Okay, I just discovered that using make SUBDIR hides error code, so I've
checked build logs carefully and I think my updates are good to go in.
However, it's good to be aware of this:
$ make SUBDIR=archivers/unrar fake
...
===> Faking installation for unrar-4.20v1
install -c -s -o root -g bin -m 555 /home/ports/obj/unrar-4.20/unrar/unrar
/home/ports/obj/unrar-4.20/fake-i386/usr/local/bin
install -d -o root -g bin -m 755
/home/ports/obj/unrar-4.20/fake-i386/usr/local/share/doc/unrar
install -c -o root -g bin -m 444 /home/ports/obj/unrar-4.20/unrar/readme.txt
/home/ports/obj/unrar-4.20/unrar/license.txt
/home/ports/obj/unrar-4.20/fake-i386/usr/local/share/doc/unrar
exit 1
*** Error 1 in /usr/ports/archivers/unrar (Makefile:36 'do-install')
*** Error 1 in archivers/unrar (/usr/ports/infrastructure/mk/bsd.port.mk:2716
'/home/ports/obj/unrar-4.20/fake-i386/.fake_done')
*** Error 1 in archivers/unrar (/usr/ports/infrastructure/mk/bsd.port.mk:2360
'fake')
===> Exiting archivers/unrar with an error
$ echo $?
0
$ cd archivers/unrar && make fake
===> Faking installation for unrar-4.20v1
install -c -s -o root -g bin -m 555 /home/ports/obj/unrar-4.20/unrar/unrar
/home/ports/obj/unrar-4.20/fake-i386/usr/local/bin
install -d -o root -g bin -m 755
/home/ports/obj/unrar-4.20/fake-i386/usr/local/share/doc/unrar
install -c -o root -g bin -m 444 /home/ports/obj/unrar-4.20/unrar/readme.txt
/home/ports/obj/unrar-4.20/unrar/license.txt
/home/ports/obj/unrar-4.20/fake-i386/usr/local/share/doc/unrar
exit 1
*** Error 1 in /usr/ports/archivers/unrar (Makefile:36 'do-install')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2716
'/home/ports/obj/unrar-4.20/fake-i386/.fake_done')
*** Error 1 in /usr/ports/archivers/unrar
(/usr/ports/infrastructure/mk/bsd.port.mk:2360 'fake')
$ echo $?
1
$ cvs -nq diff -u
Index: Makefile
===================================================================
RCS file: /cvs/ports/archivers/unrar/Makefile,v
retrieving revision 1.58
diff -u -r1.58 Makefile
--- Makefile 10 Mar 2013 22:28:05 -0000 1.58
+++ Makefile 16 May 2013 11:53:38 -0000
@@ -33,6 +33,7 @@
${INSTALL_DATA_DIR} ${PREFIX}/share/doc/unrar
${INSTALL_MAN} ${WRKSRC}/readme.txt ${WRKSRC}/license.txt \
${PREFIX}/share/doc/unrar
+ exit 1
NO_TEST= Yes
$
On Thu, May 16, 2013 at 12:26:38PM +0100, Mikolaj Kucharski wrote:
> On Wed, May 15, 2013 at 03:40:07PM +0100, Stuart Henderson wrote:
> > Could you at least make sure it doesn't break building/packaging them?
> > cpan is a mass of tangled dependencies and often we find that updating
> > one thing means updating a handful of others.
> >
>
> I think all Perl YAML updates which I've sent don't break building nor
> packaging of any dependent ports.
>
> I did following:
>
> echo ">>>>> $pkgpath <<<<<"
> make SUBDIR=$pkgpath clean=depends
> make SUBDIR=$pkgpath clean=packages
> make SUBDIR=$pkgpath package
> ret=$?
> if [ $ret == 0 ]
> then
> echo "MAKE PACKAGE FOR $pkgpath SUCCEEDED"
> else
> echo "MAKE PACKAGE FOR $pkgpath FAILED: $ret"
> fi
> make SUBDIR=$pkgpath clean=depends
>
>
> The list which I went through:
>
>
> devel/p5-YAML
> devel/p5-YAML-Perl
> devel/p5-YAML-Shell
> devel/p5-YAML-Syck
> devel/p5-YAML-Tiny
> devel/p5-YAML-XS
> audio/squeezecenter
> converters/p5-Catalyst-View-JSON
> converters/p5-JSON-Any
> databases/p5-Catalyst-Controller-Rose
> databases/p5-Rose-DB
> databases/p5-SQL-Translator
> databases/ports-readmes-dancer
> devel/p5-B-Deobfuscate
> devel/p5-Config-Any
> devel/p5-Config-Auto
> devel/p5-Data-Random
> devel/p5-Devel-ebug
> devel/p5-File-Attributes
> devel/p5-FormValidator-Simple
> devel/p5-Module-Install
> devel/p5-SDL
> devel/p5-SVN-Web
> devel/p5-TAP-Harness-Archive
> devel/p5-Test-Base
> devel/p5-Test-YAML-Valid
> devel/svk
> mail/p5-Mail-SPF-Test
> multimedia/p5-Data-AMF
> net/p5-OSPF-LSDB
> productivity/davical
> sysutils/ns4
> sysutils/shelldap
> www/p5-Catalyst-Runtime
> www/p5-Dancer
> www/p5-HTML-FormFu
> www/p5-HTTP-BrowserDetect
> www/p5-Plack-Middleware-ReverseProxy
> www/p5-Plagger
>
>
--
best regards
q#
