Joel E. Denny wrote:
> On Tue, 4 Aug 2009, Karl Berry wrote:
>
>> It definitely would be nice, and I've contemplated doing so from time to
>> time, but it never seemed like a big enough deal with which to occupy
>> rms's time discussing.
>
> I understand.
>
>> So, the consensus is that we shou
Joel E. Denny wrote:
> For Bison, update-copyright is not enough. Our backend skeletons contain
> m4 macro invocations with lists of copyright years as arguments. I've a
> written a script and a makefile target to handle these, but I would like
> it to be called by maint.mk's update-copyright tar
For Bison, update-copyright is not enough. Our backend skeletons contain
m4 macro invocations with lists of copyright years as arguments. I've a
written a script and a makefile target to handle these, but I would like
it to be called by maint.mk's update-copyright target rule so we don't
forg
On Tue, 4 Aug 2009, Karl Berry wrote:
> 5. Inconsistent line prefix. In many cases, it's a C-style comment such
> that "/*" appears only on the first line. In some cases, it's just
> inconsistent indentation.
>
> For better or worse, it's very common to have /* or just * preceding
On Tue, 4 Aug 2009, Karl Berry wrote:
> It definitely would be nice, and I've contemplated doing so from time to
> time, but it never seemed like a big enough deal with which to occupy
> rms's time discussing.
I understand.
> So, the consensus is that we should make (C) completely optional?
modules which cannot be compiled with c++ (regex, gettimeofday) should depend
on no-c++
Joel E. Denny wrote:
> These patches make update-copyright functionality a little easier to
> understand and a little more powerful. Once these are applied, I will
> feel better about writing a patch that allows "(C)" to be omitted.
>
>>From 61c92eb1ecb5b23c667e026cdb72cce1eaf6f847 Mon Sep 17 00:0
Joel E. Denny wrote:
> On Wed, 5 Aug 2009, Eric Blake wrote:
>
>> > +if ( diff --version < /dev/null 2>&1 | grep GNU ) 2>&1 > /dev/null;
>> > then
>> > + compare() { diff -u "$@"; }
>>
>> 'diff -u' is required by POSIX 2008. Rather than filtering on
>> whether 'diff --version' includes GNU, it wo
These patches make update-copyright functionality a little easier to
understand and a little more powerful. Once these are applied, I will
feel better about writing a patch that allows "(C)" to be omitted.
>From 61c92eb1ecb5b23c667e026cdb72cce1eaf6f847 Mon Sep 17 00:00:00 2001
From: Joel E. Den
On 08/05/2009 10:19 PM, Paolo Bonzini wrote:
Here is the patch. The point is to distinguish whether console handles
are opened for input ("input buffers") or output ("screen buffers"), and
avoid waiting on the latter. The reason is that the screen buffers can
be waited on---and the effect is the
Here is the patch. The point is to distinguish whether console handles
are opened for input ("input buffers") or output ("screen buffers"), and
avoid waiting on the latter. The reason is that the screen buffers can
be waited on---and the effect is the same as waiting on the
corresponding inpu
On Wed, 5 Aug 2009, Eric Blake wrote:
> > +if ( diff --version < /dev/null 2>&1 | grep GNU ) 2>&1 > /dev/null;
> > then
> > + compare() { diff -u "$@"; }
>
> 'diff -u' is required by POSIX 2008. Rather than filtering on
> whether 'diff --version' includes GNU, it would be better to
> filter on
> +if ( diff --version < /dev/null 2>&1 | grep GNU ) 2>&1 > /dev/null;
> then
> + compare() { diff -u "$@"; }
'diff -u' is required by POSIX 2008. Rather than filtering on
whether 'diff --version' includes GNU, it would be better to
filter on whether 'diff -u' is accepted on the command line,
to
On Tue, 4 Aug 2009, Joel E. Denny wrote:
> > [BTW, it'd be good to use a "compare" function as defined in
> > test-vc-list-files-cvs.sh, in case diff is not available or
> > does not honor the -u option. ]
>
> Ok.
The following patch implements that. Unlike test-vc-list-files-cvs.sh, I
decided
14 matches
Mail list logo
