cvs-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Cvs-dev] Re: cvs-passwd patch


From: Mark D. Baushke
Subject: [Cvs-dev] Re: cvs-passwd patch
Date: Tue, 10 Oct 2006 08:25:45 -0700

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

P J P <address@hidden> writes:

> On Mon, 9 Oct 2006, Mark D. Baushke wrote:
> > Nope. It generates lots of derived objects that will not be availble in
> > my tree and the patch will fail to apply cleanly. In addition, the
> > doc/stamp* files should probably not be sent along as they are not
> > really all that useful.
> 
>    But then, how should I do it? I mean, is there any way, I can avoid
> those extra files. Do I have to 'rm' them, before creating patch
> myself??

I would probably do a 'make clean' before generating a diff between
the virgin sources from the tarball I gave you and your build tree.

As a test, you might try doing the patch operation on a virgin copy of
the sources yourself to see if it works for you. If you have problems
getting your patch to work, then I/we would also have problems.

In general, avoid sending diffs of binary files like *.pdf which are not
even in the repository in any case.

The doc/stamp* files are not a big deal, send them if you wish. They
will end up getting regenerated in any case using the dates of the
modified files as applied in our trees. If they apply cleanly to a
virgin set of sources, they will be okay on our end.

> > The different is that you will get square brackets and the word aborted
> > around an 'error (1, 0, "message");' which is desirable to clients that
> > are looking for errors in the output.
> 
>    Okay, got it working. I was missing, '\[' & '\]' before square
> brackets, in the expected ouput of the tests.

Yes, that can be tricky.

> > There are some other problems with the existing source as well in that
> > you are able to send the -e switch from the client to the server without
> > asking if the server understands that option or not.
> 
>   Hmmn okay! I'll look into it!!

Good.

Remember that you will need/want to do a full 'make check' without
having PWCVSROOT set to ensure that everything we would normally see
works and then a run without it set to ensure that all of your other
tests work correctly.

        Good luck,
        -- Mark
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.4 (FreeBSD)

iD8DBQFFK7t5Cg7APGsDnFERAv2+AJwIiGlwZN9SYj3YgzKj+RVWqBjbzACgqBjZ
54WXXqLYHa4+DRzKK/PKrV0=
=uwtT
-----END PGP SIGNATURE-----




reply via email to

[Prev in Thread] Current Thread [Next in Thread]