Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Received-SPF: neutral (herse.apache.org: 195.212.29.139 is neither permitted
 nor denied by domain of mark.hindess@googlemail.com)
Message-Id: <200701111400.l0BE0khd020760@d06av02.portsmouth.uk.ibm.com>
In-reply-to: <636fd28e0701101323l67f272e5s220ba234ac1764ba@mail.gmail.com> 
References: <636fd28e0701091724k10cf61f5y2899258ebfd55632@mail.gmail.com>
 <2c9597b90701100534j8101758pb3300863642cad84@mail.gmail.com>
 <F30CF05D-5CC8-4E15-B5C5-EBEDE5E30A8A@pobox.com>
 <636fd28e0701101323l67f272e5s220ba234ac1764ba@mail.gmail.com>
Comments: In-reply-to "Alex Blewitt" <alex.blewitt@gmail.com>
   message dated "Wed, 10 Jan 2007 21:23:35 +0000."
From: Mark Hindess <mark.hindess@googlemail.com>
To: dev@harmony.apache.org, "Alex Blewitt" <alex.blewitt@gmail.com>
Subject: Re: [OT] Attempting to build classlibs on a Mac 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Thu, 11 Jan 2007 14:00:47 +0000


On 10 January 2007 at 21:23, "Alex Blewitt" <alex.blewitt@gmail.com> wrote:
> On 10/01/07, Geir Magnusson Jr. <geir@pobox.com> wrote:
> >
> > On Jan 10, 2007, at 8:34 AM, Alexei Zakharov wrote:
> >
> > > Alex :
> >
> > >> In any case, when compiling this on a Mac, it was
> > >> falling over in the Ant build looking for an "excludes.Mac OS
> > >> X.ppc.ibm" file. I had to copy one of the existing ones over just to
> > >> move past that hurdle.
> > >
> > > This is because nobody else was trying to run tests on Mac. In your
> > > case you just need to create an empty file with such name if you want
> > > to run SegmentTest. BTW we need to add such files (probably as well as
> > > "excludes.Mac OS X.x86.ibm" "excludes.Mac OS X.x86.drlvm") to every
> > > module if we want to run tests under MacOS.
> >
> > That's going to be a barrier for people.  Can we be optimistic rather
> > than pessimistic, and change the build to use, in the absence of the
> > specific platform file, a "generic" or "base" exclude list?
> >
> > (We talked about this a while ago, I think...)
> 
> Should probably be additive, too, rather than replace it. For example,
> the SegmentTest was commented out in all platforms. That would have
> been better solved by having a generic/base ommission of SegmentTest,
> whilst still allowing a platform-specific excludes file that allows
> extra tests to also be excluded. But I don't believe you should have
> to re-list SegmentTest in the platform-specific excludes file just
> because it was in the platform-agnostic excludes file when it was
> written.

+1

I'm looking at porting to freebsd/ia32, linux/s390, linux/s390x, aix/ppc32,
and aix/ppc64 - keeping these exclude list up to date with 
platform-agnostic issues is going to get tedious pretty fast.

-Mark.