From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1745 invoked from network); 14 Oct 2001 19:04:18 -0000 Received: from unknown (HELO sunsite.dk) (130.225.247.90) by ns1.primenet.com.au with SMTP; 14 Oct 2001 19:04:18 -0000 Received: (qmail 13367 invoked by alias); 14 Oct 2001 19:03:20 -0000 Mailing-List: contact zsh-workers-help@sunsite.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 16037 Received: (qmail 13351 invoked from network); 14 Oct 2001 19:03:18 -0000 To: zsh-workers@sunsite.auc.dk (Zsh hackers list) Subject: Re: differences between the two branches In-reply-to: "Bart Schaefer"'s message of "Sat, 13 Oct 2001 07:39:37 -0000." <1011013073937.ZM368@candle.brasslantern.com> Date: Sat, 14 Oct 2000 20:02:48 +0100 From: Peter Stephenson Message-Id: <20001014190253.4AC4E14283@pwstephenson.fsnet.co.uk> Bart Schaefer wrote: > } Hmm, a bit of perl could probably help here. Thanks. I've just been looking through this, except for a few which went way before 4.0.2 which I haven't bothered to look at again (some in the list below still predate 4.0.2). This obviously dupliates Oliver's selection to a certain extent, but I'm a sucker for a perl script. Here's a list of things which could go, if it's not too much work, and those that probably shouldn't. I will not be moving other people's patches to 4.0 myself, now or ever, because I'm too lazy and too disorganised. So decide based on the following, or otherwise. For the same reason, there's a middle category between `could go' and `probably shouldn't go' called `my brain hurts' (you have to imagine the knotted handkerchief, tank top and ape-like posture). Some of the ones in the last category are simply there because they're new and would actually be entirely harmless to tranfer. Could go if not too much work ============================ > - * unposted: Completion/compinstall: spotted by Bart: assignments > - when finding $fpath if not set in current shell were completely > - garbled. That's me. That should go across, but I'll have to do it by hand. > only in current/ChangeLog: 16xxx > only in branch/ChangeLog: 16018 Oliver: I think these are probably the same, so it's probably gone in anyway. It fixes compilation for Darwin, which would certainly be a good idea. > only in current/ChangeLog: 15812 Bart: doc changes for parameters. Might make sense to copy these over. > only in current/ChangeLog: 15739 > only in current/ChangeLog: 15740 Clint: compaudit, probably OK, maybe even necessary on Debian. > only in current/ChangeLog: 15721 Me: LICENCE: Oliver pointed this out and it's now in. > only in current/ChangeLog: 15713 Wayne: _rpm: probably OK. > only in current/ChangeLog: 15702 Bart: Doc improvements, probably could go across. > only in current/ChangeLog: 15676 Bart: minor test tweaks. Might be a good idea since they remove some unnecessary failures. > only in current/ChangeLog: 15653 Sven: Restore traditional insert-tab behaviour. This was certainly a bug in that it disagreed with the doc, so probably ought to be in 4.0. > only in current/ChangeLog: 15638 Sven: correction with matchers. Definitely a bug, so if this is in 4.0 should go across. > only in current/ChangeLog: 15596 Wayne: silence compiler, fine if applicable. > only in current/ChangeLog: 15583 Me: installfns.sh installed CVS stuff, may need to go across. > only in current/ChangeLog: 15508 Me: _cvs tags didn't work with one command, should go across if not already there. > only in current/ChangeLog: 15488 Bart: doc tweaks, probably reasonable to apply these. > only in current/ChangeLog: 15375 Bart: explain setopt CORRECT, could probably go across. > only in current/ChangeLog: 15354 Bart: Doc indexing, useful but I don't know how hard this is to transfer. > only in current/ChangeLog: 15279 Me: expand an expansion description, probably OK to push back. > only in current/ChangeLog: 15181 Andrej: kpic -> KPIC; I thought this *was* for 4.0.2? > only in current/ChangeLog: 15180 Andrej: fix AIX configuration, should almost certainly go across if it's been tested (does sourceforge have AIX?) > only in current/ChangeLog: 15099 Andrej: test suite handles `unimplemented', be nice to have this in 4.0 but may well need some work. > only in current/ChangeLog: 15057 Bart: PRINT_EXIT_VALUE: didn't work, so probably this ought to go back. > only in current/ChangeLog: 15023, 15027 > only in current/ChangeLog: 15038 Bart: fix POSIX `for', looks OK to copy. Andrej: test POSIX for syntax, depends on fix. My brain hurts just thinking about it ===================================== > only in current/ChangeLog: 16002 Sven: Fix expand style with prefix: probably OK, but I don't know if it's relevant or easy to port. > only in current/ChangeLog: 15949 > only in current/ChangeLog: 15994 Me: list-suffixes and other stuff in compinstall. I may simply copy the whole of compinstall across (see above). > only in current/ChangeLog: 15944 Sven: `suffix' in expand style always respected. Arguably a bug fix, but arguably not. > only in current/ChangeLog: 15836 > only in current/ChangeLog: 15806 > only in current/ChangeLog: 15809 Wayne: `r' problem reported by Bart. Dunno if this was in 4.0. > only in current/ChangeLog: 15650 > only in current/ChangeLog: 15669 Sven: approximation and ~ problems. Dunno, it's too hard to work out the consequences. > only in current/ChangeLog: 15615 Sven: complicated matcher thing it would take me half an hour to understand. > only in current/ChangeLog: 15584 Sven: zwc problem, dunno if this is relevant. > only in current/ChangeLog: 15498 Me: Consequent on moving zle builtins to zshzle. Must check if that was already in 4.0. > only in current/ChangeLog: 15198 Bart: _tilde return value, OK if applicable, but may well not be. > only in current/ChangeLog: 15094 Clint: his prompt theme, dunno. > only in current/ChangeLog: 15060 Bart: A02alias: can't keep track of whether this applies or not. > only in current/ChangeLog: 15050, 15054 Bart: POSIX exit statuses. Change of behaviour, so not sure. Probably shouldn't go ===================== > only in current/ChangeLog: users/4157 (plus workers/15674) Bart: upgrade to _man, probably too much out of sync with 4.0 by now. 15674 looks like a bug fix in any case. > only in current/ChangeLog: unposted; based on 14679 (me) and 14693 (Bart) Andrej: autoconf 2.50 fixes. Not necessary; for consistency, we're sticking with the the old version for 4.0. > +2001-09-21 Andrew Main (Zefram) > + > + * Doc/Zsh/contrib.yo, Functions/Misc/tetris: Tetris game for ZLE. > + Not necessary. > only in current/ChangeLog: 15608 > only in current/ChangeLog: Bart: 15106, small changes in 15152 > only in current/ChangeLog: 15945 Bart: improves _cd. Sven: improve _cd. I can't actually work out what the combined effect is, but there are almost certainly too many _cd, _tilde etc. changes. > only in current/ChangeLog: Back out 15266 Me: That didn't go in anyway. > only in current/ChangeLog: 16006 Clint: _mailboxes. New style, so doesn't need to be transferred. I'd prefer all new behaviour like that, unless strictly necessary for compatibility, to appear in upgrades rather than bug fixes. > only in current/ChangeLog: 16005 Clint: mount: there's a lot of this, let's not get into this on 4.0, it may still be dragging on. > only in current/ChangeLog: 16000 Clint: new _debchange, not necessary but harmless. > only in current/ChangeLog: 15995 Sven: show directories with globbed files by default. I don't think we should change this in mid-4.0. > only in current/ChangeLog: 15991 Sven: file completion display. I'm not worried about this, myself. > only in current/ChangeLog: 15987 Oliver: tcp.c, irrelevant. > only in current/ChangeLog: 15980 Me: zcalc, irrelevant. > only in branch/ChangeLog: 15974 > only in current/ChangeLog: 15931 > only in current/ChangeLog: 15917 > only in current/ChangeLog: 15844 Me: 15974 only on branch because it documents the TRAPEXIT bug which is fixed in 4.1 by 15844, 15917, 15931, but not necessarily stable, so I'll leave this. > only in current/ChangeLog: 15973 Oliver: printf: irrelevant > only in current/ChangeLog: 15964 Andrej: tcp.c, irrelevant. > only in current/ChangeLog: 15946 Sven: compresult.c, don't think this is an issue for 4.0 > only in current/ChangeLog: 15919 Clint: tcp.c, irrelevant. > only in current/ChangeLog: 15895 Bart: zftp has changed too much. > only in current/ChangeLog: 15886 Clint: tcp.c, irrelevant. > only in current/ChangeLog: 15851, 15948 Oliver: printf, irrelevant. > only in current/ChangeLog: 15762 > only in current/ChangeLog: 15763 > only in current/ChangeLog: 15766 > only in current/ChangeLog: 15768 > only in current/ChangeLog: 15770 > only in current/ChangeLog: 15772 > only in current/ChangeLog: 15779 > only in current/ChangeLog: 15780 > only in current/ChangeLog: 15783 Clint: tcp.c, irrelevant. > only in current/ChangeLog: 15765 Clint: [[:ascii:]], not necessary. > only in current/ChangeLog: 15734 Zefram: zle user key maps, overdue but too late now for 4.0. > only in current/ChangeLog: 15722 Me: _sh, not necessary. > only in current/ChangeLog: 15720 Bart/me: zmv: not necessary. > only in current/ChangeLog: 15708 Bart: quoted parameter parsing, probably best left alone for 4.0. > only in current/ChangeLog: 15647 Me: zcalc, irrelevant. > only in current/ChangeLog: 15606 Oliver: update _dd, depends on new -w argument to _values. > only in current/ChangeLog: 15597 Sven: _values -w, new option so probably not necessary. > only in current/ChangeLog: 15586 > only in current/ChangeLog: 15590 Me, Oliver: aliases in = context. Not changed in 4.0. > only in current/ChangeLog: 15588 Sven: _values stuff which gives me a headache. > only in current/ChangeLog: 15585 Me: bash functions, not necessary. > only in current/ChangeLog: 15549 Sven: Y03arguments.ztst resulting from other change only > only in current/ChangeLog: 15526 Sven: complicated _arguments stuff. > only in current/ChangeLog: 15511 Me: zcalc, irrelevant. > only in current/ChangeLog: 15489 > only in current/ChangeLog: 15509 Sven, Bart: _describe technicality. > only in current/ChangeLog: 15507 Me: zcalc, irrelevant. > only in current/ChangeLog: 15477 > only in current/ChangeLog: 15482 > only in current/ChangeLog: 15484 > only in current/ChangeLog: 15485 Sven: long/short options thing. Not necessary to backport this. > only in current/ChangeLog: 15472 Bart: remove doc for change which never happened even on 4.1. > only in current/ChangeLog: 15468 Sven: improve _ssh, probably not necessary. > only in current/ChangeLog: 15432 Me: another bash function, not necessary. > only in current/ChangeLog: 15389 > only in current/ChangeLog: 15407 > only in current/ChangeLog: 15409 > only in current/ChangeLog: 15415 Sven: long/short options complexity (15409 was a core dump), only 4.1. > only in current/ChangeLog: 15327 > only in current/ChangeLog: 15329 > only in current/ChangeLog: 15334 Me: line numbering, thing we should steer clear of putting this into 4.0. > only in current/ChangeLog: 15305 Bart: xtrace in zle funcs, probably unnecessary. > only in current/ChangeLog: 15291, 15292 > only in current/ChangeLog: 15304 Me/Bart: fix `scalar=foo; (( scalar = 0 ))', too complicated to try to backport. > only in current/ChangeLog: 15288 Me: bash functions, not necessary. > only in current/ChangeLog: 15278 (Sven), 15390 Sven: New file completion stuff, not necessary. > only in current/ChangeLog: 15271 Bart: pcre, not necessary. > only in current/ChangeLog: 15266 Me: wrong anyway. > only in current/ChangeLog: 15265 Me: new `for' syntax, not necessary. > only in current/ChangeLog: 15211 > only in current/ChangeLog: 15234 > only in current/ChangeLog: 15242 > only in current/ChangeLog: 15250 Clint: pcre, not necessary. > only in current/ChangeLog: 15228 Clint: AC_CHECK_LIB -> AC_SEARCH_LIBS. Best to leave this, I should think. > only in current/ChangeLog: 15169 Oliver/Me: I said explicitly I wouldn't put this in 4.0 > only in current/ChangeLog: 15123 Sven: _cd and _tilde, I'm sick of this, either we leave well alone or copy the current versions straight back to 4.0. > only in current/ChangeLog: 15087 > only in current/ChangeLog: 15115 Sven: handle disowning stopped jobs, new feature, probably leave out though I don't think it's really a problem. > only in current/ChangeLog: 15056 Bart: magic space change, deliberately left out, I believe. > only in current/ChangeLog: 15030 Me: definitely non-POSIX for, not to go in. > only in current/ChangeLog: 15029 Sven: _man improvements, probably too complicated. > only in current/ChangeLog: 14952 Sven: zwc stuff, probably best left out. > only in current/ChangeLog: 14939 Me: was complete crap. > only in current/ChangeLog: 14931 Andrej: zftp, prresumably tcp.c-dependent. > only in current/ChangeLog: 14915 > only in current/ChangeLog: 14929 Me: can't de-readonlify a special; make positional parameter arrays (but not their elements) readonly. Not to be put into 4.0. -- Peter Stephenson Work: pws@csr.com Web: http://www.pwstephenson.fsnet.co.uk