FYI: use `&&' to separate commands in the untar process [libtool--release--2.0--patch-56]

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Applied to branch-2-0 (I didn't make this oversight in the other branches). * looking for [EMAIL PROTECTED]/libtool--release--2.0--patch-55 to compare with * comparing to [EMAIL PROTECTED]/libtool--release--2.0--patch-55 M libltdl/Makefile.am

Re: Remove double negation error messages.

Quoting Ralf Wildenhues ([EMAIL PROTECTED]): > Thanks for reporting this. I agree that double negation is suboptimal. > However, since the upper bound is chosen on rather pragmatic grounds > (we are pretty sure that five digits are portable), I'd rather not set > it in stone. In doubt we might r

Re: patch to fix missing quoting

Ralf Wildenhues wrote: Sorry for self-reply. OK, better patch: does the quoting only in one place, plus complains loudly and fails if $libobj contains special characters. Works for me. -- --Per Bothner [EMAIL PROTECTED] http://per.bothner.com/

'make clean' fails if missing testsuite

'make clean' fails if there is no tests/testsuite directory. (This is cvs libtool.) I have an attached a patch for Makefile, but unfortunately the 'clean-local' rule comes from automake (1.9.3, in my case). -- --Per Bothner [EMAIL PROTECTED] http://per.bothner.com/ --- Makefile~ 2005-03-1

FYI: Fix copy-with-tar invocations [libtool--release--2.0--patch-55]

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Applied to branch-1-4, branch-2-0 and HEAD. * looking for [EMAIL PROTECTED]/libtool--release--2.0--patch-54 to compare with * comparing to [EMAIL PROTECTED]/libtool--release--2.0--patch-54 M libltdl/Makefile.am M libtoolize.m4sh M Change

enhance PGI pgf77 shared linking

We don't want MAIN_ to be referenced in a libtool-created Fortran library, linking into a C `main' program would fail without a stub. OK to apply this patch (all branches) which makes the HEAD testsuite pass for pgcc/pgCC/pgf77 (branch-1-5 testsuite is insufficient w.r.t f77)? BTW, many thanks to

Re: Remove double negation error messages.

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ralf Wildenhues wrote: | | * config/ltmain.m4sh (func_mode_link): Reword double negation in | error message. Reported by Jesper Louis Andersen <[EMAIL PROTECTED]>. | | 0|[[1-9]]|[[1-9]][[0-9]]|[[1-9]][[0-9]][[0-9]]|[[1-9]][[0-9]][[0-9]][[0-

Re: patch to fix missing quoting

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ralf Wildenhues wrote: | | 2005-03-06 Per Bothner <[EMAIL PROTECTED]>, | Ralf Wildenhues <[EMAIL PROTECTED]> | | Allow special characters in source file names. Will break unless | `-o target' is also given, which must not cont

Re: FYI: HEAD fails stress test on darwin

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ralf Wildenhues wrote: |>If you intended to test this then we need to look at making this test |>conditional on OS. | | | Nope, removing this. I've checked in the following trivial patch. | And I just reran the test and it passes, thank you. Peter - --

FYI: weed out AM_CPPFLAGS

Applied to HEAD and branch-2-0. (we need `-I. -I$(srcdir)' because the ltdl user might have put nostdinc as global Automake option.) This broke compilation with Portland's pgcc. Regards, Ralf * libltdl/Makefile.am: $(builddir) is not defined, replace by `.'. Index: libltdl/Makefile.am

Protect your possessions with Personal Home Inventory.com

Visit us online at www.PersonalHomeInventory.com. If you do not wish to receive email from us, simply respond to this email with

Re: Remove double negation error messages.

Hi Jesper, * Jesper Louis Andersen wrote on Thu, Mar 10, 2005 at 11:56:14AM CET: > $echo "$modename: CURRENT \`$current' is not a nonnegative integer" > > (grep for it, there are a couple of places) > > Should rather be something along the lines of: > > $current must be a positive integer less

Re: patch to fix missing quoting

Sorry for self-reply. * Ralf Wildenhues wrote on Wed, Mar 09, 2005 at 06:53:07PM CET: > > OK. I've got something half-working here. It allows merely adding > weird characters to the source file names. You absolutely have to > specify a "nice" output name with `-o'. Maybe libtool should refuse

RE: IRIX 6.5 and exporting symbols.

Albert Chin wrote: > On Wed, Mar 09, 2005 at 02:51:31PM +0100, Ralf Wildenhues wrote: >> * Peter Ekberg wrote on Wed, Mar 09, 2005 at 02:11:40PM CET: >>> This makes it work for me, the namespace is sadly polluted, >>> but there is nothing we can do about that, short of hacking >>> the gcc specs fil

RE: IRIX 6.5 and exporting symbols.

Ralf Wildenhues wrote: > Michael -- this might be of interest for you. Quoting > generously since > the ML archives don't provide me with a link yet. > *snip* > Preprocess a file with __GNUC__ (and __GNUC_MINOR__ if necessary). > Maybe we can abandon that idea in favor of just running the linker