Re: 341-gary-libtoolize-recognise-LTDL_INIT

[Gary V. Vaughan]

Hi Eric,

Thanks for the review.

On 26 Jan 2008, at 02:58, Eric Blake wrote:
> According to Gary V. Vaughan on 1/24/2008 9:43 PM:
> | Okay to commit?
>
> The patch looks okay, and makes the situation is better than it was,  
> so I
> think it is worth committing.

Committed.

>  However,...
>
> |   @@ -921,7 +921,7 @@
> |       func_echo "and rerunning libtoolize."
> |          fi
> |        elif test -z "$m4dir"; then
> |   -      if $opt_ltdl && test "${ltdlprefix}m4" != "$m4dir"; then
> |   +      if test "${ltdlprefix}m4" != "$m4dir" && $seen_ltdl ||
> $opt_ltdl; then
> |       acmacrodir="${ltdlprefix}m4"
> |          else
> |       acmacrodir="$aclocaldir"
>
> Technically, we only need to use ${ltdlprefix}m4 if we are  
> distributing
> ltdl as a subproject, because that is the only mode where the ltdl
> directory must be self-contained.  It would be nice if nonrecursive  
> and
> recursive modes would cooperate with an explicit $aclocaldir (and  
> silence
> the recommendation about setting AC_CONFIG_AUX_DIR([$ 
> {ltdlprefix}config]),
> since these two modes imply that ltdl is sharing infrastructure with  
> the
> rest of the project, rather than the current scenario of the rest of  
> the
> project having to conform to libtool's directory structure.

Coming in the next patch in the series...

> |   +# _LT_AT_LTDL_SETUP
> |   +# -----------------
> |   +m4_pushdef([_LT_AT_LTDL_SETUP],
>
> Even though this is an internal factorization, a one-liner comment  
> stating
> that prior to the actual definition would be helpful.


This is a common idiom in the libtool testsuite.  None-the-less, I  
added a
one line comment to this one occurrence before I committed.

Cheers,
        Gary
--
  ())_.              Email me: address@hidden
  ( '/           Read my blog: http://blog.azazil.net
  / )=         ...and my book: http://sources.redhat.com/autobook
`(_~)_

PGP.sig
Description: This is a digitally signed message part