Merge eecd76dd92 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/_extractors.py

@@ -2379,6 +2379,7 @@ from .xhamster import (
     XHamsterEmbedIE,
     XHamsterUserIE,
 )
+from .xiaohongshu import XiaoHongShuIE
 from .ximalaya import (
     XimalayaIE,
     XimalayaAlbumIE

yt_dlp/extractor/xiaohongshu.py

@@ -0,0 +1,85 @@
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    js_to_json,
+    url_or_none,
+)
+from ..utils.traversal import traverse_obj
+
+
+class XiaoHongShuIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.xiaohongshu.com/explore/(?P<id>[a-f0-9]+)'
+    IE_DESC = '小红书'
+    _TESTS = [{
+        'url': 'https://www.xiaohongshu.com/explore/6411cf99000000001300b6d9',
+        'md5': '2a87a77ddbedcaeeda8d7eae61b61228',
+        'info_dict': {
+            'id': '6411cf99000000001300b6d9',
+            'ext': 'mp4',
+            'uploader_id': '5c31698d0000000007018a31',
+            'description': '#今日快乐今日发[话题]# #吃货薯看这里[话题]# #香妃蛋糕[话题]# #小五卷蛋糕[话题]# #新手蛋糕卷[话题]#',
+            'title': '香妃蛋糕也太香了吧🔥不需要卷❗️绝对的友好',
+            'tags': ['今日快乐今日发', '吃货薯看这里', '香妃蛋糕', '小五卷蛋糕', '新手蛋糕卷'],
+            'duration': 101.726,
+            'thumbnail': r're:https?://sns-webpic-qc\.xhscdn\.com/\d+/[a-z0-9]+/[\w]+',
+        }
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        initial_state = self._search_json(
+            r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', display_id, transform_source=js_to_json)
+
+        note_info = traverse_obj(initial_state, ('note', 'noteDetailMap', display_id, 'note'))
+        video_info = traverse_obj(note_info, ('video', 'media', 'stream', ('h264', 'av1', 'h265'), ...))
+
+        formats = []
+        for info in video_info:
+            format_info = traverse_obj(info, {
+                'fps': 'fps',
+                'width': 'width',
+                'height': 'height',
+                'vcodec': 'videoCodec',
+                'acodec': 'audioCodec',
+                'abr': 'audioBitrate',
+                'vbr': 'videoBitrate',
+                'audio_channels': 'audioChannels',
+                'tbr': 'avgBitrate',
+                'format': 'qualityType',
+                'filesize': 'size',
+                'duration': ('duration', {lambda x: float_or_none(x, scale=1000)})
+            })
+
+            formats.extend(traverse_obj(info, (('mediaUrl', ('backupUrls', ...)), {
+                lambda url: url_or_none(url) and {'url': url, 'ext': 'mp4', **format_info}})))
+
+        thumbnails = []
+        for image_info in traverse_obj(note_info, ('imageList', ...)):
+            thumbnail_info = traverse_obj(image_info, {
+                'height': 'height',
+                'width': 'width'
+            })
+            for url in traverse_obj(image_info, (('urlDefault', 'urlPre'), {url_or_none})):
+                thumbnails.append({
+                    'url': url,
+                    **thumbnail_info
+                })
+
+        return {
+            'id': display_id,
+            'formats': formats or [{
+                'url': self._html_search_meta(['og:video'], webpage, fatal=True),
+                'ext': 'mp4'
+            }],
+            'thumbnails': thumbnails or [{
+                'url': self._html_search_meta(['og:image'], webpage, default=None)
+            }],
+            'title': self._html_search_meta(['og:title'], webpage, default=None),
+            **traverse_obj(note_info, {
+                'title': 'title',
+                'description': 'desc',
+                'tags': ('tagList', ..., 'name', {str}),
+                'uploader_id': ('user', 'userId'),
+            }),
+        }

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)