From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14172 invoked from network); 9 Mar 1999 08:03:29 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 9 Mar 1999 08:03:29 -0000 Received: (qmail 21853 invoked by alias); 9 Mar 1999 08:02:59 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 5693 Received: (qmail 21846 invoked from network); 9 Mar 1999 08:02:58 -0000 Date: Tue, 9 Mar 1999 09:02:08 +0100 (MET) Message-Id: <199903090802.JAA12276@beta.informatik.hu-berlin.de> From: Sven Wischnowsky To: zsh-workers@sunsite.auc.dk In-reply-to: Peter Stephenson's message of Mon, 08 Mar 1999 18:54:05 +0100 Subject: Re: Completion TODO (was: pws-11) Peter Stephenson wrote: > Sven Wischnowsky wrote: > > The current state: condition codes, some of them modifying. Saving and > > restoring of the special parameters is done automatically on function > > entry and exit (and nowhere else). > > The problems are: modifying condition codes are considered a bad thing > > and saving/restoring has to be done explicitly. > > I haven't followed what's wrong with having the parameters restored > automatically unless compstate[restore] is altered. I was alluding to the restoring inside a function when one has multiple tests. > If the modification of the parameters is simply to be removed from [[-style > tests, then I still think a builtin is the best way of handling it. > `compset -i 3' ignores three more characters from $PREFIX, or > whatever. This could allow tests with side effects, like `compset -I > ' to ignore a pattern at the head of $PREFIX if it matches, > or whatever. Then you also could have e.g. `compset -n 2 -1', narrow > to words 2 to -1 (same as shifting words and decrementing CURRENT), > `compset -N "start_pat" "optional_end_pat"', narrow to the range of > words between the patterns with the end pattern optional. If it's > much more complicated than that, I can't imagine I would use it much. I had forgotten to include this in the list of suggestions (although, I think, it was the first suggestion ever made). The builtin should have a return value usable for `if compset ...; then' as a `side effect', right? > I'm not really enthusiastic about more subscripting hieroglyphics. ;-) Bye Sven -- Sven Wischnowsky wischnow@informatik.hu-berlin.de