Time for zsh-3.1.5 ?

Time for zsh-3.1.5 ?

[Bart Schaefer]

Here's a list of all the changes I have to zsh-3.1.4.  There have been a
few others to port to or fix the build process on other platforms, which I
didn't pick up.  Some of these are fixes for fairly serious bugs; what's
the prognosis for an official 3.1.5 appearing soon?

Are there any of the following that are deemed unsuitable for inclusion in
the next release?

(Sorry I don't have all the article numbers.)
__________

Fix "for subdir in $(SUBDIRS)" in clean.mk and config.mk, which broke bash.

PWS's patches from zsh-workers 4209 and 4212 for POSIX character ranges in
glob patterns.

PWS's patch from 4172 to replace inerrflush() and discard_input() with
herrflush() and to eliminate the lastc global entirely.  This makes zsh
discard all buffered input, rather than just up to the end of the current
line, on a history or parse error.  (Supercedes PWS's patch from 4095 for
unintentional shell exit when `lastc' is incorrectly set.)

PWS's patches from zsh-workers 4191 and 4196 to abort processing on parse
errors except in interactive shells or init files (except in ksh mode).

Bernd Eggink's patch from 4192 for mishandled "select" input.

My patch to stop broken variants of "su" from sending zsh into the wrong
emulation mode by providing a bogus argv[0].

PWS's patch from zsh-workers 4317 for improper local variable `unset'.

Zefram's patch to handle years > 1999 correctly in ztrftime().

PWS's patch from zsh-workers 4250 to regenerate signal names on reconfigure.

Goran Larsson's patch from zsh-workers 4105 to nul-terminate a string.

My patch to fix "compctl -S ''" (make it work again).

Sven's patch from 4147 for strange completinword behavior.

Sven's patch from zsh-workers 4140 for starting menu completion following a
compctl that uses -U.

PWS's correction from zsh-workers 4128 of insertion of spaces when using
completeinword and alwaystoend together.

PWS's patch from zsh-workers 4124 for excess adding of slashes when completing.

PWS's patch from zsh-workers 4315 for history search segfault.

Wayne Davison's patch for spaceinline().

Wayne Davison's misc. up/down -line patches from zsh-workers 4086.

Wayne Davison's patches from zsh-workers 4104 and (an earlier one whose number
I've lost) to make history-search-backward more like the 3.0.5 version and to
use zmult sensibly in the search functions (which previously ignored it).

My patch to pass widget name and optionally zmult as parameters to user-
defined widgets.

My patch to fix error in handling of negated digit arguments.

Sven's patch from zsh-workers 4283 to make optional the second string in r[]
and R[] extended compctl patterns.

My patch to wrap execzlefunc() in PERMALLOC { ... } LASTALLOC.

My patch for excess "make" verbosity in compiling manpages and info.

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

Re: Time for zsh-3.1.5 ?

[Zefram]

Bart Schaefer wrote:
>                                                                  what's
>the prognosis for an official 3.1.5 appearing soon?

Sorry, I've been a bit busy.  3.1.5 and 3.0.6 will be out as soon as I
can manage; should be done in the next few weeks.

>My patch to stop broken variants of "su" from sending zsh into the wrong
>emulation mode by providing a bogus argv[0].

I'm very doubtful about what the ideal `correct' behaviour is here.
For the moment I'll be leaving it as it is.

>My patch to pass widget name and optionally zmult as parameters to user-
>defined widgets.

I don't think this is the right mechanism to use.  I'll consider it
more fully before making a final decision, but this probably won't be
in the release.

>My patch for excess "make" verbosity in compiling manpages and info.

make displaying all commands it executes is a feature, not a bug.
Hiding the echoes would be good, however.

-zefram

Re: Time for zsh-3.1.5 ?

[Bart Schaefer]

On Aug 26, 10:09am, Zefram wrote:
} Subject: Re: Time for zsh-3.1.5 ?
} 
} Sorry, I've been a bit busy.  3.1.5 and 3.0.6 will be out as soon as I
} can manage; should be done in the next few weeks.

Ok.  BTW, I forgot to add all my man page patches to that list.

} >My patch to pass widget name and optionally zmult as parameters to user-
} >defined widgets.
} 
} I don't think this is the right mechanism to use.

It is, nevertheless, extremely useful, particularly access to zmult
(and to the knowledge of whether a multiplier was explicitly given or
just happens to be 1 as the default).  I encourage that something of
the kind be included.

} >My patch for excess "make" verbosity in compiling manpages and info.
} 
} make displaying all commands it executes is a feature, not a bug.

Whatever ... you'll note that I added @echo of the commands that actually
do any interesting work, if an @ somewhere else has prevented make from
doing so.  I don't think my patch suppresses any useful output; apply it
and run a make and see what you think.

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