devel/cscout - failes: __nofail__

[Ion-Mihai Tetcu]

On Thu, 17 Jul 2008 18:04:08 -0400
Joe Marcus Clarke <marcus at marcuscom.com> wrote:

> On Fri, 2008-07-18 at 00:57 +0300, Ion-Mihai Tetcu wrote:
> > On Thu, 17 Jul 2008 17:50:28 -0400
> > Joe Marcus Clarke <marcus at marcuscom.com> wrote:
> > 
> > > On Fri, 2008-07-18 at 00:36 +0300, Ion-Mihai Tetcu wrote:
> > > > Hi,
> > > > 
> > > > 
> > > > For this port I get in postPortBuild
> > > >  "${STATUS}" != "SUCCESS" -a "${STATUS}" != "DUD"
> > > > but build_ports.Last_Fail_Reason = nofail
> > > > AND I'm building with -plistcheck
> > > > 
> > > > Any hints why this behaviour? It seems inconsistent to me.
> > > 
> > > In this case, the port status should be LEFTOVER, but the build
> > > was successful.  Admittedly, the fail reason should probably be
> > > PLIST in this case.  But you can just as easily check for a
> > > status of LEFTOVER, and override the fail reason.
> > 
> > Hmm, OK, but why I'm getting nofail only for this port? For all
> > other ports it seems to be plist or mtree or install error.
> 
> Ah, I see.  The problem is that the leftovers are in ${PREFIX}/etc
> which is ignored by plistcheck (in buildscript).  This was probably a
> safety net used on pointyhat to appease some critical ports.
> 
> I'm hesitant to take it out since it's still in pointyhat.  But you
> could remove the check, and see what impact it has.  If ports are
> well-behaved now, it is probably time to remove the check from
> pointyhat.

I'll try to find the time for it, thanks.

> > > > The only exotic thing I see is, ie:
> > > >  > grep cscout_incs.h /usr/ports/devel/cscout/pkg-plist
> > > > %%PORTEXAMPLES%%%%EXAMPLESDIR%%/.cscout/cscout_incs.h
> > > > @unexec if cmp -s %D/etc/cscout/cscout_incs.h
> > > > %D/etc/cscout/cscout_incs.FBSD_GCC ; then  rm -f
> > > > %D/etc/cscout/cscout_incs.h; fi
> > > 
> > > If you're asking as to why there are leftovers, my guess is the
> > > two files don't have the same contents.
> > 
> > Nah, that's maintainer's business and the BotMail was sent ;-)
> 
> Actually, it's much easier than that.  The problem is that he's
> deleted the original files before doing the cmp test.  If he moved
> the two cmp tests to line 5, it would work.

Yeh.


-- 
IOnut - Un^d^dregistered ;) FreeBSD "user"
  "Intellectual Property" is   nowhere near as valuable   as "Intellect"
FreeBSD committer -> itetcu at FreeBSD.org, PGP Key ID 057E9F8B493A297B
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 195 bytes
Desc: not available
URL: <http://marcuscom.com/pipermail/tinderbox-list/attachments/20080718/66fb3b53/attachment.bin>