[pitivi/ges: 204/287] Make quote_uri bulletproof and RFC 2396 compliant

[Jean-FranÃois Fortin Tam <jfft src gnome org>]

commit 298a23ae251d8a15053ba693c01e8924c7af3b4c
Author: Jean-FranÃois Fortin Tam <nekohayo gmail com>
Date:   Wed Jan 11 11:56:42 2012 -0500

    Make quote_uri bulletproof and RFC 2396 compliant
    
    This prevents duplicating clips by dnd in the Media Library,
    and also allows getting thumbnail hashes with perfect accuracy.

 pitivi/medialibrary.py |    7 +++++--
 pitivi/utils/misc.py   |   13 ++++++++++++-
 pitivi/utils/ui.py     |    1 +
 3 files changed, 18 insertions(+), 3 deletions(-)
---
diff --git a/pitivi/medialibrary.py b/pitivi/medialibrary.py
index 0fcb0f0..a8850c8 100644
--- a/pitivi/medialibrary.py
+++ b/pitivi/medialibrary.py
@@ -161,6 +161,10 @@ class MediaLibrary(Signallable, Loggable):
 
         The uri will be analyzed before being added.
         """
+        # Ensure we really have a correctly encoded URI according to RFC 2396.
+        # Otherwise, in some cases we'd get rogue characters that break
+        # searching for duplicates
+        uri = quote_uri(uri)
         if uri in self._sources:
             # uri is already added. Nothing to do.
             return
@@ -220,8 +224,7 @@ class MediaLibrary(Signallable, Loggable):
         """
         uri = info.get_uri()
         if self._sources.get(uri, None) is not None:
-            raise MediaLibraryError("We already have a info for this URI",
-                    uri)
+            raise MediaLibraryError("We already have info for this URI", uri)
         self._sources[uri] = info
         self._ordered_sources.append(info)
         self.nb_imported_files += 1
diff --git a/pitivi/utils/misc.py b/pitivi/utils/misc.py
index f367c12..dfb5c88 100644
--- a/pitivi/utils/misc.py
+++ b/pitivi/utils/misc.py
@@ -135,8 +135,19 @@ def get_filesystem_encoding():
 
 
 def quote_uri(uri):
+    """
+    Encode a URI according to RFC 2396, without touching the file:/// part.
+    """
     parts = list(urlsplit(uri, allow_fragments=False))
-    parts[2] = quote(parts[2])
+    # Make absolutely sure the string is unquoted before quoting again!
+    raw = unquote(parts[2])
+    # For computing thumbnail md5 hashes in the source list, we must adhere to
+    # RFC 2396. However, urllib's quote method only uses alphanumeric and "/"
+    # as their safe chars. We need to add both the reserved and unreserved chars
+    RFC_2396_RESERVED = ";/?:@&=+$,"
+    RFC_2396_UNRESERVED = "-_.!~*'()"
+    URIC_SAFE_CHARS = "/" + "%" + RFC_2396_RESERVED + RFC_2396_UNRESERVED
+    parts[2] = quote(raw, URIC_SAFE_CHARS)
     uri = urlunsplit(parts)
     return uri
 
diff --git a/pitivi/utils/ui.py b/pitivi/utils/ui.py
index 0de63c9..1510843 100644
--- a/pitivi/utils/ui.py
+++ b/pitivi/utils/ui.py
@@ -179,6 +179,7 @@ def beautify_info(info):
 
 
 def info_name(info):
+    """Return a human-readable filename (without the path and quoting)."""
     return escape(unquote(os.path.basename(info.get_uri())))