Minor filename encoding improvement in a common case

2 files changed, 10 insertions, 0 deletions

diff --git a/test/test_utils.py b/test/test_utils.py
index 4208ee653..e806a10c5 100644
--- a/test/test_utils.py
+++ b/test/test_utils.py
@@ -61,6 +61,13 @@ class TestUtil(unittest.TestCase):
 			for fbc in forbidden:
 				self.assertTrue(fbc not in sanitize_filename(fc, restricted=True))
 
+		# Handle a common case more neatly
+		self.assertEqual(sanitize_filename(u'大声带 - Song', restricted=True), u'Song')
+		self.assertEqual(sanitize_filename(u'总统: Speech', restricted=True), u'Speech')
+		# .. but make sure the file name is never empty
+		self.assertTrue(sanitize_filename(u'-', restricted=True) != u'')
+		self.assertTrue(sanitize_filename(u':', restricted=True) != u'')
+
 	def test_ordered_set(self):
 		self.assertEqual(orderedSet([1,1,2,3,4,4,5,6,7,3,5]), [1,2,3,4,5,6,7])
 		self.assertEqual(orderedSet([]), [])
diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 3339f56ec..4ace22c2f 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -218,6 +218,9 @@ def sanitize_filename(s, restricted=False):
 	while '__' in result:
 		result = result.replace('__', '_')
 	result = result.strip('_')
+	# Common case of "Foreign band name - English song title"
+	if restricted and result.startswith('-_'):
+		result = result[2:]
 	if not result:
 		result = '_'
 	return result