From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 596 invoked from network); 8 Mar 1999 10:13:05 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 8 Mar 1999 10:13:05 -0000 Received: (qmail 28831 invoked by alias); 8 Mar 1999 10:12:48 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 5678 Received: (qmail 28819 invoked from network); 8 Mar 1999 10:12:46 -0000 Date: Mon, 8 Mar 1999 11:11:47 +0100 (MET) Message-Id: <199903081011.LAA11500@beta.informatik.hu-berlin.de> From: Sven Wischnowsky To: zsh-workers@sunsite.auc.dk Subject: Comments on new style completion Thinking some more about Andrej's suggestions for `_path_files' I thought about making this user-configurable and then was reminded about Peter's comment about passing options to the completion code. Maybe we should add a (or more than one) associative array(s), nicely putting everything into one basket, e.g.: compconfig[dump_file] what is now COMPDUMP compconfig[correct] COMPCORRECT compconfig[correct_prompt] CCPROMPT compconfig[correct_orig] CCORIG (may be extensible to: show the original string as the first/last `match') Then we could have keys for different `styles' that should be used by various functions, e.g.: compconfig[path_style] say how `_path_files' should behave Here I would prefer `compstyle[paths]=...'. And since I already mentiond the correction stuff: should we change this so that the maximum number of errors allowed is less than or equal to the number of characters in the word to complete (or that plus one)? Bye Sven -- Sven Wischnowsky wischnow@informatik.hu-berlin.de