http://code.google.com/p/googlecl/issues/detail?id=473 contains a
patch that does some more sane checking on the contents of an image
file. Without this patch, I can not upload files that have the
extension in upper case, like .JPG. Renaming to .jpg works, but this
solutions seems a little ugly.
The proposed patch checks the mime-type of the file by reading the
contents, using the standard Python mimetypes library. The result
should be much more reliable than checking the extension of the
Could someone please review the patch and merge it if suitable?