alloca2.test has
grep 'Makefile.am:1:.*AC_PROG_LIBTOOL' stderr
but I think that nowadays LT_INIT is what appears in automake's error
message about lack of libtool?
Cheers,
Patrick
In tests/Makefile.am, there is
XFAIL_TESTS = \
all.test\
auxdir2.test\
cond17.test \
txinfo5.test
and I am getting confused by:
GNU Make 3.81
env VER
On Sat, Oct 28, 2006 at 01:48:23AM +0200, Ralf Wildenhues wrote:
> Second try.
>
> * Patrick Welche wrote on Wed, Oct 25, 2006 at 07:44:21PM CEST:
> > ...
> > Making all in doc
> > gmake[1]: Entering directory `/usr/src/local/automake/doc'
> > ...
&g
On Wed, Oct 25, 2006 at 09:35:49PM +0200, Ralf Wildenhues wrote:
> Never mind what I wrote. Please.
> *blush*
Oops - sequential reading :-/
Time for a pint and a head scratch...
Cheers,
Patrick
On Wed, Oct 25, 2006 at 09:33:34PM +0200, Ralf Wildenhues wrote:
> * Patrick Welche wrote on Wed, Oct 25, 2006 at 07:44:21PM CEST:
> > Building automake, after
> >
> > gmake distclean
> > cvs update
> > sh bootstrap
> > configure
>
> Bzzt. At this p
Building automake, after
gmake distclean
cvs update
sh bootstrap
configure
gmake
...
Making all in doc
gmake[1]: Entering directory `/usr/src/local/automake/doc'
...
config.status: creating Makefile
config.status: creating src/Makefile
config.status: creating config.h
config.status: executing dep
Hello Ralf!
On Sun, Sep 03, 2006 at 09:33:20AM +0200, Ralf Wildenhues wrote:
> I say we drop the test and require that install-sh be executable.
> After all, this is so much easier to do.
>
> OK to apply?
Sounds good (I didn't understand the advantage of non-executable install-sh)
- though you h
In nobase.test:
# Remove the 'x' bits just for fun. (Peter O'Gorman said: "If an
# upstream source tarball does not use the autotools, it is sometimes
# easier to build if it is modified to use them. Then a patch is
# generated, and applied to the unpacked clean sources during the
# build. patch(
On a recent NetBSD/i386 system with today's cvs autoconf/automake:
=== Running test ./aclocal.test
+ pwd
/usr/src/local/automake/tests/testSubDir
+ set -e
+ aclocal-1.9a -Werror --output=fred
+ test -f fred
+ aclocal-1.9a -Werror --unknown-option
+ grep help stderr
aclocal: Try `/usr/src/local/aut
Not many clues here.. highlights:
gmake MAKE=gmake VERBOSE=-x TESTS=tar2.test check
...
checking how to create a pax tar archive... none
...
config.status: creating Makefile
{ test ! -d tar2-1.0 || { find tar2-1.0 -type d ! -perm -200 -exec chmod u+w {} ';' &&
rm -fr tar2-1.0; }; }
mkdir tar2-1.0
On Fri, Mar 05, 2004 at 12:23:50PM +0100, JRBCAST wrote:
> Hi,
>
> I have been trying to disable the -g flag that autoconf uses when
> compiling my GNU project. I have tried --enable-debug=no --disable-debug
> and none works. I have had a look at google and some questions but no
> response...
I h
I seem to have many tests failing eg:
gmake[1]: Nothing to be done for `uninstall'.
gmake[1]: Leaving directory
`/usr/src/local/automake/tests/testSubDir/autohdr3-1.0/_build'
gmake[1]: Entering directory
`/usr/src/local/automake/tests/testSubDir/autohdr3-1.0/_build'
gmake[1]: Leaving directory
On Wed, Jan 14, 2004 at 06:28:55PM +0100, Alexandre Duret-Lutz wrote:
> >>> "Patrick" == Patrick Welche <[EMAIL PROTECTED]> writes:
>
> Patrick> Slowly giving up.. autoconf cvs with Eric Sunshine's
> Patrick> SHELL patch applied, but maybe n
Slowly giving up.. autoconf cvs with Eric Sunshine's SHELL patch applied, but
maybe not propagated to all the executables involved, and automake
cvs of today, NetBSD-1.6ZG/i386, in the automake directory:
% sh bootstrap
Found no shell that has working shell functions.
Please tell [EMAIL PROTECTED
On Sat, Oct 25, 2003 at 11:49:43PM +0200, Alexandre Duret-Lutz wrote:
> >>> "Patrick" == Patrick Welche <[EMAIL PROTECTED]> writes:
> Patrick> All attached.. There is plenty of sed between
> Patrick> "creating Makefile" and "creating s
On Fri, Oct 24, 2003 at 10:17:11PM +0200, Alexandre Duret-Lutz wrote:
>
> Ouch. I was really hoping you had used the wrong version.
> That's a lot less fun suddenly.
Indeed - still no clues here :(
> I see you are using bash (which version?), and I assume you are
> still running NetBSD.
I am s
On Tue, Oct 21, 2003 at 01:54:35PM +0200, Alexandre Duret-Lutz wrote:
> >>> "Patrick" == Patrick Welche <[EMAIL PROTECTED]> writes:
>
> [...]
>
> Patrick> So, might there be a ../ too many?
>
> I don't know. Which version of Autoconf a
With today's cvs:
...
mkdir output7-1.0
/usr/local/bin/bash ./mkinstalldirs output7-1.0/mk
mkdir -p -- output7-1.0/mk
list='mk sub'; for subdir in $list; do if test "$subdir" = .; then :; else test -d
"output7-1.0/$subdir" || mkdir "output7-1.0/$subdir" || exit 1; (cd $subdir &&
make to
On Thu, Aug 21, 2003 at 01:25:29AM +0200, Alexandre Duret-Lutz wrote:
> >>> "Patrick" == Patrick Welche <[EMAIL PROTECTED]> writes:
>
> Patrick> txinfo3.test txinfo13.test txinfo16.test txinfo18.test txinfo22.test
> Patrick> fail for me just because te
txinfo3.test txinfo13.test txinfo16.test txinfo18.test txinfo22.test
fail for me just because texput.log is left behind..
Cheers,
Patrick
(makeinfo (GNU texinfo) 4.6; TeX, Version 3.14159)
On Sat, Jul 05, 2003 at 10:23:42AM +0200, Akim Demaille wrote:
>
> > This seems to make my NetBSD sed unhappy... Thoughts?
>
> I'm installing the following.
...
> - sed -e '1{h;s/./=/g;p;x}' -e '$${p;x}'
> + sed -e '1{h;s/./=/g;p;x;}' -e '$${p;x;}'
==
I am getting
sed: 1: "1{h;s/./=/g;p;x}": extra characters at the end of x command
gmake: *** [distcheck] Error 1
which I think comes from
RCS file: /cvs/automake/automake/lib/am/distdir.am,v
retrieving revision 1.46
retrieving revision 1.47
diff -r1.46 -r1.47
...
348,350c351,353
< $(am__re
On Fri, Dec 13, 2002 at 10:58:43PM +0100, Alexandre Duret-Lutz wrote:
>
> Thank you. I'm checking this in.
>
> 2002-12-13 Alexandre Duret-Lutz <[EMAIL PROTECTED]>
>
> * tests/version7.test: Require GNU make.
...
Thank you!
Patrick
Just got (fresh cvs):
ERROR: files left in build directory after distclean:
./version7-3.141.tar.gz
Haven't looked yet why...
Cheers,
Patrick
On Thu, Nov 14, 2002 at 07:43:36PM +0100, Alexandre Duret-Lutz wrote:
>
> No. This Nestable stuff is only temporary, until further work
> on the way variables and targets are stored is done.
>
> See the following if you want an explanation why we need
> Netstable at the moment.
> http://mail.gnu
On Thu, Nov 14, 2002 at 05:32:49PM +0100, Alexandre Duret-Lutz wrote:
>
> Patrick> v5.6.0, and you guessed right: everything passes/works
> Patrick> happily on a perl v5.8.0rc2 machine.
>
> Thanks! Here is what I've found in the Perl 5.6.1 changelog:
That was quick! Now it has gone past am__i
On Thu, Nov 14, 2002 at 11:48:27AM +0100, Alexandre Duret-Lutz wrote:
> >>> "Patrick" == Patrick Welche <[EMAIL PROTECTED]> writes:
>
> Patrick> With cvs code from now:
> Patrick> % sh bootstrap
> Patrick> Bareword "am__installdirs"
With cvs code from now:
% sh bootstrap
Bareword "am__installdirs" not allowed while "strict subs" in use at ./automake.tmp
line 4742.
BEGIN not safe after errors--compilation aborted at ./automake.tmp line 7682.
# Deal with installdirs target.
sub handle_installdirs ()
{
$output_rules .=
On the computers without TeX installed, all tests pass fine, but the one
with TeX hangs at
PASS: texinfo10.test
[1] + Stopped (tty input)gmake check
(is it running texinfo10, or texinfo13?)
apparently because it is running "tex --version". On my version of TeX,
that gets:
% tex --version
On Mon, Aug 19, 2002 at 11:15:09AM +0200, Alexandre Duret-Lutz wrote:
> On Tue, Aug 13, 2002 at 02:29:47PM +0100, Patrick Welche wrote:
> > On NetBSD, I get a failure for vtexi.test. The odd thing is that the test
> > mentions:
> >
> > # Test for bug reported by Jim
On NetBSD, I get a failure for vtexi.test. The odd thing is that the test
mentions:
# Test for bug reported by Jim Meyering:
# When I ran automake-0.29 on textutils,
# I noticed that doc/Makefile.in had
# textutils.info: textutils.texi
# instead of
# textutils.info: textutils.texi version.tex
On Thu, Jul 25, 2002 at 11:21:31PM +0200, Alexandre Duret-Lutz wrote:
> patrick> - comment6 needs gnu make
> >>
> >> Which make are you using? I'd rather teach Automake to warn
> >> about this construct if it's not portable.
>
> It turns out it was easier to fix the user input rather than t
On Tue, Jul 23, 2002 at 06:32:34PM +0200, Alexandre Duret-Lutz wrote:
>
> What environment is `me'?
Ooops - NetBSD-1.6D/i386
> Cleaning configure.lino in distclean is correct, but not
> autom4te.cache. See the comment about this a few lines below
> the place you patched.
Ah yes..
> patrick
Some "make check" tests failed for me.. Essentially
- libtool3, subobj9, subpkg do a distcheck, which finds autoconf's
configure.lineno left over. Presumably autom4te.cache should also
get distcleaned away?
- comment6 needs gnu make
- still don't know what's wrong with gnits2 and gnits3 (prob
On Tue, Jul 23, 2002 at 04:16:52PM +0100, Patrick Welche wrote:
> On Fri, Jul 19, 2002 at 11:47:36AM +0200, Akim Demaille wrote:
> > >>>>> "mcmahill" == mcmahill <[EMAIL PROTECTED]> writes:
> >
> > mcmahill> I'm ending up with a fil
On Fri, Jul 19, 2002 at 11:47:36AM +0200, Akim Demaille wrote:
> > "mcmahill" == mcmahill <[EMAIL PROTECTED]> writes:
>
> mcmahill> I'm ending up with a file called '=build/configure.lineno'
> mcmahill> after 'make distcheck' which of course causes the distcheck
> mcmahill> to fail. This is
With cvs automake of Jun 14 15:05 GMT:
% ./configure
...
configure: error: Autoconf 2.52 or better is required
bach: {237} autoconf --version
autoconf (GNU Autoconf) 2.53b
...
?!
Cheers,
Patrick
On Sun, Feb 17, 2002 at 06:48:13PM -0700, Tom Tromey wrote:
> > "adl" == Alexandre Duret-Lutz <[EMAIL PROTECTED]> writes:
>
> adl> I managed to reproduce this (on NetBSD 1.5.2), and reduced the
> adl> failure to the following script:
>
> adl> % cat foo.sh
> adl> set -e
> adl> for x in a; do
Scrap my earlier mumblings and the patch! Having been put right again by
Christos:
When this option is on, if a simple command fails for any of the
reasons listed in Consequences of Shell Errors or returns an exit
status value >0, and is not part of the compound list following a
while, until, or
Actually, you really don't want set -e in make.test:
for flag in '' -w; do
MAKE="$save $flag" ./configure
fgrep 'am__include = #' Makefile && exit 1
touch configure.in
$MAKE $flag
fgrep 'am__include = #' Makefile && exit 1
rm -f config.cache
done
With set -e, you're saying
On Thu, Feb 14, 2002 at 04:36:36PM +0100, Alexandre Duret-Lutz wrote:
> I managed to reproduce this (on NetBSD 1.5.2), and reduced the
> failure to the following script:
>
> % cat foo.sh
> set -e
> for x in a; do
>BAR="foo"
>false && echo true
>echo mumble
> done
> % sh -x foo.sh
> +
On Wed, Feb 13, 2002 at 06:54:07PM +0100, Alexandre Duret-Lutz wrote:
>
> Patrick> config.status: creating Makefile
> Patrick> + fgrep am__include = # Makefile
> Patrick> FAIL: make.test
>
> So, what does this Makefile contains?
>
> Could you run
> grep am__include /usr/src/local/automake/t
On Wed, Feb 13, 2002 at 09:31:19PM +0100, Alexandre Duret-Lutz wrote:
> Nope, this code comes from Automake, and `am__include = include'
> is the expected output.
>
> The question is why does
>
> fgrep 'am__include = #' Makefile && exit 1
>
> abort the test?
>
> Can you check how the followi
On Wed, Feb 13, 2002 at 06:03:41PM +0100, Alexandre Duret-Lutz wrote:
> >>> "Patrick" == Patrick Welche <[EMAIL PROTECTED]> writes:
>
> [...]
>
> Patrick> Hints on where to delve/how to fix?
>
> Yep: could you run the same test with VERBOSE=x?
Under NetBSD-1.5ZA/i386, gmake 3.79.1, $MAKE==gmake, automake cvs of 13 Feb:
=== Running test ./make.test
checking for a BSD compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for gawk... no
checking for mawk... no
checking for nawk... no
checking
Sorry about the previous noise: setting MAKE=gmake avoided the problem
(which was that %OBJ% in lib/am/yacc.am was ../parse.c, so
sed s/y\.tab\.c/../parse.c/ wasn't happy)
Cheers,
Patrick
yaccvpath.test is failing for me because lib/am/yacc.am:26 contains
sed '/^#/ s/y\.tab\.c/%OBJ%/' y.tab.c >%OBJ%
and %OBJ% ends up as ../parse.c - the slash isn't escaped..
Cheers,
Patrick
I have automake 1.4s installed, and just tried to make a freshly checked out
cvs copy. After running configure, make gives:
Making all in .
cd . && /usr/src/local/automake/aclocal --acdir=m4
Can't locate Automake/General.pm in @INC (@INC contains:
/usr/local/share/automake /usr/local/lib/perl5/5
During the subdir5.test, a project is configured and run. Later a
subdirectory is added and populated, and the root Makefile.am and
configure.in suitably ammended. Then automake, autoconf, configure are NOT
run, JUST gmake. The idea is to see whether use is made of the Makefile.am
in the subdirect
Having just read
"FYI: TESTS_ENVIRONMENT patch"
over in libtool reminds me that essentially the same trivial problem happens
in automake, but I don't see the solution for a nice neat patch..
On my system, make = BSD make, gmake = GNU make.
For some reason ansi.test fails for me if I
gmake ch
On Tue, Apr 24, 2001 at 12:02:35PM -0400, Pavel Roskin wrote:
> I wrote a couple of days ago:
> > I just did a cvs checkout of automake, all the tests pass, but believe it
> > or not:
...
> This patch has been tested with Perl perl5.005_03 and 5.6.0. Ok to apply?
...
> +++ automake.in
...
> +sub
I just did a cvs checkout of automake, all the tests pass, but believe it
or not:
# automake --version
Automake::register_language() called too early to check prototype at
/usr/local/bin/automake line 741.
Automake::register_language() called too early to check prototype at
/usr/local/bin/autom
52 matches
Mail list logo
