[commonmistakes] Detect BOMs at the beginning of URLs

Reported at https://bugzilla.redhat.com/show_bug.cgi?id=1093517 .

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 55ca0d6e4667928ad062ffbc3b26a19a7f819ccd..fb1e7f325c5c9e781bae5b3136fa71e4db50b92e 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -74,7 +74,7 @@ from .collegehumor import CollegeHumorIE
 from .collegerama import CollegeRamaIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .comcarcoff import ComCarCoffIE
 from .collegerama import CollegeRamaIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .comcarcoff import ComCarCoffIE

-from .commonmistakes import CommonMistakesIE
+from .commonmistakes import CommonMistakesIE, UnicodeBOMIE

 from .condenast import CondeNastIE
 from .cracked import CrackedIE
 from .criterion import CriterionIE
 from .condenast import CondeNastIE
 from .cracked import CrackedIE
 from .criterion import CriterionIE

diff --git a/youtube_dl/extractor/commonmistakes.py b/youtube_dl/extractor/commonmistakes.py
index dbbf27a7486ab78a239df5a21a761b911a1e4a6e..2f86e2381f447faa7b42d6056e685bc04f101f9c 100644 (file)
--- a/youtube_dl/extractor/commonmistakes.py
+++ b/youtube_dl/extractor/commonmistakes.py
@@ -27,3 +27,20 @@ class CommonMistakesIE(InfoExtractor):
         if not self._downloader.params.get('verbose'):
             msg += ' Add -v to the command line to see what arguments and configuration youtube-dl got.'
         raise ExtractorError(msg, expected=True)
         if not self._downloader.params.get('verbose'):
             msg += ' Add -v to the command line to see what arguments and configuration youtube-dl got.'
         raise ExtractorError(msg, expected=True)

+
+
+class UnicodeBOMIE(InfoExtractor):
+        IE_DESC = False
+        _VALID_URL = r'(?P<bom>\ufeff)(?P<id>.*)$'
+
+        _TESTS = [{
+            'url': '\ufeffhttp://www.youtube.com/watch?v=BaW_jenozKc',
+            'only_matching': True,
+        }]
+
+        def _real_extract(self, url):
+            real_url = self._match_id(url)
+            self.report_warning(
+                'Your URL starts with a Byte Order Mark (BOM). '
+                'Removing the BOM and looking for "%s" ...' % real_url)
+            return self.url_result(real_url)