X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Futils.py;h=90e0ed9ab7eb36d4c9f30efffdd2868e9cfe6d47;hb=70a1165b32acf253905109e9b4f245295d67af1f;hp=0f49d602eab1296dced15449a949c26dc4408d66;hpb=a2aaf4dbc6e5f5d345329b5a845111851453b6a6;p=youtube-dl

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 0f49d602e..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:
@@ -252,15 +252,12 @@ def sanitize_open(filename, open_mode):
             raise
 
         # In case of error, try to remove win32 forbidden chars
-        alt_filename = os.path.join(
-            re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part)
-            for path_part in os.path.split(filename)
-        )
+        alt_filename = sanitize_path(filename)
         if alt_filename == filename:
             raise
         else:
             # An exception here should be caught in the caller
-            stream = open(encodeFilename(filename), open_mode)
+            stream = open(encodeFilename(alt_filename), open_mode)
             return (stream, alt_filename)
 
 
@@ -322,13 +319,20 @@ def sanitize_path(s):
     if unc_or_drive:
         norm_path.pop(0)
     sanitized_path = [
-        re.sub('[/<>:"\\|\\\\?\\*]', '#', path_part)
+        path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|\.$)', '#', path_part)
         for path_part in norm_path]
     if unc_or_drive:
         sanitized_path.insert(0, unc_or_drive + os.path.sep)
     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 = []
@@ -344,7 +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'):