Merge branch 'master' of github.com:rg3/youtube-dl

[youtube-dl] / youtube_dl / compat.py

diff --git a/youtube_dl/compat.py b/youtube_dl/compat.py
index 0243949a44468582a55fb151ec40ccf3a6850e9a..b8aaf5a461c9e3ca2884c748ebb3225a2fd9fe29 100644 (file)
--- a/youtube_dl/compat.py
+++ b/youtube_dl/compat.py
@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import binascii
@@ -64,8 +65,8 @@ except ImportError:  # Python 2
     import htmlentitydefs as compat_html_entities
 
 try:  # Python >= 3.3
-    from compat_html_entities import html as compat_html_entities_html5
-except ImportError:
+    compat_html_entities_html5 = compat_html_entities.html5
+except AttributeError:
     # Copied from CPython 3.5.1 html/entities.py
     compat_html_entities_html5 = {
         'Aacute': '\xc1',
@@ -2594,15 +2595,19 @@ except ImportError:  # Python < 3.3
             return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
-if sys.version_info >= (2, 7, 3):
+try:
+    args = shlex.split('中文')
+    assert (isinstance(args, list) and
+            isinstance(args[0], compat_str) and
+            args[0] == '中文')
     compat_shlex_split = shlex.split
-else:
+except (AssertionError, UnicodeEncodeError):
     # Working around shlex issue with unicode strings on some python 2
     # versions (see http://bugs.python.org/issue1548891)
     def compat_shlex_split(s, comments=False, posix=True):
         if isinstance(s, compat_str):
             s = s.encode('utf-8')
-        return shlex.split(s, comments, posix)
+        return list(map(lambda s: s.decode('utf-8'), shlex.split(s, comments, posix)))
 
 
 def compat_ord(c):