From 9d7ded6419089c1bf252496073f73ad90ed71004 Mon Sep 17 00:00:00 2001 From: Awal Garg Date: Sun, 8 Oct 2023 01:57:23 +0200 Subject: [PATCH] [utils] `js_to_json`: Fix `Date` constructor parsing (#8295) Authored by: awalgarg, Grub4K --- test/test_utils.py | 7 ++++++- yt_dlp/utils/_utils.py | 2 +- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/test/test_utils.py b/test/test_utils.py index fd612ff86..77040f29c 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -1209,6 +1209,9 @@ class TestUtil(unittest.TestCase): on = js_to_json('\'"\\""\'') self.assertEqual(json.loads(on), '"""', msg='Unnecessary quote escape should be escaped') + on = js_to_json('[new Date("spam"), \'("eggs")\']') + self.assertEqual(json.loads(on), ['spam', '("eggs")'], msg='Date regex should match a single string') + def test_js_to_json_malformed(self): self.assertEqual(js_to_json('42a1'), '42"a1"') self.assertEqual(js_to_json('42a-1'), '42"a"-1') @@ -1220,11 +1223,13 @@ class TestUtil(unittest.TestCase): self.assertEqual(js_to_json('`${name}"${name}"`', {'name': '5'}), '"5\\"5\\""') self.assertEqual(js_to_json('`${name}`', {}), '"name"') - def test_js_to_json_map_array_constructors(self): + def test_js_to_json_common_constructors(self): self.assertEqual(json.loads(js_to_json('new Map([["a", 5]])')), {'a': 5}) self.assertEqual(json.loads(js_to_json('Array(5, 10)')), [5, 10]) self.assertEqual(json.loads(js_to_json('new Array(15,5)')), [15, 5]) self.assertEqual(json.loads(js_to_json('new Map([Array(5, 10),new Array(15,5)])')), {'5': 10, '15': 5}) + self.assertEqual(json.loads(js_to_json('new Date("123")')), "123") + self.assertEqual(json.loads(js_to_json('new Date(\'2023-10-19\')')), "2023-10-19") def test_extract_attributes(self): self.assertEqual(extract_attributes(''), {'x': 'y'}) diff --git a/yt_dlp/utils/_utils.py b/yt_dlp/utils/_utils.py index ba6242380..3dc17bf59 100644 --- a/yt_dlp/utils/_utils.py +++ b/yt_dlp/utils/_utils.py @@ -2744,7 +2744,7 @@ def js_to_json(code, vars={}, *, strict=False): code = re.sub(r'(?:new\s+)?Array\((.*?)\)', r'[\g<1>]', code) code = re.sub(r'new Map\((\[.*?\])?\)', create_map, code) if not strict: - code = re.sub(r'new Date\((".+")\)', r'\g<1>', code) + code = re.sub(rf'new Date\(({STRING_RE})\)', r'\g<1>', code) code = re.sub(r'new \w+\((.*?)\)', lambda m: json.dumps(m.group(0)), code) code = re.sub(r'parseInt\([^\d]+(\d+)[^\d]+\)', r'\1', code) code = re.sub(r'\(function\([^)]*\)\s*\{[^}]*\}\s*\)\s*\(\s*(["\'][^)]*["\'])\s*\)', r'\1', code)