mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-11-04 08:35:12 +00:00
Code Issues Projects Releases Wiki Activity

[jsinterp] Fix escape in regex

Browse Source
This commit is contained in:
pukkandan
2022-09-01 13:14:04 +05:30
parent b505e8517a
commit 05deb747bb
4 changed files with 16 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
test/test_jsinterp.py
View File

@@ -352,6 +352,11 @@ class TestJSInterpreter(unittest.TestCase):
''')
self.assertEqual(jsi.call_function('x').flags & re.I, re.I)
jsi = JSInterpreter('''
function x() { let a=/,][}",],()}(\[)/; return a; }
''')
self.assertEqual(jsi.call_function('x').pattern, r',][}",],()}(\[)')
def test_char_code_at(self):
jsi = JSInterpreter('function x(i){return "test".charCodeAt(i)}')
self.assertEqual(jsi.call_function('x', 0), 116)

4
test/test_youtube_signature.py
View File

@@ -122,6 +122,10 @@ _NSIG_TESTS = [
'https://www.youtube.com/s/player/113ca41c/player_ias.vflset/en_US/base.js',
'cgYl-tlYkhjT7A', 'hI7BBr2zUgcmMg',
),
(
'https://www.youtube.com/s/player/c57c113c/player_ias.vflset/en_US/base.js',
'M92UUMHa8PdvPd3wyM', '3hPqLJsiNZx7yA',
),
]

2
yt_dlp/extractor/youtube.py
View File

@@ -2702,7 +2702,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _extract_n_function_code(self, video_id, player_url):
player_id = self._extract_player_info(player_url)
func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.08.19.2')
func_code = self.cache.load('youtube-nsig', player_id, min_ver='2022.09.1')
jscode = func_code or self._load_player(video_id, player_url)
jsi = JSInterpreter(jscode)

11
yt_dlp/jsinterp.py
View File

@@ -245,11 +245,12 @@ class JSInterpreter:
counters[_MATCHING_PARENS[char]] += 1
elif not in_quote and char in counters:
counters[char] -= 1
elif not escaping and char in _QUOTES and in_quote in (char, None):
if in_quote or after_op or char != '/':
in_quote = None if in_quote and not in_regex_char_group else char
elif in_quote == '/' and char in '[]':
in_regex_char_group = char == '['
elif not escaping:
if char in _QUOTES and in_quote in (char, None):
if in_quote or after_op or char != '/':
in_quote = None if in_quote and not in_regex_char_group else char
elif in_quote == '/' and char in '[]':
in_regex_char_group = char == '['
escaping = not escaping and in_quote and char == '\\'
after_op = not in_quote and char in OP_CHARS or (char.isspace() and after_op)