From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24948 invoked from network); 13 Mar 2000 16:52:25 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 13 Mar 2000 16:52:25 -0000 Received: (qmail 29485 invoked by alias); 13 Mar 2000 16:52:14 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 10121 Received: (qmail 29460 invoked from network); 13 Mar 2000 16:52:11 -0000 From: "Bart Schaefer" Message-Id: <1000313165200.ZM2467@candle.brasslantern.com> Date: Mon, 13 Mar 2000 16:52:00 +0000 In-Reply-To: <200003131042.LAA17110@beta.informatik.hu-berlin.de> Comments: In reply to Sven Wischnowsky "Re: Default fpath" (Mar 13, 11:42am) References: <200003131042.LAA17110@beta.informatik.hu-berlin.de> X-Mailer: Z-Mail (5.0.0 30July97) To: Sven Wischnowsky , zsh-workers@sunsite.auc.dk Subject: Re: Default fpath MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii On Mar 13, 11:42am, Sven Wischnowsky wrote: } Subject: Re: Default fpath } } Bart Schaefer wrote: } } > Note 1 -- While we're on the topic of zcompile, it could use: } > (1) a way to append to an existing .zwc, rather like `ar' works, and } } That's a bit problematic because of the little/big-endian thing. I.e. we } can't just append to a wordcode file (and I don't want to have } multiple headers). I was afraid of that. } But of course we could make it read/change/re-write } such files, also allowing deletion of functions. Is it worth it? I } mean, creating wordcode files is quite fast for me... Perhaps not. Actually, what would suit me fine is a way to dump out a .zwc of (a subset of the) functions defined in a running zsh. } > (2) an option to say whether each function should be autoloaded zsh-style } > or ksh-style, so that the right thing happens regardless of the run- } > time setting of kshautoload.) } } But currently it is the same as for loading from a normal file, } wouldn't it probably be confusing if the wordcode file said how to } load it? Actually I'd like it if the text file could "say how to load it" as well, but that wouldn't be portable -- the whole point of kshautoload is to be able to write functions that can load into both zsh and ksh. So if I set kshautoload, I'm probably in a situation where I load ksh functions. It would be possible to require that zsh function text files include some token that specifies NOT using kshautoload, but that would put the burden on the ordinary zsh writer, which feels wrong. There's no such problem with .zwc files -- only zsh will load them, and the how-to token is invisible to the user. Consider the current problem of initializing the completion system with kshautoload set. (I.e., you can't.) The kshautoload user could solve this problem by precompiling the completion system for zsh loading. The one thing I can't decide is whether the default should be initialized from the setting of kshautoload, or if it should always default to zsh. } But don't get me wrong: I never use kshautoload, so I } wouldn't be against allowing it. Per-function or per-wordcode file? Per source file used to build the wordcode file would be the correct granularity, which I guess means per-function? -- Bart Schaefer Brass Lantern Enterprises http://www.well.com/user/barts http://www.brasslantern.com