Configure cleanups

In change #30919, Nicholas added some cleanups for 'try'. He
however did
not catch all, which isn't his fault, but unveils a somewhat
unclean
state of affairs.

There are several approaches to the cleanup.
The common flow of control is to write the test program in
try.c, then
compile it, if that succeeded, run it, read the exit code,
and clean up.

But the cleanup's are a mess. Here's the possible used
combo's:

$rm -f core core.try.* try.core
$rm -f core try.core core.try.*
$rm -f try.*
$rm -f try*
$rm -f try.c
rm -f try.c a.out
$rm -f try.[co]
$rm -f try.* core core.* *.core *.core.*
$rm -f try.c try
$rm -f try.c try.o'
$rm -f try.c tryout.c'
$rm -f try$_exe try.c;
$rm -f try$_exe try.c core core.* try.core;
$rm -f try.h
$rm -f try.* .out
$rm -f try try.*
$rm -f try.* try
$rm -f try try.*
$rm -f try try.c
$rm -f try try.* core
$rm -f try.* try core
$rm -f try.* try core core.try.*
$rm -f try try.* cplusplus$$
$rm -f try.* try$exe_ext
$rm -f try try.* glibc.ver
$rm -f try try$_o
$rm -f try try$_o try.c
$rm -f try try.* .out core
$rm -f try try.* .out core head.c mtry
$rm try.*

I'm thinking about creating a macro that does something
like

cleanup_try="$rm -f try try$exe_ext a.out try.c try.o
.out core core.* try.core"

for extended cleanup's,

$cleanup_try head.c mtry

should still work.
This however is a huge task. Any objections before I start?

-- 
H.Merijn Brand         Amsterdam Perl Mongers (http://amsterdam.pm.org/
)
using & porting perl 5.6.2, 5.8.x, 5.9.x   on HP-UX
10.20, 11.00, 11.11,
& 11.23, SuSE 10.0 & 10.2, AIX 4.3 & 5.2, and
Cygwin. http://qa.perl.org
http://mirrors.de
velooper.com/hpux/            http://www.test-smoke.org
                        http
://www.goldmark.org/jeff/stupid-disclaimers/

On Wed, 18 Apr 2007 13:22:18 +0200, "H.Merijn
Brand" <h.m.brandxs4all.nl>
wrote:

> In change #30919, Nicholas added some cleanups for
'try'. He however did
> not catch all, which isn't his fault, but unveils a
somewhat unclean
> state of affairs.
> 
> There are several approaches to the cleanup.
> The common flow of control is to write the test program
in try.c, then
> compile it, if that succeeded, run it, read the exit
code, and clean up.
> 
> But the cleanup's are a mess. Here's the possible used
combo's:

Here would be the suggested cleanup:
16155 2007-04-18 14:15 h
ttp://www.xs4all.nl/~hmbrand/Configure-rm_try.diff

That would imply a change of 83 metaunits. Worth the
trouble?

> This however is a huge task. Any objections before I
start?

-- 
H.Merijn Brand         Amsterdam Perl Mongers (http://amsterdam.pm.org/
)
using & porting perl 5.6.2, 5.8.x, 5.9.x   on HP-UX
10.20, 11.00, 11.11,
& 11.23, SuSE 10.0 & 10.2, AIX 4.3 & 5.2, and
Cygwin. http://qa.perl.org
http://mirrors.de
velooper.com/hpux/            http://www.test-smoke.org
                        http
://www.goldmark.org/jeff/stupid-disclaimers/

On Wed, 18 Apr 2007 09:04:48 -0400 (EDT), Andy Dougherty
<dougheralafayette.edu> wrote:

> On Wed, 18 Apr 2007, H.Merijn Brand wrote:
> 
> > In change #30919, Nicholas added some cleanups for
'try'. He however did
> > not catch all, which isn't his fault, but unveils
a somewhat unclean
> > state of affairs.
> > 
> > There are several approaches to the cleanup.
> > The common flow of control is to write the test
program in try.c, then
> > compile it, if that succeeded, run it, read the
exit code, and clean up.
> 
> > I'm thinking about creating a macro that does
something like
> > 
> > cleanup_try="$rm -f try try$exe_ext a.out
try.c try.o .out core core.* try.core"
> 
> > This however is a huge task. Any objections before
I start?
> 
> No objections from me.  It sounds like a good plan. 
Looking forward and 
> dreaming a bit, this might also make it easier later to
extend the cleanup 
> process to do something useful in cases where tests
fail. 

Open for suggestions.
Above suggestion is now in as change #31040 (76 files
changed)
Enjoy the clean feeling 

mc_units snapshot uploaded to $CPAN

-- 
H.Merijn Brand         Amsterdam Perl Mongers (http://amsterdam.pm.org/
)
using & porting perl 5.6.2, 5.8.x, 5.9.x   on HP-UX
10.20, 11.00, 11.11,
& 11.23, SuSE 10.0 & 10.2, AIX 4.3 & 5.2, and
Cygwin. http://qa.perl.org
http://mirrors.de
velooper.com/hpux/            http://www.test-smoke.org
                        http
://www.goldmark.org/jeff/stupid-disclaimers/

This changes looks 'broken' as it causes a whole bunch of:
    ./protochk: line 7: -f: command not found
during configure.

On 4/23/07, H.Merijn Brand <h.m.brandxs4all.nl> wrote:
> On Wed, 18 Apr 2007 09:04:48 -0400 (EDT), Andy
Dougherty
> <dougheralafayette.edu> wrote:
>
> > On Wed, 18 Apr 2007, H.Merijn Brand wrote:
> >
> > > In change #30919, Nicholas added some
cleanups for 'try'. He however did
> > > not catch all, which isn't his fault, but
unveils a somewhat unclean
> > > state of affairs.
> > >
> > > There are several approaches to the cleanup.
> > > The common flow of control is to write the
test program in try.c, then
> > > compile it, if that succeeded, run it, read
the exit code, and clean up.
> >
> > > I'm thinking about creating a macro that does
something like
> > >
> > > cleanup_try="$rm -f try try$exe_ext
a.out try.c try.o .out core core.* try.core"
> >
> > > This however is a huge task. Any objections
before I start?
> >
> > No objections from me.  It sounds like a good
plan.  Looking forward and
> > dreaming a bit, this might also make it easier
later to extend the cleanup
> > process to do something useful in cases where
tests fail.
>
> Open for suggestions.
> Above suggestion is now in as change #31040 (76 files
changed)
> Enjoy the clean feeling 
>
> mc_units snapshot uploaded to $CPAN
>
> --
> H.Merijn Brand         Amsterdam Perl Mongers (http://amsterdam.pm.org/
)
> using & porting perl 5.6.2, 5.8.x, 5.9.x   on HP-UX
10.20, 11.00, 11.11,
> & 11.23, SuSE 10.0 & 10.2, AIX 4.3 & 5.2,
and Cygwin. http://qa.perl.org
> http://mirrors.de
velooper.com/hpux/            http://www.test-smoke.org
>                         http
://www.goldmark.org/jeff/stupid-disclaimers/
>