Erik Max Francis <[EMAIL PROTECTED]> writes: > From looking at the commands, two other features occurred to me: > > - A pipe-headers command, which instead of piping the entire > contents of a message, simply pipes the headers. This can be > helpful when you want to do header processing with an external > program, but don't care about the (sometimes enormous) body.
Good suggestion, I've made note of it. I don't think many users of the current 'pipe' actually care about the body anyway. It might also help users who like to 'pipe' to subprograms that cause SIGPIPE, which in turn triggers an exception/deferred delivery in TMDA, which in turn leads to a post to this list. :-) > - Along the same lines, lazy body processing. If the body of a > message is never needed by any filter rule, it is simply not > processed. This is pretty easy to do with the current Python mail > classes -- which only use the body if you tell them to -- and is an > obvious optimization so it may already be in place. We've discussed this before I believe, but I can't find the thread. It's not possible currently, but I think it would make sense for the future. The body is processed for a few other general things such as determining the size of the message to decide whether or not to attach it to the confirmation request. So, it would be a combination of working through these cases, and then some code reorganization. _____________________________________________ tmda-users mailing list ([EMAIL PROTECTED]) http://tmda.net/lists/listinfo/tmda-users
