExtractor):
     }]
     _GEO_BYPASS = False
     _HOME_PAGE_URL = 'https://www.fox.com/'
-    _API_KEY = '6E9S4bmcoNnZwVLOHywOv8PJEdu76cM9'
+    _API_KEY = '6E9S4bmcoNnZwVLOHywOv8PJEdu76cM9'  # sports: 'cf289e299efdfa39fb6316f259d1de93'
     _access_token = None
     _device_id = compat_str(uuid.uuid4())
+    _XML_NS = {
+        'vmap': 'http://www.iab.net/videosuite/vmap',
+        'yospacenet': 'http://www.yospace.com/extension',
+    }
 
     def _call_api(self, path, video_id, data=None):
         headers = {
@@ -69,7 +94,7 @@ class FOXIE(InfoExtractor):
             headers['Authorization'] = 'Bearer ' + self._access_token
         try:
             return self._download_json(
-                'https://api3.fox.com/v2.0/' + path,
+                urljoin('https://api3.fox.com/v2.0/', path),
                 video_id, data=data, headers=headers)
         except ExtractorError as e:
             if isinstance(e.cause, HTTPError) and e.cause.status == 403:
@@ -103,8 +128,8 @@ class FOXIE(InfoExtractor):
             'previewpassmvpd?device_id=%s&mvpd_id=TempPass_fbcfox_60min' % self._device_id,
             video_id)['accessToken']
 
-        video = self._call_api('watch', video_id, data=json.dumps({
-            'capabilities': ['drm/widevine', 'fsdk/yo'],
+        video = self._call_api('https://prod.api.video.fox/v2.0/watch', video_id, data=json.dumps({
+            'capabilities': ['fsdk/yo/v3'],
             'deviceWidth': 1280,
             'deviceHeight': 720,
             'maxRes': '720p',
@@ -119,13 +144,16 @@ class FOXIE(InfoExtractor):
             'privacy': {'us': '1---'},
             'siteSection': '',
             'streamType': 'vod',
-            'streamId': video_id}).encode('utf-8'))
-
-        title = video['name']
-        release_url = video['url']
+            'streamId': video_id,
+        }).encode())
 
         try:
-            m3u8_url = self._download_json(release_url, video_id)['playURL']
+            if playback_url := traverse_obj(video, ('playbackUrl', {url_or_none})):
+                xml_data = self._download_xml(playback_url, video_id)
+                stream = xml_data.find('vmap:Extensions/vmap:Extension/yospacenet:Stream', self._XML_NS)
+                m3u8_url = join_nonempty('https://', stream.get('urlDomain'), stream.get('urlSuffix'), delim='')
+            else:
+                m3u8_url = self._download_json(video['url'], video_id)['playURL']
         except ExtractorError as e:
             if isinstance(e.cause, HTTPError) and e.cause.status == 403:
                 error = self._parse_json(e.cause.response.read().decode(), video_id)
@@ -133,9 +161,11 @@ class FOXIE(InfoExtractor):
                     self.raise_geo_restricted(countries=['US'])
                 raise ExtractorError(error['description'], expected=True)
             raise
-        formats = self._extract_m3u8_formats(
-            m3u8_url, video_id, 'mp4',
-            entry_protocol='m3u8_native', m3u8_id='hls')
+
+        if not m3u8_url or m3u8_url == 'https://':
+            raise ExtractorError('Unable to extract m3u8 url')
+
+        formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4', m3u8_id='hls')
 
         data = try_get(
             video, lambda x: x['trackingData']['properties'], dict) or {}
@@ -160,7 +190,7 @@ class FOXIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'title': title,
+            'title': video.get('name'),
             'formats': formats,
             'description': video.get('description'),
             'duration': duration,

yt_dlp/extractor/foxsports.py

@@ -5,7 +5,7 @@ from ..utils import float_or_none, make_archive_id, smuggle_url
 
 
 class FoxSportsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?foxsports\.com/watch/(?P<id>[\w-]+)'
+    _VALID_URL = r'https?://(?:www\.)?foxsports\.com/watch/(?P<id>play-[\w-]+)'
     _TESTS = [{
         'url': 'https://www.foxsports.com/watch/play-612168c6700004b',
         'info_dict': {

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)