From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 5535 invoked from network); 12 May 2000 08:31:51 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 12 May 2000 08:31:51 -0000 Received: (qmail 25392 invoked by alias); 12 May 2000 08:31:37 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 11340 Received: (qmail 25363 invoked from network); 12 May 2000 08:31:35 -0000 From: "Bart Schaefer" Message-Id: <1000512083115.ZM5065@candle.brasslantern.com> Date: Fri, 12 May 2000 08:31:14 +0000 X-Mailer: Z-Mail (5.0.0 30July97) To: zsh-workers@sunsite.auc.dk Subject: Test (re)numbering MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii I'd like to propose we renumber to the following numbering scheme: 00 : cd (the example test) 01-29 : Shell syntax and other features usable in scripts (no modules!) 30 : zmodload 31-49 : As 01-29, but for features loaded from modules 50 : zpty 51- : Interactive-only features (e.g. completion) This at least approximately follows a logical progression where most things aren't attempted until other tests have been run on which they might depend. This would let us specify that certain test failures (e.g. 30 and 50) abort all tests that come after them, on the grounds that the remaining tests can't possibly succeed. Of course, the whole test system is likely to fail if anything in 00-30 fails, but short of rewriting ztst.zsh in Bourne I don't think we can do anything about that. I could easily be convinced that we should reserve more than 29 numbers for basic shell tests, but I don't think fewer is wise. Of course, the numbers are presently ordered lexically, so 100-299, 1000-2999, etc. are available if we need to shove more tests in "below" 30. Which, if we start writing tests for lots of combinations of setopts, we might need ... Under the above scheme, we'd invent 30zmodload and 50zpty, and rename: 50cd.ztst --> 00cd.ztst 51xtrace.ztst --> 29xtrace.ztst 52zregexparse.ztst --> 31zregexparse.ztst and we'd also document the numbering scheme in a new Test/README or in 00cd.ztst or both. Aside: I wonder why zregexparse doesn't deserve it's own module ... -- Bart Schaefer Brass Lantern Enterprises http://www.well.com/user/barts http://www.brasslantern.com