[Mono-devel-list] Nant / OJB.NET / Maverick.NET / NVelocity

[Paolo Molaro]

On 10/06/03 Ian MacLean wrote:
> >Yes, the DirectoryScanner.cs code is what holds up the compilation on a
> >mono system.
> >
> >And this is a reported issue, a couple months ago when I was looking
> >into using NAnt, I saw on the mailing list archive a thread from someone
> >saying they got it working, but a compile that took 2 seconds from the
> >commandline with mcs took 57 seconds through NAnt.
> > 
> >
> ok - I must have missed that. However now its great that we know where 
> the issues is and whats causing it.

Please, whoever tracks down the issue, file a bug in bugzilla.ximian.com
so mono can be fixed (and, yes, if nant takes 55 seconds to run a build
it is a bug in mono) or better, provide also the fix.

One of the unwritten rules of the Mono project is to assume the issues
reported by users and programmers are in the mono code. It is too easy
(and lame) to assume the issue is in nant or in other projects. This
doesn't mean that we're not confident in our code or that the programs
that use mono can't be improved: it just means that we get to test
more of the mono codebase thanks to the people that run their
applications on mono. This is very important and one of the things that
made mono successful: widespread testing (in the same way widespread
testing helps making the Linux kernel stable). Sometimes some mono
contributors forget this.

Please, even if NAnt can workaround the specific issue, file a bug with
a possibly simple test case in the mono bugzilla database, so we can
make mono better.

Thanks.

lupus

-- 
-----------------------------------------------------------------
lupus at debian.org                                     debian/rules
lupus at ximian.com                             Monkeys do it better