X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=90e0ed9ab7eb36d4c9f30efffdd2868e9cfe6d47;hb=70a1165b32acf253905109e9b4f245295d67af1f;hp=e82e3998a7c30d8ae14c9561b51946df0cbebcff;hpb=dc03a42537cba83597ca8acb2bbe03f686f2136c;p=youtube-dl

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index e82e3998a..90e0ed9ab 100644
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -75,7 +75,7 @@ def preferredencoding():
     try:
         pref = locale.getpreferredencoding()
         'TEST'.encode(pref)
-    except:
+    except Exception:
         pref = 'UTF-8'
 
     return pref
@@ -127,7 +127,7 @@ def write_json_file(obj, fn):
             except OSError:
                 pass
         os.rename(tf.name, fn)
-    except:
+    except Exception:
         try:
             os.remove(tf.name)
         except OSError:
@@ -326,6 +326,13 @@ def sanitize_path(s):
     return os.path.join(*sanitized_path)
 
 
+def sanitize_url_path_consecutive_slashes(url):
+    """Collapses consecutive slashes in URLs' path"""
+    parsed_url = list(compat_urlparse.urlparse(url))
+    parsed_url[2] = re.sub(r'/{2,}', '/', parsed_url[2])
+    return compat_urlparse.urlunparse(parsed_url)
+
+
 def orderedSet(iterable):
     """ Remove all duplicates from the input iterable """
     res = []
@@ -341,7 +348,7 @@ def _htmlentity_transform(entity):
     if entity in compat_html_entities.name2codepoint:
         return compat_chr(compat_html_entities.name2codepoint[entity])
 
-    mobj = re.match(r'#(x?[0-9]+)', entity)
+    mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity)
     if mobj is not None:
         numstr = mobj.group(1)
         if numstr.startswith('x'):
@@ -1804,18 +1811,3 @@ class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
             return None  # No Proxy
         return compat_urllib_request.ProxyHandler.proxy_open(
             self, req, proxy, type)
-
-
-def url_sanitize_consecutive_slashes(url):
-    """Sanitize URLs with consecutive slashes
-
-    For example, transform both
-        http://hostname/foo//bar/filename.html
-    and
-        http://hostname//foo/bar/filename.html
-    into
-        http://hostname/foo/bar/filename.html
-    """
-    parsed_url = list(compat_urlparse.urlparse(url))
-    parsed_url[2] = re.sub(r'/{2,}', '/', parsed_url[2])
-    return compat_urlparse.urlunparse(parsed_url)