bug-cvs
[Top][All Lists]
Advanced

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

Re: a few Q's


From: Mark D. Baushke
Subject: Re: a few Q's
Date: Sat, 25 Oct 2003 12:42:28 -0700

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

Hi Christos,

It would probably be best to pass your changes to diff to the diffutils
maintainers <address@hidden>. We try to keep our sources in sync
with them. Read the srclist.txt file for more details. However, I can
safely say that any changes to the diff directory that are not real bugs
will not be fixed except thru the diffutils folks.

Support for the vasprintf() function is going to need to be pulled from
the gnulib project, so gnulib/m4/vasprintf.m4 and the associated
implementations for when it is not normally supported on a target host.

Christos Zoulas <address@hidden> writes:

> On Oct 24,  5:18pm, address@hidden (Derek Robert Price) wrote:
> -- Subject: Re: a few Q's
> 
> And here I go, as promised [patch 1]:
> 
> 1. use xasprintf() in most places where it is easily done, and simplify
>    the code in the process. This actually eliminates *a lot* of duplicate
>    string scanning. Don't use the ugly slop code that makes assumptions
>    about the lengths of strings.
> 2. simplify and factor out common repository path manipulation code.
> 3. use const everywhere possible, to make sure that strings don't get
>    unintentionally modified.
> 
> There is a variety of indentation styles in the code. What is the
> sanction style for continuation lines?

It is presently in a bit of flux. See the HACKING file for the goal, but
in the mean time doing as little reformatting to a file is probably the
best idea.

> If you need any more explanations, please let me know. If/when this is
> accepted, I will proceed the second part (setuid changes).

I have not had the opportunity to look thru all of the patch yet and I
suspect that Derek and/or Larry may also want to give it a look too.

> Yes, this is a big diff, but the existing string manipulation in the
> code was ugly and unmaintainable. Using xasprintf() in a lot of places,
> make the test easier to read and make changes to.

Yes, that is certainly one reason I would favor moving to using
xasprintf() myself.

        -- Mark

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (FreeBSD)

iD8DBQE/mtIk3x41pRYZE/gRAt2HAJ9nytu7+GIX8PazsHelt6EbqcKFbgCeIfA7
UTD0gNlanGzqxVhrtCYdhmk=
=urbB
-----END PGP SIGNATURE-----




reply via email to

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