autotools@FreeBSD.org. This is for individuals willing to put in
the hard work to generate patches and work with portmgr as we put
them through -exp runs and work through port breakage that they
create. (A huge number of ports depend on these, so -exp runs are
required.)
Thanks to aDe@ for all the thankless work he's done over the years
to tame these beasts.
Hat: portmgr
Specifically, newer autoconf (> 2.13) has different semantic of the
configure target. In short, one should use --build=CONFIGURE_TARGET
instead of CONFIGURE_TARGET directly. Otherwise, you will get a warning
and the old semantic may be removed in later autoconf releases.
To workaround this issue, many ports hack the CONFIGURE_TARGET variable
so that it contains the ``--build='' prefix.
To solve this issue, under the fact that some ports still have
configure script generated by the old autoconf, we use runtime detection
in the do-configure target so that the proper argument can be used.
Changes to Mk/*:
- Add runtime detection magic in bsd.port.mk
- Remove CONFIGURE_TARGET hack in various bsd.*.mk
- USE_GNOME=gnometarget is now an no-op
Changes to individual ports, other than removing the CONFIGURE_TARGET hack:
= pkg-plist changed (due to the ugly CONFIGURE_TARGET prefix in * executables)
- comms/gnuradio
- science/abinit
- science/elmer-fem
- science/elmer-matc
- science/elmer-meshgen2d
- science/elmerfront
- science/elmerpost
= use x86_64 as ARCH
- devel/g-wrap
= other changes
- print/magicfilter
GNU_CONFIGURE -> HAS_CONFIGURE since it's not generated by autoconf
Total # of ports modified: 1,027
Total # of ports affected: ~7,000 (set GNU_CONFIGURE to yes)
PR: 126524 (obsoletes 52917)
Submitted by: rafan
Tested on: two pointyhat 7-amd64 exp runs (by pav)
Approved by: portmgr (pav)
ports, which then results in an implicit dependency on gawk, with the
obvious bad things happening should gawk be later removed.
Note that because this is an edge case, which most folks will not run
into, and because it doesn't change the default package, I am not
bumping PORTREVISION as suggested in the original patch.
PR: 125101
Who: matthias <dot> andree <at> gmx <dot> de
The versioned autotools are now strictly for building other ports in the tree.
Likewise, the gnu- autotools are for runtime dependencies for IDEs, and others,
where unmodified cross-platform capabilities are desired.
Take back maintainership. I'll be scouring the PR database shortly,
if I miss any, or someone beats me to it, please feel free to reassign
appropriately.
or interest to look after these things any more.
To whomever picks these up, I salute you. I'll reset the relevant
PRs to ports-bugs in a couple of weeks if no-one has taken over
maintainership.
Farewell, autotool hell.
of the autotools project to remove all but the latest versions of
libtool, autoconf, and automake, by 31st December 2004.
Note that autoconf259 and automake18 aren't yet accessible through the
relevant USE_* knobs, but will be as soon as bsd.autotools.mk comes into
existence.
2. Fixed bug that Knu pointed out, which was the bad dirs were captured
in bsd.port.mk (to be fixed RSN, in testing now). Fixed with the magic
of ... symlinks! (I always thought symlinks to . were lame, and they are,
but sometimes they can get you out of a corner. Temporarily.)
3. The question is, will anything break trying to recursively go down the
tree following symlinks? If so, that tool needs to be fixed. A symlink
to '.' (or '..') just plain should not be followed.
Bottom line is, AUTOCONF_DIR, AUTOMAKE_DIR, & ACLOCAL_DIR work again.
As a side effect of this fix, you can now download the entire internet
onto 1 floppy disk.
Put the damn support files in the right place. Modifying $(datadir) was
not the right way to do it. I knew there was something about this port that
bugged me. I am content now. Love is all you need. Give peace a chance.
list by bsd.port.mk insert anti foot-shooting device, which prevents
infinite fork loop when the user defines corresponding USE_XXX in global
make.conf, command line or environment.
Similar devices should probably be inserted into ports that might be inserted
into dependency list by others bsd.foo.mk files (bsd.ruby.mk, bsd.python.mk
and so on.)
since I'm doing most of the updating, and am working on a
port/Mk/bsd.<gnublah>.mk to move some cruft around.
Sponsored by: Mr. Smith and Mr. Wesson. :)
was causing problems with a few ports determining that .c was the suffix
for executables. The fix is to expand it to '*.c | *.C' so that m4(1)
doesn't try to outsmart us.
Noticed by: George W. Dinolt <gdinolt@pacbell.net>
