Hello Sam, I'm afraid there is some confusion here. Just for the record, I thought you want to determine the version number in the time when the configure script is running? If you want to achieve that, of course you have to remove the checks: > -AS_LITERAL_IF([$3], [], [m4_warn([syntax], [AC_INIT: not a literal: $3])]) But you also have to face the real problems: grep the autoconf CVS (amd Automake CVS) for occurrences of the AC_PACKAGE_* macros. Then you did changes like this: > -m4_ifndef([AC_PACKAGE_NAME], > - [m4_define([AC_PACKAGE_NAME], [$1])]) > +[m4_ifndef([AC_PACKAGE_NAME], > + [m4_define([AC_PACKAGE_NAME], $1)]) I don't understand what you mean by this. Evaluating the parameter once more by m4 has nothing to do with the fact that you want to execute it later. > well, in this thread 4 persons (Eric, Bob, Bruno, Bruce) concurred with > me [...] Well, they all supported your opinion that there should be a way to store version number in a separate file. Two of them admitted that they are misusing the obsolete syntax of AC_INIT and AM_INIT_AUTOMAKE, in order to achieve that. But I guess that they won't insist on the misfeature that the version number is determined when the configure script runs. Alexandre has offered a more elegant solution then my one. Do you see any problems with that? (Besides the note that m4_normalize should be documented.) Have a nice day, Stepan _______________________________________________ Autoconf mailing list Autoconf@xxxxxxx http://lists.gnu.org/mailman/listinfo/autoconf
