.cvsignore

[Russ Allbery]

Jeffrey M Vinocur <jeff at litech.org> writes:
> On Fri, 1 Mar 2002, Russ Allbery wrote:

>> Does anyone object to my adding appropriate .cvsignore files to the INN
>> repository?  I'm working on a new version of make manifest-check that
>> would benefit from them, not to mention the standard benefits for
>> developers.

> I would, in fact, like it very much.

If anyone has any objections, yell, or I'll probably do it fairly soon.
It's been minorly bugging me for a while and the only reason why I didn't
do it is because I didn't want to distribute .cvsignore files in the
distribution and it would have created a lot more check-manifest noise.
But now that I have a real program to do the manifest checking, that's no
longer a problem.

>> The .cvsignore files would not be included in the distribution
>> tarballs.  They would be included in the snapshots

> Hmm.  Any reason for this?  It shouldn't cause any problems, but we
> remove the CVS directories, right?  So why not .cvsignore as well?

Ah, yeah, good point.  I forgot that we already took care of the CVS
directories; I could handle .cvsignore the same way.

I'd actually like to ship lib/parsedate.c and innfeed/config_y.[ch] as
part of the distribution as well, so that people don't have to have yacc
or bison to build INN, so I'm probably going to hack on the packaging a
bit and also make the snapshots look more like a distribution tarball.

Probably would be easy enough to set up a nightly build of INN around here
too that just builds it and runs the test suite and sends any problems to
some mailing list.  Hm.  I likely won't get to that for a bit, though.

-- 
Russ Allbery (rra at stanford.edu)             <http://www.eyrie.org/~eagle/>