From 011b4a04db2a636c3ef0a0ad4e2d3ae482c9fd76 Mon Sep 17 00:00:00 2001 From: bashonly <88596187+bashonly@users.noreply.github.com> Date: Wed, 31 Jul 2024 16:19:30 -0500 Subject: [PATCH] [ie/youtube] Fix `n` function name extraction for player `20dfca59` (#10611) Closes #10608 Authored by: bashonly --- test/test_youtube_signature.py | 4 ++++ yt_dlp/extractor/youtube.py | 26 ++++++++++++++++++++------ 2 files changed, 24 insertions(+), 6 deletions(-) diff --git a/test/test_youtube_signature.py b/test/test_youtube_signature.py index ae167d16d..d37df7a2e 100644 --- a/test/test_youtube_signature.py +++ b/test/test_youtube_signature.py @@ -175,6 +175,10 @@ _NSIG_TESTS = [ 'https://www.youtube.com/s/player/3400486c/player_ias.vflset/en_US/base.js', 'lL46g3XifCKUZn1Xfw', 'z767lhet6V2Skl', ), + ( + 'https://www.youtube.com/s/player/20dfca59/player_ias.vflset/en_US/base.js', + '-fLCxedkAk4LUTK2', 'O8kfRq1y1eyHGw', + ), ] diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py index 1a3e286c6..4993ce397 100644 --- a/yt_dlp/extractor/youtube.py +++ b/yt_dlp/extractor/youtube.py @@ -3174,18 +3174,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return ret def _extract_n_function_name(self, jscode): + # Examples (with placeholders nfunc, narray, idx): + # * .get("n"))&&(b=nfunc(b) + # * .get("n"))&&(b=narray[idx](b) + # * b=String.fromCharCode(110),c=a.get(b))&&c=narray[idx](c) + # * a.D&&(b="nn"[+a.D],c=a.get(b))&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("") + # * a.D&&(PL(a),b=a.j.n||null)&&(b=narray[0](b),a.set("n",b),narray.length||nfunc("") funcname, idx = self._search_regex( r'''(?x) (?: \.get\("n"\)\)&&\(b=| (?: b=String\.fromCharCode\(110\)| - ([a-zA-Z0-9$.]+)&&\(b="nn"\[\+\1\] - ),c=a\.get\(b\)\)&&\(c= - ) - (?P[a-zA-Z0-9$]+)(?:\[(?P\d+)\])?\([a-zA-Z0-9]\)''', - jscode, 'Initial JS player n function name', group=('nfunc', 'idx')) - if not idx: + (?P[a-zA-Z0-9_$.]+)&&\(b="nn"\[\+(?P=str_idx)\] + ),c=a\.get\(b\)\)&&\(c=| + \b(?P[a-zA-Z0-9_$]+)= + )(?P[a-zA-Z0-9_$]+)(?:\[(?P\d+)\])?\([a-zA-Z]\) + (?(var),[a-zA-Z0-9_$]+\.set\("n"\,(?P=var)\),(?P=nfunc)\.length)''', + jscode, 'n function name', group=('nfunc', 'idx'), default=(None, None)) + if not funcname: + self.report_warning('Falling back to generic n function search') + return self._search_regex( + r'''(?xs) + ;\s*(?P[a-zA-Z0-9_$]+)\s*=\s*function\([a-zA-Z0-9_$]+\) + \s*\{(?:(?!};).)+?["']enhanced_except_''', + jscode, 'Initial JS player n function name', group='name') + elif not idx: return funcname return json.loads(js_to_json(self._search_regex(