On Tue, 11 Nov 2003, Alexandre Duret-Lutz wrote: > >>> "Paul" == Paul Eggert <eggert@xxxxxxxxxxx> writes: > > Paul> Pavel Roskin <proski@xxxxxxx> writes: > >> I've seen an announcement in autotools-announce@xxxxxxx that Automake > >> 1.7.9 has been released with Autoconf 2.59.... > >> OK, I give up. I used to be an active co-maintainer of Autoconf, > >> and now I cannot get the latest public release! > > Paul> Clearly Autoconf 2.59 has not been released. I don't know why 2.59 > Paul> was mentioned in the Automake 1.7.9 announcement, nor do I know why > Paul> the Autoconf CVS ChangeLog says "Version 2.59". However, since 2.59 > Paul> has not yet been released, I assume that possibly other fixes will be > Paul> integrated before 2.59 comes sout. > > Since the CVS is tagged, the ChangeLog and NEWS updated, and the > 2.59 diff posted I think it's pretty clear that Akim intended to > release 2.59. (I remember I urged him to, so we don't last with > incorrect abs_* value too long.) Tagging the cvs only occurs at > the end of `make cvs-dist'. It was so clear that I though it > had been announced, hence the mention in Automake's NEWS. > > It's likely that Akim did all this in an hurry (the incorrect > date in NEWS tends to support that) and forgot to upload and > announce it. Beside, lrde.epita.fr had some network problems > lately, so maybe he could simply not upload it. > > I can see the GPG-signed tarballs of Autoconf 2.59 in Akim's > account. But since Akim should be back tonight or tomorrow, I > guess it's better we wait for him. Thanks for clarification. But let's look at the configure script inside ftp://sources.redhat.com/pub/automake/automake-1.7.9.tar.bz2 #! /bin/sh # Guess values for system-dependent variables and create Makefiles. # Generated by GNU Autoconf 2.59 for GNU Automake 1.7.9. Maybe if Autoconf 2.59 has not been released, the real release should be using Automake 1.7.9 for recursiveness :-) Well, it looks like things are not that bad as they seemed this morning. P.S. Wrong autoconf-maintainers address is also my fault. I think it should be correct now. -- Regards, Pavel Roskin