浏览代码

[YouTube] Fix signature function extraction for `2f1832d2`
* `_` was omitted from patterns
* thx yt-dlp/yt-dlp#11801

Co-authored-by: bashonly

dirkf 8 月之前
父节点
当前提交
eeafbbc3e5
共有 2 个文件被更改,包括 21 次插入12 次删除
  1. 9 0
      test/test_youtube_signature.py
  2. 12 12
      youtube_dl/extractor/youtube.py

+ 9 - 0
test/test_youtube_signature.py

@@ -79,6 +79,11 @@ _SIG_TESTS = [
         '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
         'MyOSJXtKI3m-uME_jv7-pT12gOFC02RFkGoqWpzE0Cs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
     ),
+    (
+        'https://www.youtube.com/s/player/2f1832d2/player_ias.vflset/en_US/base.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        '0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xxAj7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJ2OySqa0q',
+    ),
 ]
 
 _NSIG_TESTS = [
@@ -210,6 +215,10 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/f8f53e1a/player_ias.vflset/en_US/base.js',
         'VTQOUOv0mCIeJ7i8kZB', 'kcfD8wy0sNLyNQ',
     ),
+    (
+        'https://www.youtube.com/s/player/2f1832d2/player_ias.vflset/en_US/base.js',
+        'YWt1qdbe8SAfkoPHW5d', 'RrRjWQOJmBiP',
+    ),
 ]
 
 

+ 12 - 12
youtube_dl/extractor/youtube.py

@@ -1587,19 +1587,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         # ;N&&(N=sig(decodeURIComponent(N)),J.set(R,encodeURIComponent(N)));return J};
         # {var H=u,k=f.sp,v=sig(decodeURIComponent(f.s));H.set(k,encodeURIComponent(v))}
         funcname = self._search_regex(
-            (r'\b(?P<var>[a-zA-Z0-9$]+)&&\((?P=var)=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\((?P=var)\)\)',
-             r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*(?P<arg>[a-zA-Z0-9$]+)\s*\)\s*{\s*(?P=arg)\s*=\s*(?P=arg)\.split\(\s*""\s*\)\s*;\s*[^}]+;\s*return\s+(?P=arg)\.join\(\s*""\s*\)',
-             r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)(?:;[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\))?',
+            (r'\b(?P<var>[\w$]+)&&\((?P=var)=(?P<sig>[\w$]{2,})\(decodeURIComponent\((?P=var)\)\)',
+             r'(?P<sig>[\w$]+)\s*=\s*function\(\s*(?P<arg>[\w$]+)\s*\)\s*{\s*(?P=arg)\s*=\s*(?P=arg)\.split\(\s*""\s*\)\s*;\s*[^}]+;\s*return\s+(?P=arg)\.join\(\s*""\s*\)',
+             r'(?:\b|[^\w$])(?P<sig>[\w$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)(?:;[\w$]{2}\.[\w$]{2}\(a,\d+\))?',
              # Old patterns
-             r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
-             r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
-             r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
+             r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[\w$]+)\(',
+             r'\b[\w]+\s*&&\s*[\w]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[\w$]+)\(',
+             r'\bm=(?P<sig>[\w$]{2,})\(decodeURIComponent\(h\.s\)\)',
              # Obsolete patterns
-             r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
-             r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
-             r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
-             r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
-             r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
+             r'("|\')signature\1\s*,\s*(?P<sig>[\w$]+)\(',
+             r'\.sig\|\|(?P<sig>[\w$]+)\(',
+             r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[\w$]+)\(',
+             r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[\w$]+)\(',
+             r'\bc\s*&&\s*[\w]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[\w$]+)\('),
             jscode, 'Initial JS player signature function name', group='sig')
 
         jsi = JSInterpreter(jscode)
@@ -1699,7 +1699,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     (?P<name>(?!\d)[a-zA-Z\d_$]+)\s*=\s*function\((?!\d)[a-zA-Z\d_$]+\)
                     \s*\{(?:(?!};).)+?(?:
                         ["']enhanced_except_ |
-                        return\s*(?P<q>"|')[a-zA-Z\d-]+_w8_(?P=q)\s*\+\s*[a-zA-Z0-9_$]+
+                        return\s*(?P<q>"|')[a-zA-Z\d-]+_w8_(?P=q)\s*\+\s*[\w$]+
                     )
                 ''', jscode, 'Initial JS player n function name', group='name')
         if not idx: