Re: xvasprintf: don't depend directly on xalloc-die

[Simon Josefsson]

Sorry, it seems the xalloc-die logic has changed, so my patch may be
wrong.  Specifically:

2005-07-15  Paul Eggert  <address@hidden>

        * modules/xalloc (Depends-on): Add xalloc-die.
        * modules/xvasprintf (Depends-on): Add xalloc-die.

I can't find the discussion behind this change.  Why was a hard
dependency on xalloc-die added?  Wasn't the point of separating
xalloc-die from the xmalloc module to allow applications to supply its
own xalloc_die function?  Right now I can't avoid the gnulib
xalloc_die implementation in my library, and I want my own
die-function.

Thanks!

Simon Josefsson <address@hidden> writes:

> Just like xmalloc doesn't have a hard dependency on xalloc-die, I
> don't think xvasprintf should have.  Will install in a few days unless
> someone objects.  (In Shishi, I use a custom die-function, but
> xvasprintf drag in the gnulib defined one... ungood.)
>
> 2005-08-10  Simon Josefsson  <address@hidden>
>
>       * modules/xvasprintf (Depends-on): Don't force xalloc-die.
>
> --- xvasprintf        16 Jul 2005 21:29:14 +0200      1.4
> +++ xvasprintf        10 Aug 2005 16:57:14 +0200      
> @@ -9,7 +9,6 @@
>  
>  Depends-on:
>  vasprintf
> -xalloc-die
>  
>  configure.ac:
>