From ad01fa6ccadd1ecade8002e937492a141d3b8f25 Mon Sep 17 00:00:00 2001 From: dirkf Date: Wed, 19 Jun 2024 21:44:27 +0100 Subject: [PATCH] [jsinterp] Add Debugger from yt-dlp * https://github.com/yt-dlp/yt-dlp/commit/8f53dc4 * thx pukkandan --- test/test_jsinterp.py | 8 ++++--- test/test_youtube_signature.py | 4 ++++ youtube_dl/extractor/common.py | 1 - youtube_dl/jsinterp.py | 42 +++++++++++++++++++++++++++++++++- 4 files changed, 50 insertions(+), 5 deletions(-) diff --git a/test/test_jsinterp.py b/test/test_jsinterp.py index 91b12f544..da8e98020 100644 --- a/test/test_jsinterp.py +++ b/test/test_jsinterp.py @@ -577,9 +577,11 @@ class TestJSInterpreter(unittest.TestCase): def test_unary_operators(self): jsi = JSInterpreter('function f(){return 2 - - - 2;}') self.assertEqual(jsi.call_function('f'), 0) - # fails - # jsi = JSInterpreter('function f(){return 2 + - + - - 2;}') - # self.assertEqual(jsi.call_function('f'), 0) + jsi = JSInterpreter('function f(){return 2 + - + - - 2;}') + self.assertEqual(jsi.call_function('f'), 0) + # https://github.com/ytdl-org/youtube-dl/issues/32815 + jsi = JSInterpreter('function f(){return 0 - 7 * - 6;}') + self.assertEqual(jsi.call_function('f'), 42) """ # fails so far def test_packed(self): diff --git a/test/test_youtube_signature.py b/test/test_youtube_signature.py index f45dfec7c..cafba7a5c 100644 --- a/test/test_youtube_signature.py +++ b/test/test_youtube_signature.py @@ -158,6 +158,10 @@ _NSIG_TESTS = [ 'https://www.youtube.com/s/player/b7910ca8/player_ias.vflset/en_US/base.js', '_hXMCwMt9qE310D', 'LoZMgkkofRMCZQ', ), + ( + 'https://www.youtube.com/s/player/590f65a6/player_ias.vflset/en_US/base.js', + '1tm7-g_A9zsI8_Lay_', 'xI4Vem4Put_rOg', + ), ] diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index b10e84416..9b0016d07 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -3033,7 +3033,6 @@ class InfoExtractor(object): transform_source=transform_source, default=None) def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs): - # allow passing `transform_source` through to _find_jwplayer_data() transform_source = kwargs.pop('transform_source', None) kwfind = compat_kwargs({'transform_source': transform_source}) if transform_source else {} diff --git a/youtube_dl/jsinterp.py b/youtube_dl/jsinterp.py index 86d902248..e258ebd00 100644 --- a/youtube_dl/jsinterp.py +++ b/youtube_dl/jsinterp.py @@ -14,6 +14,7 @@ from .utils import ( remove_quotes, unified_timestamp, variadic, + write_string, ) from .compat import ( compat_basestring, @@ -220,6 +221,42 @@ class LocalNameSpace(ChainMap): return 'LocalNameSpace%s' % (self.maps, ) +class Debugger(object): + ENABLED = False + + @staticmethod + def write(*args, **kwargs): + level = kwargs.get('level', 100) + + def truncate_string(s, left, right=0): + if s is None or len(s) <= left + right: + return s + return '...'.join((s[:left - 3], s[-right:] if right else '')) + + write_string('[debug] JS: {0}{1}\n'.format( + ' ' * (100 - level), + ' '.join(truncate_string(compat_str(x), 50, 50) for x in args))) + + @classmethod + def wrap_interpreter(cls, f): + def interpret_statement(self, stmt, local_vars, allow_recursion, *args, **kwargs): + if cls.ENABLED and stmt.strip(): + cls.write(stmt, level=allow_recursion) + try: + ret, should_ret = f(self, stmt, local_vars, allow_recursion, *args, **kwargs) + except Exception as e: + if cls.ENABLED: + if isinstance(e, ExtractorError): + e = e.orig_msg + cls.write('=> Raises:', e, '<-|', stmt, level=allow_recursion) + raise + if cls.ENABLED and stmt.strip(): + if should_ret or not repr(ret) == stmt: + cls.write(['->', '=>'][should_ret], repr(ret), '<-|', stmt, level=allow_recursion) + return ret, should_ret + return interpret_statement + + class JSInterpreter(object): __named_object_counter = 0 @@ -416,7 +453,7 @@ class JSInterpreter(object): except Exception as e: if allow_undefined: return JS_Undefined - raise self.Exception('Cannot get index {idx:.100}'.format(**locals()), expr=repr(obj), cause=e) + raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e) def _dump(self, obj, namespace): try: @@ -438,6 +475,7 @@ class JSInterpreter(object): _FINALLY_RE = re.compile(r'finally\s*\{') _SWITCH_RE = re.compile(r'switch\s*\(') + @Debugger.wrap_interpreter def interpret_statement(self, stmt, local_vars, allow_recursion=100): if allow_recursion < 0: raise self.Exception('Recursion limit reached') @@ -797,6 +835,8 @@ class JSInterpreter(object): def eval_method(): if (variable, member) == ('console', 'debug'): + if Debugger.ENABLED: + Debugger.write(self.interpret_expression('[{}]'.format(arg_str), local_vars, allow_recursion)) return types = { 'String': compat_str,