You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Jeff Trawick <tr...@attglobal.net> on 2003/06/12 20:48:20 UTC
libtool 1.4.2 vs. libtool 1.5 for httpd releases
One problem I know of with 1.5 is that it forces our configure to fail
if some standard C++ autoconf checks fail. Since we don't care about
C++, this isn't cool for Apache users.
Based on discussions on the libtool mailing list, I expect that a future
1.5.x release will allow packages like ours to disable C++ checking (or
require that apps that care about C++ enable such checking). But no
guesses from me about when such a release will become available.
Meanwhile, are there problems with 1.4.2 that our users have experienced?
Maybe there is no win here, but if 1.4.2 has no known problems then
maybe we should stick with that until the use of libtool 1.5 doesn't
bring with it any C++-related checks.
Re: libtool 1.4.2 vs. libtool 1.5 for httpd releases
Posted by Greg Stein <gs...@lyra.org>.
On Thu, Jun 12, 2003 at 03:01:36PM -0400, Jeff Trawick wrote:
> Justin Erenkrantz wrote:
> >On Thu, Jun 12, 2003 at 02:48:20PM -0400, Jeff Trawick wrote:
> >>Meanwhile, are there problems with 1.4.2 that our users have experienced?
> >>
> >>Maybe there is no win here, but if 1.4.2 has no known problems then
> >>maybe we should stick with that until the use of libtool 1.5 doesn't
> >>bring with it any C++-related checks.
> >
> >You need at least 1.4.3 because it adds the sed checks for Solaris that
> >are required.
>
> good info!
>
> >We can have our release script to enforce only with 1.4.3, but I think
> >buildconf should still allow 1.5. -- justin
>
> certainly... I was only concerned with which libtool was used to build
> the release tarballs
The Subversion project does this: autogen.sh (basically, buildconf) allows
1.4.x, but the release tarballs are required to use libtool 1.4.3. Works out
nicely. Not sure if the checks allow 1.5 or not. *shrug*
I'm a big +1 on falling back to 1.4.3 for the releases. The 1.5 stuff breaks
third party modules (like SVN) in certain build environments.
Cheers,
-g
--
Greg Stein, http://www.lyra.org/
Re: libtool 1.4.2 vs. libtool 1.5 for httpd releases
Posted by Jeff Trawick <tr...@attglobal.net>.
Justin Erenkrantz wrote:
> On Thu, Jun 12, 2003 at 02:48:20PM -0400, Jeff Trawick wrote:
>
>>Meanwhile, are there problems with 1.4.2 that our users have experienced?
>>
>>Maybe there is no win here, but if 1.4.2 has no known problems then
>>maybe we should stick with that until the use of libtool 1.5 doesn't
>>bring with it any C++-related checks.
>
>
> You need at least 1.4.3 because it adds the sed checks for Solaris that
> are required.
good info!
> We can have our release script to enforce only with 1.4.3, but I think
> buildconf should still allow 1.5. -- justin
certainly... I was only concerned with which libtool was used to build
the release tarballs
Re: libtool 1.4.2 vs. libtool 1.5 for httpd releases
Posted by Justin Erenkrantz <ju...@erenkrantz.com>.
On Thu, Jun 12, 2003 at 02:48:20PM -0400, Jeff Trawick wrote:
> Meanwhile, are there problems with 1.4.2 that our users have experienced?
>
> Maybe there is no win here, but if 1.4.2 has no known problems then
> maybe we should stick with that until the use of libtool 1.5 doesn't
> bring with it any C++-related checks.
You need at least 1.4.3 because it adds the sed checks for Solaris that
are required.
We can have our release script to enforce only with 1.4.3, but I think
buildconf should still allow 1.5. -- justin
RE: libtool 1.4.2 vs. libtool 1.5 for httpd releases
Posted by Sander Striker <st...@apache.org>.
> From: Jeff Trawick [mailto:trawick@attglobal.net]
> Sent: Thursday, June 12, 2003 8:48 PM
> One problem I know of with 1.5 is that it forces our configure to fail
> if some standard C++ autoconf checks fail. Since we don't care about
> C++, this isn't cool for Apache users.
>
> Based on discussions on the libtool mailing list, I expect that a future
> 1.5.x release will allow packages like ours to disable C++ checking (or
> require that apps that care about C++ enable such checking). But no
> guesses from me about when such a release will become available.
>
> Meanwhile, are there problems with 1.4.2 that our users have experienced?
>
> Maybe there is no win here, but if 1.4.2 has no known problems then
> maybe we should stick with that until the use of libtool 1.5 doesn't
> bring with it any C++-related checks.
Agreed. I was kind of expecting 1.5 would give our users less grief :(
I'll bump back to 1.4.3 for next release I'm RM'ing.
Sander