Re: Macro writing

2003-06-07 Thread Steve M. Robbins
On Thu, Jun 05, 2003 at 11:00:23AM +0200, Mattias Brändström wrote: > Hello again! > > Now I have a revised version of my macro. If anyone have the time to say > anything about it it would be greatly appricated! > > AC_DEFUN([RKC_XERCES], [ > > AC_ARG_WITH(xerces-prefix, [ --with-xerces-prefix=

Re: Library path with AC_CHECK_LIB

2003-06-07 Thread moseley
On Sat, Jun 07, 2003 at 02:23:15AM -0400, Assar Westerlund wrote: > Bill Moseley <[EMAIL PROTECTED]> writes: > > I was building our package on a Solaris 2.6 machine and ended up with: > > > > ld: fatal: library -lz: not found > > > > On this machine libz is in /usr/local/lib. > > > I can only

Re: Library path with AC_CHECK_LIB

2003-06-07 Thread Assar Westerlund
[EMAIL PROTECTED] writes: > I looked and didn't find anything telling. All I see is this: > > configure:11703: checking for gzread in -lz > configure:11734: gcc -o conftest -g -O2 -Imaybe/include -Lmaybe/lib > conftest.c -lz -lm >&5 > configure:11737: $? = 0 > configure:11740: test -s confte

Re: Using AC_C_INLINE blows up C++ tests

2003-06-07 Thread Assar Westerlund
Bob Friesenhahn <[EMAIL PROTECTED]> writes: > If a configure script uses AC_C_INLINE to test the C compiler, and > then later switches the current language to C++ in order to perform > C++ tests, the C++ tests will fail if AC_C_INLINE re-defines 'inline' > to some other string. > > It would be use

Re: Using AC_C_INLINE blows up C++ tests

2003-06-07 Thread Bob Friesenhahn
On 7 Jun 2003, Assar Westerlund wrote: > Bob Friesenhahn <[EMAIL PROTECTED]> writes: > > If a configure script uses AC_C_INLINE to test the C compiler, and > > then later switches the current language to C++ in order to perform > > C++ tests, the C++ tests will fail if AC_C_INLINE re-defines 'inli

menu driven configure-script

2003-06-07 Thread Mikael Hansen
Why not let autoconf create a configure-script with the possibilbity to set all the options via a menu (like "make menuconfig" for the linux kernel)? /Mike

Re: menu driven configure-script

2003-06-07 Thread Philip Graham Willoughby
On 2003-06-07 19:05:47 +, Mikael Hansen wrote: > Why not let autoconf create a configure-script with the possibilbity to > set all the options via a menu (like "make menuconfig" for the linux kernel)? Configure scripts need to run on many more platforms than the linux kernel config program. I

Re: Using AC_C_INLINE blows up C++ tests

2003-06-07 Thread Assar Westerlund
Bob Friesenhahn <[EMAIL PROTECTED]> writes: > The main Autoconf issue is that each feature test program includes all > the #defines generated from any previous test. This means that if > AC_C_INLINE has been executed previously, then a define like > > #define inline __inline > > may be automatic

Re: Using AC_C_INLINE blows up C++ tests

2003-06-07 Thread Bob Friesenhahn
This patch looks like just what the doctor ordered. Bob On 7 Jun 2003, Assar Westerlund wrote: > Bob Friesenhahn <[EMAIL PROTECTED]> writes: > > The main Autoconf issue is that each feature test program includes all > > the #defines generated from any previous test. This means that if > > AC_C_

Re: menu driven configure-script

2003-06-07 Thread Richard Dawe
Hello. Mikael Hansen wrote: > > Why not let autoconf create a configure-script with the possibilbity > to set all the options via a menu (like "make menuconfig" for the linux > kernel)? Which options? --with-*, --enable-*? Or do you mean "override what configure detects"? Bye, Rich =] -- Rich

Re: menu driven configure-script

2003-06-07 Thread Jose Roman Bilbao
It would be great to have such an option as it is really uncomfortable to write sometimes four or five lines to run configure properly with the desired options. Roman. El sáb, 07-06-2003 a las 21:36, Richard Dawe escribió: > Hello. > > Mikael Hansen wrote: > > > > Why not let autoconf create a

Re: Library path with AC_CHECK_LIB

2003-06-07 Thread Bill Moseley
On Sat, Jun 07, 2003 at 12:25:14PM -0400, Assar Westerlund wrote: > [EMAIL PROTECTED] writes: > > I looked and didn't find anything telling. All I see is this: > > > > configure:11703: checking for gzread in -lz > > configure:11734: gcc -o conftest -g -O2 -Imaybe/include -Lmaybe/lib > > confte

Re: Using AC_C_INLINE blows up C++ tests

2003-06-07 Thread Kevin Ryde
Assar Westerlund <[EMAIL PROTECTED]> writes: > > +AH_VERBATIM([inlind], [ > +#ifndef __cplusplus > +]) It's probably cleaner to use an AH_VERBATIM for the whole thing. I've has some joy from th following (and no description are in the AC_DEFINEs). AH_VERBATIM(inline, [/* Define as `__inline' if

Re: Using AC_C_INLINE blows up C++ tests

2003-06-07 Thread Kevin Ryde
Assar Westerlund <[EMAIL PROTECTED]> writes: > > Wouldn't that require autoconf to write multiple config.h for multiple > languages? That probably has to be the way to go eventually. But if there's only a few conflicts between C and C++ then maybe often they could share, like now. "const" from A

Getting a world distributable configure script....

2003-06-07 Thread Calvin Arndt
I use autoconf/make versions 2.52/1.6.1 The configure script created by the build system absolutely and completely fails on systems with 2.13/1.4. In addition my configure.in fails when used with automake 1.7 Are there any tip or tricks out there for creating a Build System that actually ha

Re: Getting a world distributable configure script....

2003-06-07 Thread Assar Westerlund
Calvin Arndt <[EMAIL PROTECTED]> writes: > I use autoconf/make versions 2.52/1.6.1 > > The configure script created by the build system absolutely and > completely fails on systems with 2.13/1.4. In addition my > configure.in fails when used with automake 1.7 This is confusing. The generated