Subject: pkg/36259: sysutils/xentools30 has test == portability problems
To: None <email@example.com, firstname.lastname@example.org,>
From: None <kre@munnari.OZ.AU>
Date: 05/02/2007 07:35:00
>Synopsis: sysutils/xentools30 has test == portability problems
>Arrival-Date: Wed May 02 07:35:00 +0000 2007
>Originator: Robert Elz
>Release: NetBSD 3.99.15 (pkgsrc current within past 12 hours)
Prince of Songkla University
System: NetBSD jade.coe.psu.ac.th 3.99.15 NetBSD 3.99.15 (GENERIC-1.696-20060125) #8: Wed Jan 25 04:59:39 ICT 2006 email@example.com:/usr/obj/current/kernels/JADE_ASUS i386
With PKG_DEVELOPER set (as is the default when using pkg_comp,
which is what I do), pkgsrc checks scripts in packages for
use of (truly idiotic) "test x == y" type commands that are
apaprently becoming common in some mutant universe.
sysutils/xentools30 is (again) failing that test.
"Again" as an earlier PR was files against an earlier version
of this package. The solution at the time was to simply
turn off the tests (the scripts in question, then, not being
used or installed on NetBSD anyway).
That now looks to have been a mistake - much better to fix the
scripts, and send the fixes back upstream, so the turkeys who
perpetrate this nonsense can learn how test should be used,
and this problem may decrease over time, instead of magnifying.
Attempt to build sysutils/xentools30 with PKG_DEVELOPER set
(why don't all pkgsrc developers simply always have PKG_DEVELOPER
set in their mk.conf files?)
Expect to see ...
=> Returning to build of xentools30-3.0.4
===> Overriding tools for xentools30-3.0.4
===> Extracting for xentools30-3.0.4
===> Patching for xentools30-3.0.4
=> Applying pkgsrc patches for xentools30-3.0.4
===> Creating toolchain wrappers for xentools30-3.0.4
===> Configuring for xentools30-3.0.4
=> Fixing hardcoded paths.
=> Checking for portability problems in extracted files
ERROR: [check-portability.awk] => Found test ... == ...:
ERROR: [check-portability.awk] xm-test/ramdisk/make-release.sh: if [ "$1" == ""
ERROR: [check-portability.awk] xm-test/ramdisk/make-release.sh: elif [ "$1" == "
[gratuituous noise omitted here...]
I don't know if the script detected (this time) is an important
one, and really does need to be fixed, or just more noise that
pkgsrc could be told to not bother checking, but I'd really
strongly suggest fixing the problem, and never ignoring it!