Suggested 4.0.2 contents

Suggested 4.0.2 contents

[Bart Schaefer]

Here's what I think the 4.0.1 --> 4.0.2 ChangeLog diff should look like
based on patches so far (including a few on the dev trunk that have not
be committed to the branch yet).  This has been synchronized with the
trunk ChangeLog to make it easier to see what's different in the latest
dev version; that means the entries reflect the dates that changes were
made to the dev code, not necessarily the dates they were committed to
the branch -- I think that makes sense, as we have the CVS logs to tell
us the latter.

I have a patched-up branch sandbox from which I can commit all the bits
mentioned below that have not yet been committed, if PWS and Sven agree
that this looks right.  (Note this diff can't be patched onto the branch
ChangeLog, it has to be patched onto the 4.0.1 release ChangeLog.)

Index: ChangeLog
===================================================================
RCS file: /cvsroot/zsh/zsh/ChangeLog,v
retrieving revision 1.1193
diff -u -r1.1193 ChangeLog
--- ChangeLog	2001/06/01 14:46:51	1.1193
+++ ChangeLog	2001/06/16 17:39:08
@@ -1,3 +1,125 @@
+2001-06-14  Peter Stephenson  <pws@csr.com>
+
+	* 14921: Completion/Zsh/_zftp: autoload zfcd_match and zfget_match
+	if necessary.
+
+2001-06-13  Sven Wischnowsky  <wischnow@zsh.org>
+
+	* 14903: Src/Zle/complist.c, Src/Zle/zle_tricky.c: fixes for
+	display bugs for completion lists (cursor left one line to far
+	up or down)
+
+2001-06-13  Peter Stephenson  <pws@csr.com>
+
+	* 14893: Src/signals.c: make handler mod_export because it's
+	needed for signal queueing in various modules.  Probably
+	sensible to change handler to zhandler too --- not done yet.
+
+2001-06-13  Sven Wischnowsky  <wischnow@zsh.org>
+
+	* 14897 (and Oliver Kiddle: 14862): Completion/Base/Utility/_values,
+	Completion/Unix/Command/_mount, Completion/Unix/Type/_dir_list,
+	Src/Zle/computil.c: fix for comparguments (completing after
+	single letter options that get their argument in the next
+	word); make _values ignore some of the standard option it may
+	get; make _mount use _dir_list in some places and improve that;
+	`document' comparguments and compvalues with some comments in
+	computil.c
+
+2001-06-13  Oliver Kiddle  <opk@zsh.org>
+
+	* 14896: Completion/Zsh/Type/_command_names: don't complete read-only
+	parameters on the left side of an assignment
+
+2001-06-13  Peter Stephenson  <pws@csr.com>
+
+	* unposted: Completion/compinstall: spotted by Bart: assignments
+	when finding $fpath if not set in current shell were completely
+	garbled.
+
+2001-06-13  Sven Wischnowsky  <wischnow@zsh.org>
+
+	* 14883 (Jos Backus): Completion/Unix/Command/_mount: add code for
+	BSDish systems
+
+2001-06-12  Bart Schaefer  <schaefer@zsh.org>
+
+	* 14849: README: Request "reporter" output only if a bug cannot
+	be reproduced with 'zsh -f'.
+
+2001-06-11  Sven Wischnowsky  <wischnow@zsh.org>
+
+	* 14841: Completion/Base/Utility/_values, Src/Zle/computil.c:
+	new word parsing for compvalues; this should now handle empty
+	separators correctly, as far as it is possible
+
+2001-06-11  Peter Stephenson  <pws@csr.com>
+
+	* 14840: Doc/Zsh/options.yo: document search behaviour with
+	PATH_DIRS option.
+
+2001-06-10  Clint Adams  <clint@zsh.org>
+
+	* 14830: config.guess, config.sub:
+	upgrade to GNU's June 5 and 8 versions,
+	respectively.
+
+2001-06-09  Bart Schaefer  <schaefer@zsh.org>
+
+	* unposted (see 14826): Test/C02cond.ztst: Force group of new
+	directory created for file-mode tests to be owned by the current
+	effective group.
+
+	* 14824: Test/comptest: Avoid echotc (improvement on 14792).
+
+2001-06-08  Bart Schaefer  <schaefer@zsh.org>
+
+	* 14796: zshconfig.ac: More reliable BROKEN_KILL_ESRCH test.
+
+2001-06-07  Bart Schaefer  <schaefer@zsh.org>
+
+	* 14792: Test/comptest: Don't import the current terminal type for
+	the zpty terminal, because the current terminal might not be able
+	to run ZLE (e.g., emacs shell mode).
+
+	* 14712: Doc/Zsh/builtins.yo: Improve appearance of long hanging
+	paragraph tag for the "read" builtin.
+
+	* 14578: Src/Modules/termcap.c: Handle vagaries of tgetflag()
+	return value in new function ztgetflag().
+
+2001-06-07  Wayne Davison  <wayned@users.sourceforge.net>
+
+	* 14793: Src/hist.c: Don't mark a HIST_TMPSTORE entry with HIST_OLD
+	so that the HFILE_FAST algorithm in savehistfile() will recheck this
+	histnum later.
+
+2001-06-06  Andrej Borsenkow  <bor@zsh.org>
+
+	* 14772: David Lebel <lebel@lebel.org>: zshconfig.ac: patch
+	for dynamic linking on OpenBSD
+
+	* 14766: Src/prompt.c: uniform %l substitution
+
+	* 14679, Bart 14693, 14758, 14760: configure.ac, zshconfig.ac,
+	.distfiles, Makefile.in, configure.in, Src/mkmakemod.sh: Allow
+	processing by both autoconf-2.13 and autoconf-2.50; remove
+	config.status hack for autoconf-2.50
+
+2001-06-06  Sven Wischnowsky  <wischnow@zsh.org>
+
+	* 14736: Doc/Zsh/compsys.yo: make docs for _wanted and
+	_requested clearer (command has to allow options passed to it)
+
+	* 14733: Completion/Base/Utility/_arguments: two fixes: try all
+	actions that have to be tried (but only once), execute actions
+	if we aren't after an option already (in the same word)
+
+2001-06-05  Chmouel Boudjnah  <chmouel@mandrakesoft.com>
+
+	* 14550: Completion/Redhat/Command/_rpm: Add --nobuild and --nogpg
+	to rpm completion
+
 2001-06-01  Peter Stephenson  <pws@csr.com>
 
 	* Config/version.mk: 4.0.1 released, finally.

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net   

Re: Suggested 4.0.2 contents

[Andrej Borsenkow]

Bart Schaefer wrote:

> Here's what I think the 4.0.1 --> 4.0.2 


I currently have a patch for HEAD that should fix the reported problems 
in AIX. Unforunately, I have no feedback so far. The patch actually 
consists of

- fix a bug introduced by me at some point (changed cache variable 
name). This would prevent dynamic loading even on tested AIX 3.x. I 
guess, it must go 4.0.2 in any case

- reshuffle if's in zshconfig.ac to force AIX style if AIX functions 
were found.

- add test for base type of third argument of access(). AIX compiler is 
picky as it seems. Currently it tests for int *, unsigned long * and 
size_t * and defaults to int * if everything else fails. We currently 
use int * in code so it should not break any system but you know ...


I guess, I should commit it to HEAD anyway. You decide if it should go 
in 4.0.2.

-andrej

Re: Suggested 4.0.2 contents

[Sven Wischnowsky]

Bart Schaefer wrote:

> Here's what I think the 4.0.1 --> 4.0.2 ChangeLog diff should look like
> based on patches so far (including a few on the dev trunk that have not
> be committed to the branch yet).

Looks good, as far as I can see.

I've just committed the zhandler() patch to 4.0, but not the patch for
zcompile because that's a user-visible change and people might still
want to complain about the 0444 creation mode.

Also, there's still the job problem on Irix (14861, 14865, 14911) to
which we got no reply.


Bye
  Sven


-- 
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de

RE: Suggested 4.0.2 contents

[Andrej Borsenkow]

> 
> - fix a bug introduced by me at some point (changed cache variable 
> name). This would prevent dynamic loading even on tested AIX 3.x. I 
> guess, it must go 4.0.2 in any case
> 

I have committed it.

-andrej

Re: Suggested 4.0.2 contents

[Peter Stephenson]

"Bart Schaefer" wrote:
> Here's what I think the 4.0.1 --> 4.0.2 ChangeLog diff should look like
> based on patches so far (including a few on the dev trunk that have not
> be committed to the branch yet).

Looks fine to me.  Are you going to handle 4.0.2?  It probably stands a
better chance of working.

It would be good to get the AIX changes out of the way.  Andrej, you've
been keep track of this better than me, has Timothy Miller got the latest
set of patches to try out?

I don't think we should necessarily wait up for the IRIX problem, there's
no sign it's really under control at the moment.

-- 
Peter Stephenson <pws@csr.com>                  Software Engineer
CSR Ltd., Unit 300, Science Park, Milton Road,
Cambridge, CB4 0XL, UK                          Tel: +44 (0)1223 392070


**********************************************************************
The information transmitted is intended only for the person or
entity to which it is addressed and may contain confidential 
and/or privileged material. 
Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by 
persons or entities other than the intended recipient is 
prohibited.  
If you received this in error, please contact the sender and 
delete the material from any computer.
**********************************************************************

RE: Suggested 4.0.2 contents

[Andrej Borsenkow]

>
> It would be good to get the AIX changes out of the way.  Andrej, you've
> been keep track of this better than me, has Timothy Miller got the latest
> set of patches to try out?
>

Unfortunately, I have not heard of Timothy for some time. I got contacted by
Timo, but he had weird problem with boolcodes & friends:

> Well, can you explain this
>
> term.h:
>
> extern  char    term_parm_err[], *term_err_strings[], *Def_term,
>                 *boolnames[], *boolcodes[], *boolfnames[],
>                 *numnames[], *numcodes[], *numfnames[],
>                 *strnames[], *strcodes[], *strfnames[];
>
> config.log:
>
> configure:3367: checking if boolcodes is available
> configure:3379: cc -qlanglvl=ansi -o conftest  -O
>  conftest.c -lcurses -lm  -lc
>  1>&5
> ld: 0711-317 ERROR: Undefined symbol: boolcodes
> ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more
> information.
> configure: failed program was:
> #line 3369 "configure"
> #include "confdefs.h"
> #ifdef TERM_H_NEEDS_CURSES_H
> #include <curses.h>
> #endif
> #include <term.h>
> int main() {
> char **test = boolcodes; printf(*test);
> ; return 0; }
>
> In which library is boolcodes then?

and I got no new info so far.

-andrej

Re: Suggested 4.0.2 contents

[Bart Schaefer]

On Jun 18, 10:34am, Peter Stephenson wrote:
} Subject: Re: Suggested 4.0.2 contents
}
} Looks fine to me.  Are you going to handle 4.0.2?  It probably stands a
} better chance of working.

Er, um ... I'll go ahead and commit this stuff, and I'm happy to keep
track of which patches ought to be in there, but for this one at least
I'd prefer it if you do the actual distribution file releases.  My
schedule is a bit convoluted the next several weeks.

We can talk about it again when (I won't say "if") 4.0.3 approaches.  OK?

-- 
Bart Schaefer                                 Brass Lantern Enterprises
http://www.well.com/user/barts              http://www.brasslantern.com

Zsh: http://www.zsh.org | PHPerl Project: http://phperl.sourceforge.net