gump-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Adam R. B. Jack" <>
Subject Re: [nant-dev] NAnt and Gump -- getting closer...
Date Tue, 07 Sep 2004 19:51:27 GMT

> I don't know exactly what you're trying to achieve with Gump, but IMO
> are much easier than you think if you fully utilize NAnt's potential.

Yeah, I think we need to step back and make that clear.

Gump is attempting to emulate a "rabid" OSS developer, perform the steps a
developer would do in order to download and build NAnt, and then others -- 
from scratch. This OSS developer is particulary anal, they don't want
yesterdays build ... they want it fresh, everything the very latest. Their
objective isn't to run this, it is to monitor for changes, disconnects. If
project X breaks project Y the developer wishes to feel this pain. This
automated developer is there to reduce pain for human developers/users, and
to catch interface change pain early! If early enough perhaps the problem
can be resolved before a release is made, and a problems gets into the

Gump is a Python program that interacts with things (CVS, SVN, Shell, Ant,
Maven) and we'd like NAnt. Gump wants to bootstrap NAnt 'cos it was to also
catch NAnt changes --- not just compilation, but runtime. Being an outsider
is important. Gump is Python (not Java) so it is spearate from concerns of
the platform, and this ought make it a good candidate for working on DotNet.

We want to work with NAnt as NAnt ought be worked with, but not inside NAnt,
outside -- as a developer would be. Further, we want to allow existing NAnt
scripts -- as they exist today -- without change. This allows projects to
benefit from Gump without changing how they work. At least, that is what
we'd like.

For more/historical information see:



To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message