4.3.13 changes

4.3.13 changes

[Peter Stephenson]

I'd like to make a release soon to get out the fixes for job control and
file descriptor management.  Is there anything else important in the
main shell that needs fixing?  (If it didn't need fixing for 4.3.12, it
doesn't need fixing this time round.)

-- 
Peter Stephenson <pws@csr.com>            Software Engineer
Tel: +44 (0)1223 692070                   Cambridge Silicon Radio Limited
Churchill House, Cambridge Business Park, Cowley Road, Cambridge, CB4 0WZ, UK


Member of the CSR plc group of companies. CSR plc registered in England and Wales, registered number 4187346, registered office Churchill House, Cambridge Business Park, Cowley Road, Cambridge, CB4 0WZ, United Kingdom
More information can be found at www.csr.com. Follow CSR on Twitter at http://twitter.com/CSR_PLC and read our blog at www.csr.com/blog

Re: 4.3.13 changes

[Nikolai Weibull]

On Fri, Jul 22, 2011 at 13:08, Peter Stephenson
<Peter.Stephenson@csr.com> wrote:
> I'd like to make a release soon to get out the fixes for job control and
> file descriptor management.  Is there anything else important in the
> main shell that needs fixing?  (If it didn't need fixing for 4.3.12, it
> doesn't need fixing this time round.)

Can you please have a look at the '*:: :->file' problem that I’m
having with _arguments (see _git-add in Completion/Unix/Command/_git)?
 I need to know if this is a problem with _arguments or if I’m doing
something wrong.  The gist of it is that options don’t complete after
arguments if you use “::” instead of “:”.

Also, I’m in the process of fixing some minor nits with _git that were
introduced in the patch series that added support for third-party
commands and user commands.  They should be resolved during the day.

Re: 4.3.13 changes

[İsmail Dönmez]

[-- Attachment #1: Type: text/plain, Size: 455 bytes --]

Hi;

On Fri, Jul 22, 2011 at 1:08 PM, Peter Stephenson
<Peter.Stephenson@csr.com>wrote:

> I'd like to make a release soon to get out the fixes for job control and
> file descriptor management.  Is there anything else important in the
> main shell that needs fixing?  (If it didn't need fixing for 4.3.12, it
> doesn't need fixing this time round.)


Given the large backlog of crash fixes and other stuff a 4.3.12 release
would be nice.

Regards,
ismail

Re: 4.3.13 changes

[İsmail Dönmez]

[-- Attachment #1: Type: text/plain, Size: 603 bytes --]

On Fri, Aug 19, 2011 at 10:53 PM, İsmail Dönmez <ismail@namtrac.org> wrote:

> Hi;
>
> On Fri, Jul 22, 2011 at 1:08 PM, Peter Stephenson <
> Peter.Stephenson@csr.com> wrote:
>
>> I'd like to make a release soon to get out the fixes for job control and
>> file descriptor management.  Is there anything else important in the
>> main shell that needs fixing?  (If it didn't need fixing for 4.3.12, it
>> doesn't need fixing this time round.)
>
>
> Given the large backlog of crash fixes and other stuff a 4.3.12 release
> would be nice.
>

I mean 4.3.13 release of course...

ismail

Re: 4.3.13 changes

[Peter Stephenson]

On Fri, 22 Jul 2011 13:13:49 +0200
Nikolai Weibull <now@bitwi.se> wrote:
> Can you please have a look at the '*:: :->file' problem that I’m
> having with _arguments (see _git-add in Completion/Unix/Command/_git)?
>  I need to know if this is a problem with _arguments or if I’m doing
> something wrong.  The gist of it is that options don’t complete after
> arguments if you use “::” instead of “:”.

I finally looked at this.  The documentation says

 With two colons before the message, the words special array and the
 CURRENT special parameter are modified to refer only to the normal
 arguments when the action is executed or evaluated.

I can't see any evidence not being able to complete options thereafter
is deliberate.  This is to do with the restrict_range() calls in
ca_set_data(), I think.  The :: causes the first argument to be
ca_laststate.argbeg instead of ca_laststate.optbeg, which fits the
descxription; I can't see any other intentional effect.  It's possible
there's some miscounting later, or something, but it looks obscure.

-- 
Peter Stephenson <p.w.stephenson@ntlworld.com>
Web page now at http://homepage.ntlworld.com/p.w.stephenson/

Re: 4.3.13 changes

[Nikolai Weibull]

On Mon, Sep 12, 2011 at 21:51, Peter Stephenson
<p.w.stephenson@ntlworld.com> wrote:
> On Fri, 22 Jul 2011 13:13:49 +0200
> Nikolai Weibull <now@bitwi.se> wrote:
>> Can you please have a look at the '*:: :->file' problem that I’m
>> having with _arguments (see _git-add in Completion/Unix/Command/_git)?

> I can't see any evidence not being able to complete options thereafter
> is deliberate.  This is to do with the restrict_range() calls in
> ca_set_data(), I think.  The :: causes the first argument to be
> ca_laststate.argbeg instead of ca_laststate.optbeg, which fits the
> descxription; I can't see any other intentional effect.  It's possible
> there's some miscounting later, or something, but it looks obscure.

OK.  I’ll simply leave it as is for now.  If this gets fixed it gets
fixed.  I don’t think it’s worth trying to work around it.

Re: 4.3.13 changes

[Bart Schaefer]

On Sep 12,  8:51pm, Peter Stephenson wrote:
} Subject: Re: 4.3.13 changes
}
} On Fri, 22 Jul 2011 13:13:49 +0200
} Nikolai Weibull <now@bitwi.se> wrote:
} > Can you please have a look at the '*:: :->file' problem [...] ?
} > The gist of it is that options don't complete after
} > arguments if you use "::" instead of ":".
} 
} I can't see any evidence not being able to complete options thereafter
} is deliberate.

I suspect the internals that support _arguments assume that you don't
mix arguments with options except when the arguments are those of a
particular option; that is, that a command always looks like

	command-name command-options command-arguments

where command-options may be zero or more sets of

	option option-arguments

I seem to recall discussions on the austin-group (POSIX) list about
the correctess of allowing more options to follow command-arguments, and
incompatibilities introduced by variant utilities which allow that.

The very fact that NORMARG is defined as "_the_ position of the first
normal argument" (my emphasis) seems to imply that comparguments does
not expect there to be more than one cluster of options, with all the
non-option arguments following it.