On Jun 27, 2016 9:44 AM, "Ruediger Pluem" <rpl...@apache.org> wrote:
>
>
>
> On 06/27/2016 04:35 PM, William A Rowe Jr wrote:
> > On Mon, Jun 27, 2016 at 9:25 AM, Ruediger Pluem <rpl...@apache.org
<mailto:rpl...@apache.org>> wrote:
> >
> >
> >     On 06/27/2016 03:45 PM, wr...@apache.org <mailto:wr...@apache.org>
wrote:
> >     > Author: wrowe
> >     > Date: Mon Jun 27 13:45:02 2016
> >     > New Revision: 1750335
> >     >
> >     > URL: http://svn.apache.org/viewvc?rev=1750335&view=rev
> >     > Log:
> >     > Ensure not-selected means 'no', once an APACHE_MODULE enable_foo
is processed
> >     >
> >     > Modified:
> >     >     httpd/httpd/trunk/acinclude.m4
> >     >
> >     > Modified: httpd/httpd/trunk/acinclude.m4
> >     > URL:
http://svn.apache.org/viewvc/httpd/httpd/trunk/acinclude.m4?rev=1750335&r1=1750334&r2=1750335&view=diff
> >     >
==============================================================================
> >     > --- httpd/httpd/trunk/acinclude.m4 (original)
> >     > +++ httpd/httpd/trunk/acinclude.m4 Mon Jun 27 13:45:02 2016
> >     > @@ -375,6 +375,8 @@ AC_DEFUN([APACHE_MODULE],[
> >     >              "$force_$1" != "no" ; then
> >     >        enable_$1=$module_default
> >     >        _apmod_extra_msg=" ($module_selection)"
> >     > +  else
> >     > +     enable_$1=no
> >
> >     What if enable_$1 is set to shared by the the user of ./configure?
Wouldn't that get overriden here?
> >
> >
> > The valid cases of 'yes', 'shared', 'static', 'no', 'most' (??),
'maybe-all' (??), are all handled
> > in the various if/elif cases above this change.
>
> I don't see the handling of 'shared' above, but maybe I missed it. Hence
my question. I see static, yes, maybe-all,
> most, but not shared.

My eyes playing tricks on me, you are right.

Running some various tests, hope to complete the fix this afternoon once I
come up with any other obscure edge cases.

> > The edge case of 'few' and 'reallyall' are not handled above, and we
perhaps need to account
> > for these? In any case, the old logic seems to always result in
build-shared for unexpected
> > values, no matter how silly they are (e.g. 'none').
> >
>
> Agreed that it should be fixed.

I still don't understand what --enable-modules=few (etc) are even supposed
to mean.

Reply via email to