changing the autogen.sh approach

Mark Galassi rosalia@lanl.gov
Fri Mar 31 15:57:00 GMT 2000


Dudes, there are two ways of doing autogen.sh files:

(1) autogen.sh just calls libtoolize, aclocal, automake --add-missing,
    and autoconf.  Then you invoke "configure" yourself.

(2) autogen.sh also calls configure and passes it all the options,
    taking care to do the right thing with separate build dirs.

We are currently doing (2), but I would like to switch to doing (1),
which is cleaner and makes the autogen.sh script a total no-brainer
(as it should be).  Does anyone have strong feelings about it?


More information about the Gsl-discuss mailing list