[utils] Recognize more patterns in strip_jsonp()

Used in Youku Show pages
This commit is contained in:
Yen Chi Hsuan 2017-05-26 21:58:18 +08:00
parent 59ed87cbd9
commit 5552c9eb0f
No known key found for this signature in database
GPG key ID: 7F902A182457CA23
3 changed files with 15 additions and 1 deletions

View file

@ -1,6 +1,7 @@
version <unreleased>
Core
+ [utils] strip_jsonp() can recognize more patterns
* [postprocessor/ffmpeg] Fix metadata filename handling on Python 2 (#13182)
Extractors

View file

@ -678,6 +678,14 @@ def test_strip_jsonp(self):
d = json.loads(stripped)
self.assertEqual(d, {'status': 'success'})
stripped = strip_jsonp('window.cb && window.cb({"status": "success"});')
d = json.loads(stripped)
self.assertEqual(d, {'status': 'success'})
stripped = strip_jsonp('window.cb && cb({"status": "success"});')
d = json.loads(stripped)
self.assertEqual(d, {'status': 'success'})
def test_uppercase_escape(self):
self.assertEqual(uppercase_escape(''), '')
self.assertEqual(uppercase_escape('\\U0001d550'), '𝕐')

View file

@ -2211,7 +2211,12 @@ def parse_age_limit(s):
def strip_jsonp(code):
return re.sub(
r'(?s)^[a-zA-Z0-9_.$]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code)
r'''(?sx)^
(?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]+)
(?:\s*&&\s*(?P=func_name))?
\s*\(\s*(?P<callback_data>.*)\);?
\s*?(?://[^\n]*)*$''',
r'\g<callback_data>', code)
def js_to_json(code):