[utils] Fixup some common URL's typos in sanitize_url (closes #15649)

This commit is contained in:
Sergey M․ 2018-02-19 22:50:23 +07:00
parent 90830004c8
commit befa4708fd
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D
2 changed files with 22 additions and 3 deletions

View file

@ -57,6 +57,7 @@
read_batch_urls, read_batch_urls,
sanitize_filename, sanitize_filename,
sanitize_path, sanitize_path,
sanitize_url,
expand_path, expand_path,
prepend_extension, prepend_extension,
replace_extension, replace_extension,
@ -219,6 +220,12 @@ def test_sanitize_path(self):
self.assertEqual(sanitize_path('./abc'), 'abc') self.assertEqual(sanitize_path('./abc'), 'abc')
self.assertEqual(sanitize_path('./../abc'), '..\\abc') self.assertEqual(sanitize_path('./../abc'), '..\\abc')
def test_sanitize_url(self):
self.assertEqual(sanitize_url('//foo.bar'), 'http://foo.bar')
self.assertEqual(sanitize_url('httpss://foo.bar'), 'https://foo.bar')
self.assertEqual(sanitize_url('rmtps://foo.bar'), 'rtmps://foo.bar')
self.assertEqual(sanitize_url('https://foo.bar'), 'https://foo.bar')
def test_expand_path(self): def test_expand_path(self):
def env(var): def env(var):
return '%{0}%'.format(var) if sys.platform == 'win32' else '${0}'.format(var) return '%{0}%'.format(var) if sys.platform == 'win32' else '${0}'.format(var)

View file

@ -538,10 +538,22 @@ def sanitize_path(s):
return os.path.join(*sanitized_path) return os.path.join(*sanitized_path)
# Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of
# unwanted failures due to missing protocol
def sanitize_url(url): def sanitize_url(url):
return 'http:%s' % url if url.startswith('//') else url # Prepend protocol-less URLs with `http:` scheme in order to mitigate
# the number of unwanted failures due to missing protocol
if url.startswith('//'):
return 'http:%s' % url
# Fix some common typos seen so far
COMMON_TYPOS = (
# https://github.com/rg3/youtube-dl/issues/15649
(r'^httpss://', r'https://'),
# https://bx1.be/lives/direct-tv/
(r'^rmtp([es]?)://', r'rtmp\1://'),
)
for mistake, fixup in COMMON_TYPOS:
if re.match(mistake, url):
return re.sub(mistake, fixup, url)
return url
def sanitized_Request(url, *args, **kwargs): def sanitized_Request(url, *args, **kwargs):