Re: [PATCH] Avoid false failures caused by filesystem interaction

[Ralf Wildenhues]

Hi Charles,

* Charles Wilson wrote on Fri, Jun 25, 2010 at 06:06:30AM CEST:
> * tests/dlloader-api.at (main.c:first_open): Use uglified

Please mention the test name in parentheses.  I'm not sure how to deal
with the function names within, partly I think we've just omitted all
that if it's very short, or separate with space or colon or so.  No big
deal either way.

> name when "opening" pseudo-module.
> (main.c:last_open): Ditto.
> (main.c:main): Ditto.
> (expout): Update to reflect uglified psuedo-module names.

Typo psuedo.

> 
> Signed-off-by: Charles Wilson <...>
> ---
> This is an alternate solution to the problem addressed here:
> http://lists.gnu.org/archive/html/libtool-patches/2010-06/msg00113.html

Thanks for providing an alternative.  This is fine with me if it works,
and if Peter has no objection; I'm assuming waiting 72 hours from when
you posted the patch should be safe.

> +  /* Use a magic string to avoid possible interactions with filesystem
> +   * objects. Prepend '/' to short-circuit libltdl's search of

In very nit-picky mode, the GCS wants to have two spaces after an
end-of-sentence period.  No big deal either of course.

> +   * $shlibpath_var (e.g. PATH, LD_LIBRARY_PATH, or similar).
> +   */

Cheers,
Ralf