Hi all, Robby Griffin wrote:
I'll come up with a patch if anyone wants it, but seriously, YMMV, a lot.
I think I should commit one of my unpublishd patch on SourceForge CVS. @@ -356,8 +396,16 @@ # e.g. image/jpg (should be image/jpeg). For now we just store such # things as application/octet-streams since that seems the safest. ctype = msg.get_content_type() - fnext = os.path.splitext(msg.get_filename(''))[1] - ext = guess_extension(ctype, fnext) + # i18n file name is encoded + lcset = Utils.GetCharSet(mlist.preferred_language) + filename = Utils.oneline(msg.get_filename(''), lcset) + fnext = os.path.splitext(filename)[1] + # For safety, we should confirm this is valid ext for content-type + # but we can use fnext if we introduce fnext filtering + if mm_cfg.SCRUBBER_USE_ATTACHMENT_FILENAME_EXTENSION: + ext = fnext + else: + ext = guess_extension(ctype, fnext) if not ext: # We don't know what it is, so assume it's just a shapeless # application/octet-stream, unless the Content-Type: is With this patch, site manager can set SCRUBBER_USE_ATTACHMENT_FILENAME_EXTENSION = True in mm_cfg.py to use attachment filename extension as is specified by the original attachment. I also want to merge patch id 1027882 so that really dangerous files can be trapped by MimeDel.py. -- Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp http://weather.is.kochi-u.ac.jp/