On Sat, 2007-09-01 at 02:13 +0100, Mark Cave-Ayland wrote: > > I think the issue is the test against $SWIG. I changed it to this: > > > > if test -n x"$SWIG"x; then > > use_swig=true > > fi
Hi Charlie, Having looked at this commit, I now realise it's wrong :( According to the man page, -n tests for return string length and according to the macro, SWIG is set to an empty string. However, the resulting string would then become 'xx' which would still evaluate as true since it has a length greater than zero. The attached patch fixes this, and does a little bit of tidying up at the same time. ATB, Mark. -- ILande - Open Source Consultancy http://www.ilande.co.uk
Index: configure.in =================================================================== --- configure.in (revision 2019) +++ configure.in (working copy) @@ -157,15 +157,13 @@ use_swig=false if test x"$use_python" = xtrue || test x"$use_ruby" = xtrue; then AC_PROG_SWIG(1.3.28) - if test -n x"$SWIG"x; then - use_swig=true + if test x"$SWIG" != "x"; then + SWIG_ENABLE_CXX + AC_SUBST(SWIG) + + use_swig=true fi fi - -if test x"$use_swig" = xtrue; then - SWIG_ENABLE_CXX - AC_SUBST(SWIG) -fi AM_CONDITIONAL(ENABLE_SWIG, [ test x"$use_swig" = xtrue ]) @@ -175,8 +173,8 @@ if test x"$use_python" = xtrue; then dnl Check for Python - AM_PATH_PYTHON - SWIG_PYTHON + AM_PATH_PYTHON + SWIG_PYTHON if test x"$PYTHON" = "x"; then use_python=false
_______________________________________________ geos-devel mailing list geos-devel@geos.refractions.net http://geos.refractions.net/mailman/listinfo/geos-devel
