From 4195096ea8da8237a63e1ba3876dc8856b8605c6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Sergey=20M=E2=80=A4?= <dstftw@gmail.com>
Date: Fri, 3 Feb 2017 02:55:06 +0700
Subject: [PATCH] [utils] Improve comments processing in js_to_json (closes
 #11947)

---
 test/test_utils.py  | 24 ++++++++++++++++++++++++
 youtube_dl/utils.py | 20 +++++++++++---------
 2 files changed, 35 insertions(+), 9 deletions(-)

diff --git a/test/test_utils.py b/test/test_utils.py
index 954bb7d8b..edc712f07 100644
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -785,12 +785,24 @@ class TestUtil(unittest.TestCase):
         on = js_to_json('["abc", "def",]')
         self.assertEqual(json.loads(on), ['abc', 'def'])
 
+        on = js_to_json('[/*comment\n*/"abc"/*comment\n*/,/*comment\n*/"def",/*comment\n*/]')
+        self.assertEqual(json.loads(on), ['abc', 'def'])
+
+        on = js_to_json('[//comment\n"abc" //comment\n,//comment\n"def",//comment\n]')
+        self.assertEqual(json.loads(on), ['abc', 'def'])
+
         on = js_to_json('{"abc": "def",}')
         self.assertEqual(json.loads(on), {'abc': 'def'})
 
+        on = js_to_json('{/*comment\n*/"abc"/*comment\n*/:/*comment\n*/"def"/*comment\n*/,/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'abc': 'def'})
+
         on = js_to_json('{ 0: /* " \n */ ",]" , }')
         self.assertEqual(json.loads(on), {'0': ',]'})
 
+        on = js_to_json('{ /*comment\n*/0/*comment\n*/: /* " \n */ ",]" , }')
+        self.assertEqual(json.loads(on), {'0': ',]'})
+
         on = js_to_json('{ 0: // comment\n1 }')
         self.assertEqual(json.loads(on), {'0': 1})
 
@@ -803,15 +815,27 @@ class TestUtil(unittest.TestCase):
         on = js_to_json("['a\\\nb']")
         self.assertEqual(json.loads(on), ['ab'])
 
+        on = js_to_json("/*comment\n*/[/*comment\n*/'a\\\nb'/*comment\n*/]/*comment\n*/")
+        self.assertEqual(json.loads(on), ['ab'])
+
         on = js_to_json('{0xff:0xff}')
         self.assertEqual(json.loads(on), {'255': 255})
 
+        on = js_to_json('{/*comment\n*/0xff/*comment\n*/:/*comment\n*/0xff/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'255': 255})
+
         on = js_to_json('{077:077}')
         self.assertEqual(json.loads(on), {'63': 63})
 
+        on = js_to_json('{/*comment\n*/077/*comment\n*/:/*comment\n*/077/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'63': 63})
+
         on = js_to_json('{42:42}')
         self.assertEqual(json.loads(on), {'42': 42})
 
+        on = js_to_json('{/*comment\n*/42/*comment\n*/:/*comment\n*/42/*comment\n*/}')
+        self.assertEqual(json.loads(on), {'42': 42})
+
     def test_extract_attributes(self):
         self.assertEqual(extract_attributes('<e x="y">'), {'x': 'y'})
         self.assertEqual(extract_attributes("<e x='y'>"), {'x': 'y'})
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 6c462625b..67a847eba 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -2103,6 +2103,13 @@ def strip_jsonp(code):
 
 
 def js_to_json(code):
+    COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*'
+    SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE)
+    INTEGER_TABLE = (
+        (r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16),
+        (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
+    )
+
     def fix_kv(m):
         v = m.group(0)
         if v in ('true', 'false', 'null'):
@@ -2118,11 +2125,6 @@ def js_to_json(code):
                 '\\x': '\\u00',
             }.get(m.group(0), m.group(0)), v[1:-1])
 
-        INTEGER_TABLE = (
-            (r'^(0[xX][0-9a-fA-F]+)\s*:?$', 16),
-            (r'^(0+[0-7]+)\s*:?$', 8),
-        )
-
         for regex, base in INTEGER_TABLE:
             im = re.match(regex, v)
             if im:
@@ -2134,11 +2136,11 @@ def js_to_json(code):
     return re.sub(r'''(?sx)
         "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
         '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
-        /\*.*?\*/|//[^\n]*|,(?=\s*[\]}])|
+        {comment}|,(?={skip}[\]}}])|
         [a-zA-Z_][.a-zA-Z_0-9]*|
-        \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?|
-        [0-9]+(?=\s*:)
-        ''', fix_kv, code)
+        \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|
+        [0-9]+(?={skip}:)
+        '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code)
 
 
 def qualities(quality_ids):