Bruno Haible wrote:
> Hi Jim,
>
>> @@ -49,5 +49,5 @@ main (void)
>>x.ss_family = 42;
>>i = 42;
>>
>> - return 0;
>> + return ! (i + x.ss_family);
>> }
>
> This makes is hard to understand that the return value is really 0. (It costs
> extra brain cycles to see this.) Wouldn't it be poss
Hi Jim,
> @@ -49,5 +49,5 @@ main (void)
>x.ss_family = 42;
>i = 42;
>
> - return 0;
> + return ! (i + x.ss_family);
> }
This makes is hard to understand that the return value is really 0. (It costs
extra brain cycles to see this.) Wouldn't it be possible to write this?
(void) x;
Hi Ralf,
> Wouldn't build-aux/rerun be a good candidate for a new gnulib module?
No, why would it? Packages don't ship 'rerun'. Only maintainers have it in
their PATH somewhere.
I haven't initiated moving 'git-merge-changelog' out of gnulib, where it
doesn't belong, just to repeat the same mista
Jim Meyering wrote:
> This one is trivial and seems uncontroversial.
> Pushed.
Yes. Thanks.
The gcc option that triggers this is "-Wwrite-strings". And it triggers some
more warnings. I'm fixing these through the patch below.
test-argp.c:380: warning: passing argument 5 of 'test_optional' disc
Ralf Wildenhues wrote:
> Hi Jim,
>
> * Jim Meyering wrote on Fri, Jun 11, 2010 at 01:49:35PM CEST:
>> +warn_() { echo "$@" 1>&$stderr_fileno_; }
>> +fail_() { warn_ "$ME_: failed test: $@"; Exit 1; }
>> +skip_() { warn_ "$ME_: skipped test: $@"; Exit 77; }
>> +framework_failure_() { warn_ "$ME_: s
Hi Bruno,
* Bruno Haible wrote on Fri, Jun 11, 2010 at 10:45:29AM CEST:
> The trick that I use is a script that shows me the differences between
> the last run and the current run of a command. So, in my working directories,
> I normally don't run "./configure" and "make" any more, but
> "./rerun-
Hi Jim,
* Jim Meyering wrote on Fri, Jun 11, 2010 at 01:49:35PM CEST:
> +warn_() { echo "$@" 1>&$stderr_fileno_; }
> +fail_() { warn_ "$ME_: failed test: $@"; Exit 1; }
> +skip_() { warn_ "$ME_: skipped test: $@"; Exit 77; }
> +framework_failure_() { warn_ "$ME_: set-up failure: $@"; Exit 1; }
Yo
Eric Blake wrote:
> On 06/11/2010 09:36 AM, Jim Meyering wrote:
>> FYI, I was not happy to see the 10 new warnings like this one:
>>
>> test-inttostr.c:66:3: warning: string length '5001' is greater than the
>> length '4095' ISO C99 compilers are required to support
>> [-Woverlength-strings]
Eric Blake wrote:
> On 06/11/2010 09:36 AM, Jim Meyering wrote:
>> FYI, I was not happy to see the 10 new warnings like this one:
>>
>> test-inttostr.c:66:3: warning: string length '5001' is greater than the
>> length '4095' ISO C99 compilers are required to support
>> [-Woverlength-strings]
On 06/11/2010 09:36 AM, Jim Meyering wrote:
> FYI, I was not happy to see the 10 new warnings like this one:
>
> test-inttostr.c:66:3: warning: string length '5001' is greater than the
> length '4095' ISO C99 compilers are required to support [-Woverlength-strings]
>
>
>>From 560b837a96eee2
FYI, I was not happy to see the 10 new warnings like this one:
test-inttostr.c:66:3: warning: string length '5001' is greater than the
length '4095' ISO C99 compilers are required to support [-Woverlength-strings]
>From 560b837a96eee265a5d7f20da2277f2fdc2b6562 Mon Sep 17 00:00:00 2001
From:
Paolo Bonzini wrote:
> On 06/11/2010 01:49 PM, Jim Meyering wrote:
>> Here's a patch that does that at the expense of using
>> shell functions*before* guaranteeing we have a decent shell.
>> I think that is ok, these days.
>
> Yes, it is. There may be bugs in the implementation of shell
> functio
On 06/11/2010 01:49 PM, Jim Meyering wrote:
Here's a patch that does that at the expense of using
shell functions*before* guaranteeing we have a decent shell.
I think that is ok, these days.
Yes, it is. There may be bugs in the implementation of shell functions,
but IMO assuming the shell fu
Stefano Lattarini wrote:
> At Friday 11 June 2010, Jim Meyering wrote:
>> > diff --git a/tests/init.sh b/tests/init.sh
>> > index ef0957c..4eb4a95 100644
>> > --- a/tests/init.sh
>> > +++ b/tests/init.sh
>> > @@ -97,12 +97,15 @@ else
>> > for re_shell_ in "${CONFIG_SHELL:-no_shell}" /bin/sh ba
At Friday 11 June 2010, Jim Meyering wrote:
> > diff --git a/tests/init.sh b/tests/init.sh
> > index ef0957c..4eb4a95 100644
> > --- a/tests/init.sh
> > +++ b/tests/init.sh
> > @@ -97,12 +97,15 @@ else
> > for re_shell_ in "${CONFIG_SHELL:-no_shell}" /bin/sh bash
> > dash zsh pdksh fail do
> >
Stefano Lattarini wrote:
> The attached patch fixes a minor wart in the `tests/init.sh' shell
> library. IMO it is simple enough not to require a copyright
> assigement for gnulib (which I don't have).
>
> * tests/init.sh: Do not use `$ME_' and `skip_' before they are
> defined.
...
Hi Stefano,
Hello gnulibers.
The attached patch fixes a minor wart in the `tests/init.sh' shell
library. IMO it is simple enough not to require a copyright
assigement for gnulib (which I don't have).
Regards,
Stefano
---
* tests/init.sh: Do not use `$ME_' and `skip_' before they are
defined.
Chan
One more.
Without the patch below, I'd get these:
test-sys_socket.c:36:15: warning: variable 'i' set but not used
[-Wunused-but-set-variable]
test-sys_socket.c:35:27: warning: variable 'x' set but not used
[-Wunused-but-set-variable]
>From ef5acf202788003cdc70813756e173cd6b7dd5a1 Mon S
FYI, just pushed.
Without the following, I'd get these warnings in coreutils/gnulib-tests:
test-argmatch.c:34:6: warning: no previous declaration for 'usage'
[-Wmissing-declarations]
test-exclude.c:68:6: warning: no previous declaration for 'usage'
[-Wmissing-declarations]
>From d
I've just pushed this, too:
>From 9e1e9d8f3bac7fb52b026f0b628eb37d3a839097 Mon Sep 17 00:00:00 2001
From: Jim Meyering
Date: Fri, 11 Jun 2010 11:03:54 +0200
Subject: [PATCH] test-xvasprintf: avoid 'const' discard warnings
* tests/test-xvasprintf.c (test_xvasprintf, test_xasprintf): Use
"const" w
Hi Ben,
> Cross-compiling lib/strtod.c failed on Mingw, as reported at
> http://savannah.gnu.org/bugs/?29965:
>
> strtod.c:37: error: redefinition of 'strtod'
> /usr/i686-pc-mingw32/sys-root/mingw/include/stdlib.h:319: note:
> previous definition of 'strtod' was here
>
> The problem
Jim Meyering wrote:
> This is a strong argument for enabling -Werror when compiling tests.
> Even on Fedora 13, these warnings have been flying by:
The problem is catching the developer's attention, when some compilation
produces tons of output. The use of automake's options 'silent-rules' and
'co
This one is trivial and seems uncontroversial.
Pushed.
>From 84746b2f407af414344b9e6a611edd2e42ceeec6 Mon Sep 17 00:00:00 2001
From: Jim Meyering
Date: Fri, 11 Jun 2010 09:27:19 +0200
Subject: [PATCH] tests: avoid 'const' discard warnings in mbsstr tests
* tests/test-mbsstr1.c (main): Add "const
I've pushed these two, too:
>From cc7a4e12a15c81018321fdf653c081cb42cf05b3 Mon Sep 17 00:00:00 2001
From: Jim Meyering
Date: Fri, 11 Jun 2010 09:08:47 +0200
Subject: [PATCH 1/2] test-inttostr.c: include for use of strcmp
* tests/test-inttostr.c: Include for strcmp declaration.
---
ChangeLog
I noticed this patch go into launchpad against coreutils,
but it affects gnulib, so here we go.
This is a strong argument for enabling -Werror when compiling tests.
Even on Fedora 13, these warnings have been flying by:
test-link.h:47:5: warning: implicit declaration of function 'stat'
[-Wimpl
25 matches
Mail list logo
