From 5c62bedd6b6287429bd49d440ef0a020b7d66235 Mon Sep 17 00:00:00 2001
From: uno20001 <regisztralo111@gmail.com>
Date: Fri, 24 Jan 2020 22:57:45 +0100
Subject: [PATCH] [utils.py:js_to_json] add support for octal escape sequences

---
 test/test_utils.py  |  3 +++
 youtube_dl/utils.py | 27 +++++++++++++++++++--------
 2 files changed, 22 insertions(+), 8 deletions(-)

diff --git a/test/test_utils.py b/test/test_utils.py
index 0896f4150..72157dab6 100644
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -918,6 +918,9 @@ class TestUtil(unittest.TestCase):
         inp = '''{segments: [{"offset":-3.885780586188048e-16,"duration":39.75000000000001}]}'''
         self.assertEqual(js_to_json(inp), '''{"segments": [{"offset":-3.885780586188048e-16,"duration":39.75000000000001}]}''')
 
+        inp = '''{label: "Fran\347ais"}'''
+        self.assertEqual(js_to_json(inp), '''{"label": "Fran\u00e7ais"}''')
+
     def test_js_to_json_edgecases(self):
         on = js_to_json("{abc_def:'1\\'\\\\2\\\\\\'3\"4'}")
         self.assertEqual(json.loads(on), {"abc_def": "1'\\2\\'3\"4"})
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index f6204692a..8ed94d923 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -3982,20 +3982,31 @@ def js_to_json(code):
         (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8),
     )
 
+    def convert_escapes(m):
+        # convert Javascript's octal escape sequences (and '\0')
+        # into valid JSON escape sequences (e.g. '\347' => '\u00e7', '\0' => '\u0000')
+        if m.group(1):
+            return "\\u%04x" % int(m.group(1), 8)
+
+        # convert the remaining escape sequences
+        # into valid JSON
+        return {
+            '"': '\\"',
+            "\\'": "'",
+            '\\\n': '',
+            '\\x': '\\u00',
+        }.get(m.group(0), m.group(0))
+
     def fix_kv(m):
         v = m.group(0)
+
         if v in ('true', 'false', 'null'):
             return v
         elif v.startswith('/*') or v.startswith('//') or v == ',':
             return ""
 
         if v[0] in ("'", '"'):
-            v = re.sub(r'(?s)\\.|"', lambda m: {
-                '"': '\\"',
-                "\\'": "'",
-                '\\\n': '',
-                '\\x': '\\u00',
-            }.get(m.group(0), m.group(0)), v[1:-1])
+            v = re.sub(r'(?s)\\(?:([0-7]{1,3})|.)|"', convert_escapes, v[1:-1])
 
         for regex, base in INTEGER_TABLE:
             im = re.match(regex, v)
@@ -4006,8 +4017,8 @@ def js_to_json(code):
         return '"%s"' % v
 
     return re.sub(r'''(?sx)
-        "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"|
-        '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'|
+        "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n01234567]))*[^"\\]*"|
+        '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n01234567]))*[^'\\]*'|
         {comment}|,(?={skip}[\]}}])|
         (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*|
         \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?|