From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7936 invoked from network); 10 Dec 1998 23:42:08 -0000 Received: from ns2.primenet.com.au (HELO primenet.com.au) (7795@203.24.36.3) by ns1.primenet.com.au with SMTP; 10 Dec 1998 23:42:08 -0000 Received: (qmail 11736 invoked from network); 10 Dec 1998 09:26:40 -0000 Received: from math.gatech.edu (list@130.207.146.50) by ns2.primenet.com.au with SMTP; 10 Dec 1998 09:26:40 -0000 Received: (from list@localhost) by math.gatech.edu (8.9.1/8.9.1) id EAA17929; Thu, 10 Dec 1998 04:25:27 -0500 (EST) Resent-Date: Thu, 10 Dec 1998 04:25:27 -0500 (EST) Date: Thu, 10 Dec 1998 10:22:18 +0100 (MET) Message-Id: <199812100922.KAA03277@beta.informatik.hu-berlin.de> From: Sven Wischnowsky To: zsh-workers@math.gatech.edu In-reply-to: Sven Wischnowsky's message of Wed, 9 Dec 1998 15:44:36 +0100 (MET) Subject: Re: PATCH: wrapper functions in modules Resent-Message-ID: <"3_2wr3.0.4O4.7EvRs"@math> Resent-From: zsh-workers@math.gatech.edu X-Mailing-List: archive/latest/4740 X-Loop: zsh-workers@math.gatech.edu Precedence: list Resent-Sender: zsh-workers-request@math.gatech.edu I wrote: > ... [patch for adding wrapper functions in modules] I forgot to mention it: If a module installs such wrapper functions and there was a function called which caused the `before'-function to be executed and the module also has a `after'-function which needs to be called if the `before'-function was called we might have problems, since the shell function may unload the module so that the `after' function will not be called. So the question is: should we add a convention that the module can export a function that will be called when trying to unload a module and which says if the module can currently be unloaded? Alternatively I could change the code so that a module cannot be unloaded if it has defined `after'-functions and there are currently shell functions being executed. I think we could even change the code so that unloading a module automatically removes builtins, conditions, and wrapper functions defined by it... And another question: why is the variable containing the paths to module-directories named `module_path'? Comparing this with `fpath' I would have expected `mpath' (and `MPATH') or something like that. Bye Sven -- Sven Wischnowsky wischnow@informatik.hu-berlin.de