Hi, thanks for the helpful tips! I'm trying to combine both answers I got into a working scheme and I ran into the following question: how do I test if --enable-maintainer-mode is passed to configure? I looked at the autoconf source and found $USE_MAINTAINER_MODE and I am using it, but since I didn't see it documented anywhere I wonder if that's the right way to do this. > Here is an example project that illustrates this. It requires a > `mygen' program to build a source. If the program isn't available > but the source is available, the build will succeed. Otherwise the > user will get an error saying that a required program wasn't found. Would it be considered bad practice to test for the missing tools at configure stage rather than the building stage *if* maintainer mode is enabled? I think it would be more user-friendly if configure could catch that the build will fail rather than wait for the failure itself; I might be spoiled, but I got used to autoconf'ed packages to build without a glitch once configure stage is complete with no errors. :) Andrej _______________________________________________ Autoconf mailing list Autoconf@xxxxxxx http://lists.gnu.org/mailman/listinfo/autoconf
