[Mono-dev] Why not NAnt?
charlie at nunit.com
Tue Aug 29 12:32:21 EDT 2006
> >> No one practically uses NAnt on cygwin (it is obvious as
> NAnt never
> >> supports cygwin-considered buid/installation system and no one
> >> complained about that).
> >> Considering that there are many many additional things
> that depends
> >> on make like standalone tests, there is no better development
> >> environment than cygwin on Windows.
> >> Having build depencency just for dogfooding isn't a good idea.
> >> Every time the build got broken somehow, it will mess
> every hackers
> >> until the problem goes away. Have you joined the development and
> >> tasted that pain?
> > Are you making this statement with regard to building Mono or
> > applications that run on Mono?
> For my first paragraph, the second, and the first for everything else.
OK, that makes sense.
> > If it's the first, I agree. For applications, it seems as if
> > standardizing on NAnt is a valid choice to be made by the
> > developers and might even be a preferred choice since it allows one
> > build to work on all platforms.
> Under Mono on windows it is unlikely to work fine with
> installation (installing assemblies to mono GAC) at least
> from what I know of. Of course I might be missing something,
> especially the latest cvs version might work (anyways Gert
> would help you).
Yes... I can see that an install target would be an issue. I've always
avoided having an install target in my NAnt builds, just using them
to build the actual installer in the case of Windows.
I think it would be burdensome to need NAnt if you were merely
installing pre-built assemblies. In the Linux world, an install target
in a makefile is a perfect solution, since make is pretty much
always present. For windows, that isn't as reasonable.
I'll think more about this and add something about it to the
wiki page on cross-platform compatibility.
> Atsushi Eno
> Mono-devel-list mailing list
> Mono-devel-list at lists.ximian.com
More information about the Mono-devel-list