youtube-dl

commit befa4708fd2165b85d04002c3845adf191d34302
parent 90830004c893e2d5f0643c05af064cfc7a3b579e
Author: Sergey M․ <dstftw@gmail.com>
Date:   Mon, 19 Feb 2018 22:50:23 +0700

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

Diffstat:
M
test/test_utils.py
 | 
7
+++++++
M
youtube_dl/utils.py
 | 
18
+++++++++++++++---

2 files changed, 22 insertions(+), 3 deletions(-)
diff --git a/test/test_utils.py b/test/test_utils.py
@@ -57,6 +57,7 @@ from youtube_dl.utils import (
     read_batch_urls,
     sanitize_filename,
     sanitize_path,
+    sanitize_url,
     expand_path,
     prepend_extension,
     replace_extension,
@@ -219,6 +220,12 @@ class TestUtil(unittest.TestCase):
         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 env(var):
             return '%{0}%'.format(var) if sys.platform == 'win32' else '${0}'.format(var)
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
@@ -538,10 +538,22 @@ def sanitize_path(s):
     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):
-    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):