Browse Source

[jsinterp] Disable comment support

We need a proper lexer to be able to understand YouTube's code, which contains /* inside of strings.
For now it's sufficient to just disable comment support altogether.

Fixes #4976, fixes #4979, fixes #4980, fixes #4981, fixes #4982.
Closes #4977.
Philipp Hagemeister 10 years ago
parent
commit
3eff81fbf7
3 changed files with 18 additions and 4 deletions
  1. 11 0
      test/test_jsinterp.py
  2. 6 0
      test/test_youtube_signature.py
  3. 1 4
      youtube_dl/jsinterp.py

+ 11 - 0
test/test_jsinterp.py

@@ -70,6 +70,8 @@ class TestJSInterpreter(unittest.TestCase):
         self.assertEqual(jsi.call_function('f'), -11)
 
     def test_comments(self):
+        'Skipping: Not yet fully implemented'
+        return
         jsi = JSInterpreter('''
         function x() {
             var x = /* 1 + */ 2;
@@ -80,6 +82,15 @@ class TestJSInterpreter(unittest.TestCase):
         ''')
         self.assertEqual(jsi.call_function('x'), 52)
 
+        jsi = JSInterpreter('''
+        function f() {
+            var x = "/*";
+            var y = 1 /* comment */ + 2;
+            return y;
+        }
+        ''')
+        self.assertEqual(jsi.call_function('f'), 3)
+
     def test_precedence(self):
         jsi = JSInterpreter('''
         function x() {

+ 6 - 0
test/test_youtube_signature.py

@@ -64,6 +64,12 @@ _TESTS = [
         'js',
         '4646B5181C6C3020DF1D9C7FCFEA.AD80ABF70C39BD369CCCAE780AFBB98FA6B6CB42766249D9488C288',
         '82C8849D94266724DC6B6AF89BBFA087EACCD963.B93C07FBA084ACAEFCF7C9D1FD0203C6C1815B6B'
+    ),
+    (
+        'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js',
+        'js',
+        '312AA52209E3623129A412D56A40F11CB0AF14AE.3EE09501CB14E3BCDC3B2AE808BF3F1D14E7FBF12',
+        '112AA5220913623229A412D56A40F11CB0AF14AE.3EE0950FCB14EEBCDC3B2AE808BF331D14E7FBF3',
     )
 ]
 

+ 1 - 4
youtube_dl/jsinterp.py

@@ -30,13 +30,10 @@ class JSInterpreter(object):
     def __init__(self, code, objects=None):
         if objects is None:
             objects = {}
-        self.code = self._remove_comments(code)
+        self.code = code
         self._functions = {}
         self._objects = objects
 
-    def _remove_comments(self, code):
-        return re.sub(r'(?s)/\*.*?\*/', '', code)
-
     def interpret_statement(self, stmt, local_vars, allow_recursion=100):
         if allow_recursion < 0:
             raise ExtractorError('Recursion limit reached')