On Jun 28, 2000, Mo DeJong <[EMAIL PROTECTED]> wrote:
> On 28 Jun 2000, Alexandre Oliva wrote:
>> On Jun 28, 2000, Mo DeJong <[EMAIL PROTECTED]> wrote:
>>
>> > Ok, then why don't we just run AC_CHECK_TOOL for ranlib
>> > and any other programs that might be needed all the time?
>>
>> Because there are lots of programs out there that don't use/need
>> RANLIB.
> That is not logical. If automake ever needs ranlib then
> it should look for it with AC_CHECK_TOOL.
I'm not sure I agree with that. It sounds perfectly reasonable to me
to have automake request for the user to add AC_PROG_RANLIB to
configure.in.
>> > Libtool must run these checks because I don't get an
>> > error like this when using a libtool enabled package.
>>
>> It does.
> Then libtool does it correctly and automake is broken.
Libtool always looks for RANLIB, because the whole point of libtool is
to create libraries, including static ones. Automake, OTOH, is used
by a lot of programs that don't create any libraries at all; even by
programs that don't do any compilation. Checking for ranlib in such
situations would be a source of complaints, just like the
unconditional inclusion of `depcomp' was (or may still is, I'm not
sure it's already been fixed).
--
Alexandre Oliva Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist *Please* write to mailing lists, not to me
