autoupdate test failure

I don't know anything about autoconf, but FYI, I just CVS updated autoconf and: = ./debug-99.sh: Testing autoupdate = ./tools.m4:180: testing... --- - Mon Jul 3 23:38:53 2000 +++ stdout Mon Jul 3 23:38:53 2000 @@ -1,6 +1 @@

sed prob in foreign.at

What is foreign.at:24: AT_CHECK([libtoolize | sed -n ["s,^.*\`\(/[^']*\)'.*,\1,p"] >at-path], 0, [], ignore) trying to do? For me: quartz% libtoolize --version libtoolize (GNU libtool) 1.3c and libtoolize --version | sed -n "s,^.*\`\(/[^']*\)'.*,\1,p" gets me strictly nothing, so

Re: sed prob in foreign.at

On Thu, Jan 18, 2001 at 11:14:50AM +0100, Akim Demaille wrote: > > Could you run > > cd tests/ > ./testsuite -d -v -x 173 > # it should fail, but run this anyway: > libtoolize 2>/dev/null quartz% ./testsuite -d -v -x 172 Testing suite for GNU Autoconf 2.49c =

AC_F77_WRAPPERS prob

Just cvs updated, and this time got a new: Testing suite for GNU Autoconf 2.49c 69. ./syntax.at:27: testing AC_F77_WRAPPERS... + cat + set +vx ./syntax.at:27: autoconf --autoconf-dir .. -l $at_srcdir + at_status=0 + set +v

Re: AC_F77_WRAPPERS prob

On Thu, Jan 18, 2001 at 05:06:19PM +0100, Akim Demaille wrote: > > Arg, answered too fast, didn't pay attention to the fact you were > presenting a different failure :*( > > Expect a libtool failure, but any other failure is wrong. But if you > used `-x'. I don't know what I'm doing! So, this

Re: AC_F77_WRAPPERS prob

On Thu, Jan 18, 2001 at 05:30:13PM +0100, Akim Demaille wrote: > Patrick> I don't know what I'm doing! So, this time without -x (and it > Patrick> passed yesterday..): > > Do you still have yesterday's Autoconf? If so, it might help us > understanding what happened by diffing the two outputs on

Re: AC_F77_WRAPPERS prob

On Fri, Jan 19, 2001 at 03:11:13PM +0100, Akim Demaille wrote: > Patrick, could you try again? There were problem with AC_LANG which > might be responsible. Compared to the two days old Autoconf, the only > significant difference I see is: Yes, you got it! All 175 tests were successful (36 ski

Re: AC_F77_WRAPPERS prob

Sorry I disappeared for a while.. I just ran the old AC_F77_WRAPPERS test with source checked out just now, and it fails once again (rather than just assuming the compiler is broken) How can I help take this further? (and BTW testsuite -h doesn't print usage anymore) Cheers, Patrick On Mon, Ja

"AT_SETUP/AT_CLEANUP pair" prob?

I just tried out today's cvs tree on NetBSD-1.5S/i386. The tests give me the usual Fortran problem - fair enough I still haven't found what it's meant to do - but on the way this appeared: 2: tools.at:76 testsuite: warning: no at-check-line which means a failure happened testsuite: warnin

Re: autoconf-2.50 docs: Common Behavior/Default Includes

On Thu, May 31, 2001 at 09:26:37AM +0200, Akim Demaille wrote: ... > It might be too early to debate about this, (I mean, before 2.51), but > that's an interesting question. In addition some people seems to have > problems with our using `main ()' in tests. A friend of mine uses > Autoconf for

F77_WRAPPERS

Remember a good while back, I kept whinging about F77_WRAPPERS on my NetBSD box? Well, I just tried what's in CVS and all tests pass! (1.5W/i386) Thanks a lot, Patrick

"times" ?!

For some reason, the generated tests/testsuite contains times >at-times all over the place. What is "times"? I can't find it, and it isn't a /bin/sh builtin.. Cheers, Patrick

Re: "times" ?!

On Wed, Aug 15, 2001 at 11:52:47AM +0200, Akim Demaille wrote: > > | Tim Van Holder wrote: > | > > | > > For some reason, the generated tests/testsuite contains > | > > times >at-times > | > > all over the place. What is "times"? I can't find it, and it isn't a > | > > /bin/sh builtin.. > |

Re: "times" ?!

On Wed, Aug 15, 2001 at 12:18:20PM +0200, Akim Demaille wrote: > > | Yes - I don't have "times" anywhere.. It does say ! /bin/sh at the top > | of testsuite afterall, not some other shell. eg. of output from testsuite: > | > | 1: tools.at:47 times: not found > | FAILED near `tools.at:83'

Re: "times" ?!

quartz% cat foo.sh #! /bin/sh :; times >at-times 2>/dev/null; echo $? at_times=: times >dev/null 2>&1 && at_times=times $at_times >at-times echo $? quartz% sh foo.sh times: not found 127 times: not found 0 Cheers, Patrick

Re: "times" ?!

On Wed, Aug 15, 2001 at 01:54:42PM +0200, Tim Van Holder wrote: > > I like the former more. > > So do I. I merely provided the second in case the first > doesn't work. > > > Would you install it please? lib/autotest/general.m4. TIA. > > Once I see actual confirmation from Patrick that either

Re: "times" ?!

On Wed, Aug 15, 2001 at 04:51:04PM +0200, Tim Van Holder wrote: > > quartz% sh foo.sh > > at_times=: > > 0 > > Excellent - will install this ASAP. > > > (et etant francais aussi, nous retrouvons un petit monde > > francophone chez autoconf) > > Oh no! They're everywhere! :-) Oops - I should h

non-executable install-sh

Is this worth adding diff --git a/lib/autoconf/programs.m4 b/lib/autoconf/programs.m4 index 63f8a13..370ed31 100644 --- a/lib/autoconf/programs.m4 +++ b/lib/autoconf/programs.m4 @@ -709,7 +709,7 @@ if test -z "$MKDIR_P"; then # break other packages using the cache if that directory is #

Re: help2man missing

On Wed, Nov 12, 2008 at 07:55:28PM +0100, Ralf Wildenhues wrote: > Hi Bob, > > * Bob Friesenhahn wrote on Wed, Nov 12, 2008 at 06:33:12PM CET: > > While building libtool HEAD today, I encountered this problem: > > > > PATH=.:$PATH; export PATH; /bin/bash > > /home/bfriesen/src/gnu/libtool-head/l

make distcheck

Just a quick "Am I doing something silly?" In a project, using autoconf 2.61, automake 1.10.1, libtool 1.5.26, make distcheck succeeds. If I untar the created .tar.bz2 file, and /path/to/untar/configure && make, I get errors - somehow the manual "check build from buildir != srcdir" fails, but I th

Re: make distcheck

On Mon, Mar 02, 2009 at 08:01:59PM +0100, Ralf Wildenhues wrote: > > In a project, using autoconf 2.61, automake 1.10.1, libtool 1.5.26, > > make distcheck succeeds. If I untar the created .tar.bz2 file, and > > /path/to/untar/configure && make, I get errors - somehow the manual > > "check build fr

Re: make distcheck

On Mon, Mar 02, 2009 at 03:11:34PM -0500, Peter Johansson wrote: > Patrick Welche wrote: >> The project was broken due to liberal use of #include "../../config.h". >> configure will pop config.h in the builddir, so the source file will >> never see ../../config.h

Re: make distcheck

On Mon, Mar 02, 2009 at 09:57:28PM +0100, Ralf Wildenhues wrote: > * Patrick Welche wrote on Mon, Mar 02, 2009 at 08:52:26PM CET: > > > > I thought that running make distcheck in the source directory should > > fail? > > Not sure why you think that. distcheck d

Re: configure error

On Thu, Apr 30, 2009 at 12:44:48AM -0700, Joey Mingrone wrote: > checking for a BSD-compatible install... /usr/local/bin/ginstall -c > checking whether build environment is sane... configure: error: ls -t > appears to fail. Make sure there is not a broken > alias in your environment > configure: e

X11R7

I haven't seen breakages without this patch, but it seems logical to me... Thoughts? Patrick diff --git a/lib/autoconf/libs.m4 b/lib/autoconf/libs.m4 index cbc4218..8f7729b 100644 --- a/lib/autoconf/libs.m4 +++ b/lib/autoconf/libs.m4 @@ -236,21 +236,25 @@ m4_define([_AC_PATH_X_DIRECT], # Check X

Re: X11R7

On Fri, May 01, 2009 at 10:15:28AM -0400, Thomas Dickey wrote: > On Fri, 1 May 2009, Patrick Welche wrote: > >> I haven't seen breakages without this patch, but it seems logical to me... > > supposedly "X11R7" moved everything back into non-specific paths such as &

Re: AC_PROG_LEX behavior

On Fri, Mar 21, 2008 at 10:29:05AM -0400, Stefan Seefeld wrote: > I'm trying to use AC_PROG_LEX to discover whether flex (or lex) is > available. During testing I found out that, even when neither is > available, the LEX variable is set to ':' instead of being empty. > And sure enough, I subsequent

Re: order of programs run by autoreconf

On Tue, Sep 08, 2009 at 09:59:03AM +0200, Vincent Torri wrote: > In the autoconf manual, it is said that autoreconf runs the programs > autoconf, autoheader, aclocal, automake, libtoolize, and autopoint. > But when I run autoreconf -v to see what is running, the order is > different (that is acloca

Re: why not #include "config.h"?

On Tue, Sep 22, 2009 at 10:49:58AM +0200, Steffen Dettmer wrote: > I hope I don't ask a FAQ. If so, an URL would be appreciated. > > I've read here that someone could (should?) write > #include > specifying a system header instead of the IMHO correct > #include "config.h" > > As far as I kno

PACKAGE_URL

This almost ties in with the previous post about old versions of autoconf(!) In recent autoconf >= 2.64, I can use AC_INIT to define a PACKAGE_URL. As some of our systems only have 2.61, I thought I would # This becomes part of AC_INIT in autoconf 2.64 PACKAGE_URL="http://www.inference.phy.cam.ac

Re: PACKAGE_URL

On Mon, Apr 12, 2010 at 03:33:35PM -0400, Peter Johansson wrote: > Patrick Welche wrote: > I use the following snippet in one of my projects: > > dnl FIXME remove when we assume autoconf 2.64 > m4_ifndef([AC_PACKAGE_URL], > [AC_DEFINE([PACKAGE_URL], ["http://d

RM: to -f or not to -f

I just tried configuring libxml2 which has AC_PATH_PROG(RM, rm, /bin/rm) in its configure.in. After running configure rm: libtoolT: No such file or directory Done configuring Probably from cfgfile="${ofile}T" trap "$RM \"$cfgfile\"; exit 1" 1 2 15

VPATH test failure

Highlight (NetBSD-current/i386, autoconf head) is: ./torture.at:1223: cd $dir && $MAKE stderr: make: don't know how to make w. Stop stdout: make: stopped in /usr/src/local/autoconf/tests/testsuite.dir/240/at ./torture.at:1223: exit code was 2, expected 0 But I find it hard to start debugging as

Re: VPATH test failure

On Tue, Aug 24, 2010 at 09:19:31AM -0600, Eric Blake wrote: > On 08/24/2010 08:08 AM, Patrick Welche wrote: > >Highlight (NetBSD-current/i386, autoconf head) is: > > > >./torture.at:1223: cd $dir&& $MAKE > >stderr: > >make: don't know how to make w

Re: VPATH test failure

On Tue, Aug 24, 2010 at 08:05:04PM +0200, Ralf Wildenhues wrote: > I cannot reproduce this in NetBSD-current/i386 with git Autoconf, > see a skip instead. And now neither can I! Story is: git pull setenv PATH /bin:/usr/bin:/usr/X11R7/bin:/usr/local/bin to avoid using anything in /usr/pkg/bin ./co

Re: FW: cannot get the name of my app

On Thu, May 19, 2011 at 10:00:21PM +0200, Victor henri wrote: > I'm starting to use the autotools; everything works fine, but the only > thing I have a problem with is the name of my app. > > I'm on Ubuntu Linux and I use libbamf to get the current active app > name. It work with every running app

Re: configure checks -R and returns errors

On Tue, Mar 02, 2010 at 12:26:22PM -0700, Eric Blake wrote: > According to Vincent Torri on 10/22/2009 11:25 AM: > > Hi, > > Noticed that this old thread had no takers back when it was first asked... > > > configure checks if -R must be followed by a space when checking X. the > > log is: > > >

configure test involving plugin

glib has a configure test to check whether RTLD_GLOBAL is broken on OSF1. This involves building a plugin using libtool and then essentially AC_RUN_IFELSE on a piece of code which loads said plugin. This is all fine, and works. I was just wondering if there was a way of doing this without invoking

Re: generated version numbers

On Sat, Feb 22, 2014 at 07:47:36AM +, Harlan Stenn wrote: > But apparently something strange is going on, because after we run the > bootstrap script, if we run "make" we see: > > ... >Prerequisite `../version.m4' is newer than target `../config.h.in'. >Prerequisite `../aclocal

Re: Issues using autoconf or automake

On Mon, Mar 10, 2014 at 09:07:20PM -0700, Jim Galarowicz wrote: ... > cd . && automake-1.13 --foreign > Makefile.am:63: error: HAVE_QTLIB does not appear in AM_CONDITIONAL In http://openss.cvs.sourceforge.net/viewvc/openss/OpenSpeedShop/m4/ax_qt3.m4?revision=1.2&view=markup I see if test $f

autom4te and m4

The attached patch is necessary as if you set the environment variable M4 to "m4 --nesting-limit=2048", the regexp won't match as there is a '=' rather than a ' ' after nesting-limit. BTW if even 2048 is too small what could be wrong with my configure script? (In fact this is apr, at the autoheade

Re: autom4te and m4

On Mon, Jul 04, 2005 at 11:50:48PM -0700, Paul Eggert wrote: > Patrick Welche <[EMAIL PROTECTED]> writes: > > > The attached patch is necessary as if you set the environment variable M4 > > to "m4 --nesting-limit=2048", the regexp won't match as there is a

Re: Handling of extra macros

On Thu, Dec 01, 2005 at 09:38:31PM +0100, Alexandre Duret-Lutz wrote: > >>> "RB" == Ross Boylan <[EMAIL PROTECTED]> writes: > > RB> 5. The autoconf manual has no index entry for aclocal. There is one for > RB> aclocal.m4, and the text refers to aclocal. Is aclocal deprecated? > > aclocal is p

Re: Handling of extra macros

On Thu, Dec 01, 2005 at 04:26:40PM -0800, Russ Allbery wrote: > Patrick Welche <[EMAIL PROTECTED]> writes: > > > Isn't the long term goal "aclocal must be subsumed into autoconf"? What > > is preventing this from happening sooner rather than later? >

AC_PATH_XTRA and -rpath

I happen to use an OS (NetBSD) which uses rpath. AC_PATH_XTRA sets X_CFLAGS and X_LIBS to something sensible for compiling X, however, it misses out the necessary -rpath or -R. (libtool --config says # How to pass a linker flag through the compiler. wl="-Wl," # Flag to hardcode $libdir into a bina

Re: AC_PATH_XTRA and -rpath

; 2006-07-17 Paul Eggert <[EMAIL PROTECTED]> > > * lib/autoconf/libs.m4 (AC_PATH_XTRA): Do the check for space after > -R regardless of host. Patrick Welche reports that a space after -R > is also required for NetBSD 3.99. Not quite what I reported, NetBS

Re: AC_PATH_XTRA and -rpath

On Mon, Jul 17, 2006 at 06:54:16PM -0400, Braden McDaniel wrote: > Quoting Paul Eggert <[EMAIL PROTECTED]>: > > > Patrick Welche <[EMAIL PROTECTED]> writes: > > > > > so the -R made it, and all is well :-) > > > > > > I can't c

Re: AC_PATH_XTRA and -rpath

On Wed, Jul 19, 2006 at 12:26:36PM -0500, Albert Chin wrote: > On Mon, Jul 17, 2006 at 09:49:20PM +0100, Patrick Welche wrote: > > X_CFLAGS = -I/usr/X11R6/include > > X_EXTRA_LIBS = > > X_LIBS = -L/usr/X11R6/lib -R/usr/X11R6/lib > > X_PRE_LIBS = -lSM -lICE > >

Re: AC_PATH_XTRA and -rpath

On Thu, Jul 20, 2006 at 07:03:27PM -0700, Russ Allbery wrote: > Paul Eggert <[EMAIL PROTECTED]> writes: > > "Peter O'Gorman" <[EMAIL PROTECTED]> writes: > > >> This patch will now add -R to X_LIBS for all systems. > > > It's not supposed to. It's supposed to try -R, and add it only if it > > wor

Re: AC_PATH_XTRA and -rpath

On Fri, Aug 04, 2006 at 12:18:11AM +, Harlan Stenn wrote: > I thought he was ssaying that if /bin/ls uses RPATH then configure > should use RPATH, too. That's right - the question was how to know whether or not a system uses rpath without reinventing libtool. In fact choosing some binary we kn

netbsd sed and test 76

With today's CVS (so with AT_CHECK_AT_TITLE 2006-11-08 in ChangeLog) on NetBSD (so without GNU sed) I get 76: Macro with backslash in a test titleUNEXPECTED PASS ./autotest.at:295: $CONFIG_SHELL ./micro-suite | sed -n 's/[^:]*: \(.*[^ ]\)[]*ok.*/\1/p' Not enabli

Re: netbsd sed and test 76

On Fri, Nov 10, 2006 at 04:41:13PM +0100, Ralf Wildenhues wrote: > Hello Patrick, > > Thanks for the report; you were beaten by a couple of days. ;-) > > * Patrick Welche wrote on Fri, Nov 10, 2006 at 04:37:38PM CET: > > With today's CVS (so with AT_CHECK_AT_TIT

Re: netbsd sed and test 76

On Fri, Nov 10, 2006 at 05:06:35PM +0100, Ralf Wildenhues wrote: > No. As hinted at in > http://lists.gnu.org/archive/html/autoconf-patches/2006-11/msg00010.html > the problem is larger and requires more changes and thought. Oh dear.. and it looks as though I need to subscribe to yet another list

AC_CONFIG_HEADERS and autoheader

>From Nov 19 autoconf source, info autoconf says: -- Macro: AC_CONFIG_HEADERS (HEADER ..., [CMDS], [INIT-CMDS]) ... Usually the input file is named `HEADER.in'; however, you can override the input file name by appending to HEADER a colon-separated list of input files. Examples:

Re: AC_CONFIG_HEADERS and autoheader

On Tue, Dec 05, 2006 at 08:44:59PM +0100, Stepan Kasal wrote: > Hello, > > On Tue, Dec 05, 2006 at 10:17:57AM -0800, Paul Eggert wrote: > > Patrick Welche <[EMAIL PROTECTED]> writes: > > > Is that AC_CONFIG_HEADERS syntax simply not allowed? (missing file name) >

Re: AC_CONFIG_HEADERS and autoheader

On Wed, Dec 06, 2006 at 01:13:31PM +0100, Stepan Kasal wrote: > 2006-12-05 Stepan Kasal <[EMAIL PROTECTED]> > > * doc/autoconf.texi (Configuration Headers): Remove the > example with multiple input files. > (autoheader Invocation): Encourage `AH_BOTTOM', discouraging > m

AC_TYPE_UINT8_T and co

-- Macro: AC_TYPE_UINT8_T If `stdint.h' or `inttypes.h' defines the type `uint8_t', define `HAVE_UINT8_T'. Otherwise, define `uint8_t' to an unsigned integer type that is exactly 8 bits wide, if such a type exists. but for some reason HAVE_UINT8_T is not defined! (Today's CVS) %

Re: AC_TYPE_UINT8_T and co

On Mon, May 28, 2007 at 10:55:45AM -0600, Eric Blake wrote: > According to Patrick Welche on 5/28/2007 10:45 AM: > > (Also removes what appear to be spurious #( ) > > > > - case $ac_cv_c_int$1_t in #( > > - no|yes) ;; #( > > + case $ac_cv_c_int$1_t in > >

Re: AC_TYPE_UINT8_T and co

On Tue, May 29, 2007 at 04:54:52AM +0200, Stepan Kasal wrote: > I'm going to answer only one of your minor comments now. Thank you for the explanation of CHECK_FOO vs CHECK_FOOS! The documentation consistency change seems fine, and my configure.ac snippet becomes AC_INIT([autotype],[1.0]) AC_CONF

Re: AC_TYPE_UINT8_T and co

On Tue, May 29, 2007 at 01:31:04PM +0100, Patrick Welche wrote: > we makes sense thanks to your answer. ^^ which ___ Autoconf mailing list Autoconf@gnu.org http://lists.gnu.org/mailman/listinfo/autoconf

Re: AC_TYPE_UINT8_T and co

On Tue, May 29, 2007 at 01:31:04PM +0100, Patrick Welche wrote: > On Tue, May 29, 2007 at 04:54:52AM +0200, Stepan Kasal wrote: > > I'm going to answer only one of your minor comments now. > > Thank you for the explanation of CHECK_FOO vs CHECK_FOOS! The documentation > c

Re: AC_TYPE_UINT8_T and co

Sorry for getting email trigger happy - I seem to still be confused... If AC_TYPE_UINT8_T is run on a system without uint8_t, a typdef for uint8_t appears in config.h, but it is only created at the end of configure, so how would a subsequent test in configure see the typedef? Cheers, Patric

Re: AC_TYPE_UINT8_T and co

On Wed, May 30, 2007 at 11:31:56AM +0200, Stepan Kasal wrote: > I'm afraid there are some misunderstandings here. I'll try to > make things more clear. > > Let's start with your very first mail: > Until yesterday, the manual said: > > | -- Macro: AC_TYPE_INT8_T > | If `stdint.h' or `inttypes

Re: AC_TYPE_UINT8_T and co

On Wed, May 30, 2007 at 02:48:12PM +0200, Stepan Kasal wrote: > And yes, it sounds inconsistent. IMVHO the ac_cv_c_ prefix should be > changed to ac_cv_type_ here. What do others think? I think that - changing ac_cv_c_ to ac_cv_type_ and - changing AC_CHECK_TYPES to test for ac_cv_type_ != no r

Re: Configuring subdirectories test

On Thu, Nov 08, 2007 at 02:40:31PM +, Patrick Welche wrote: > The next thing after the chmod is > > ./configure > > which wants to write config.log to . and can't. (No a+w on 131) Actually it is ./configure --help, which shouldn't try to write

Configuring subdirectories test

I'm seeing a failure for test 131, today's CVS-head (hopefully same as git ;-) ) on NetBSD-current/i386. I'm a little confused as to how it could pass. It seems that a variety of files/directories are created, then (testsuite): # Running the outer configure recursively should provide the innermo

Re: Configuring subdirectories test

On Thu, Nov 08, 2007 at 07:58:13PM +0100, Ralf Wildenhues wrote: > Patrick, how come the `Deep package' test did not fail for you? It did, but I thought I would look at the first failure first :-) It's a bit late over here... Cheers, Patrick ___ Aut

Re: Configuring subdirectories test

On Thu, Nov 08, 2007 at 07:58:13PM +0100, Ralf Wildenhues wrote: > OK to apply, even if it looks a bit ugly? Well, it does fix the test for me... > Patrick, how come the `Deep package' test did not fail for you? > Is that only because you used an older version? Could you try current > HEAD with

autoconf from cvs

There is a potential pitfall when grabbing autoconf via CVS from the git repository which is that VERSION is computed by build-aux/git-version-gen which runs git describe. If you checked out via CVS, git rightly complains that the checkout isn't a git repository. VERSION is then UNKNOWN, rather tha

Re: Integration of autoconf and pkg-config

On Wed, Dec 19, 2007 at 10:31:16AM +0100, Stefano Sabatini wrote: > Just another little newbie question, am I supposed to copy the pkg.m4 > file containing the macro somewhere into my package (e.g. in > aclocal.m4)? I don't know about pkg.m4 specifically, but if there is a macro in it which you ne

Re: Integration of autoconf and pkg-config

On Wed, Dec 19, 2007 at 04:37:17PM +0100, Stefano Sabatini wrote: > On date Wednesday 2007-12-19 14:38:31 +0000, Patrick Welche wrote: > > On Wed, Dec 19, 2007 at 10:31:16AM +0100, Stefano Sabatini wrote: > > > Just another little newbie question, am I supposed to copy the

quoting fun

122. AC_CONFIG_FILES, HEADERS, LINKS and COMMANDS (torture.at:133): FAILED (tort ure.at:271) is failing for me, I think due to the fun quoting from config.status at eval sed \"\$ac_sed_extra\" "$ac_file_inputs" | $AWK -f "$tmp/subs.awk" >$tmp/out \ where ac_sed_extra: /^[ ]*VPATH[ ]*=/{

Re: quoting fun

On Sat, Feb 02, 2008 at 04:51:46PM +0100, Ralf Wildenhues wrote: > Please try this patch: > > I'll try to finish it tomorrow. I forgot about that thread :-/ - your patch makes all the tests pass - ignore my comment abou

Re: quoting fun

On Tue, Feb 05, 2008 at 08:51:42PM +0100, Ralf Wildenhues wrote: > * Patrick Welche wrote on Sat, Feb 02, 2008 at 06:09:03PM CET: > > - one strange thing: I tried running the testsuite -v -d 122 on its own > > in tests/testsuite.dir/122 after it passed with all the rest, an

2.53a m4sh.at:74 failure

source of 27 Mar 2002 17:20GMT NetBSD-1.5ZC/i386 Failed tests: GNU Autoconf 2.53a test suite test groups: NUM: FILENAME:LINE TEST-GROUP-NAME KEYWORDS 15: m4sh.at:29 LINENO Skipped tests: GNU Autoconf 2.53a test suite test groups: NUM: FILENAME:LINE TEST-GROUP-NAME

Re: 2.53a m4sh.at:74 failure

In testsuite: cat >test/test-1.as <<'_ATEOF' AS_INIT if false; then AS_UNSET([LINENO]) fi _AS_PREPARE echo "Line: $LINENO" grep 'Line: .*LINENO' $[0] >/dev/null || AS_ERROR([cannot find original script]) exit 0 _ATEOF But once run with testsuite -v -x 15 % cat testsuite.dir/015/test/test

Re: 2.53a m4sh.at:74 failure

On Thu, Apr 18, 2002 at 04:08:45PM +0200, Akim Demaille wrote: > Could you please try this patch? Just apply it, and... > > cd tests > make testsuite > ./testsuite -k lineno No luck: % cd tests % gmake testsuite ./autom4te --language=autotest -I . suite.at -o testsuite.

Re: 2.53a m4sh.at:74 failure

On Thu, Apr 18, 2002 at 05:40:25PM +0200, Akim Demaille wrote: > > Heck, it is right! How come I did not see that failure :( > > Try this one. It seems to do the same as before.. Patrick % gmake testsuite ./autom4te --language=autotest -I . suite.at -o testsuite.tmp mv testsuite.tmp testsuit

Re: 2.53a m4sh.at:74 failure

On Mon, Apr 22, 2002 at 11:12:32AM +0200, Akim Demaille wrote: > > * tests/m4sh.at (LINENO): Skip the test if LINENO cannot be unset. > Fixes false failures on Darwin. Still same as before, but at least the mystery as to why it worked on 1 NetBSD computer and not 2 others is: the one

Re: 2.53a m4sh.at:74 failure

On Mon, Apr 22, 2002 at 07:57:52PM +0200, Akim Demaille wrote: > I really thought you were under Darwin, all my apologies. In that > case, could run the test suite of the just released snapshot, and send > the testsuite.log to bug-autoconf? We'll start from scratch :( It's on its way.. Patrick

AC_FUNC_GETLOADAVG

I'm afraid the new version of functions.m4 doesn't work for me (reverting from 1.60 to 1.58 allows test to pass). 143. acfunctions.at:15: FAILED near `acfunctions.at:15' with +configure: error: ./getloadavg.c is missing Where should I find getloadavg.c? On my system getloadavg() is in -lc, so

Re: AC_FUNC_GETLOADAVG

And of course your failed test works for me :/ % cd tests % testsuite -v -x -d 4 ... % cd testsuite.dir/004 % cat configure.ac define([active], [ACTIVE]) % ../../autoconf -t define | sed -n '$p' configure.ac:1:define:active:ACTIVE % cat autom4te.cache/traces.0 m4trace:configure.ac:1: -1- define([

Re: AC_FUNC_GETLOADAVG

On Sun, May 12, 2002 at 10:08:54PM -0700, Paul Eggert wrote: > > From: Patrick Welche <[EMAIL PROTECTED]> > > Date: Sun, 12 May 2002 16:41:08 +0100 > > > > I'm afraid the new version of functions.m4 doesn't work for me > > (reverting from 1

Re: AC_FUNC_GETLOADAVG

On Sun, May 12, 2002 at 10:08:54PM -0700, Paul Eggert wrote: > > From: Patrick Welche <[EMAIL PROTECTED]> > > Date: Sun, 12 May 2002 16:41:08 +0100 > > > > I'm afraid the new version of functions.m4 doesn't work for me > > (reverting from 1

Re: AC_FUNC_GETLOADAVG

On Thu, May 30, 2002 at 04:53:10AM -0700, Paul Eggert wrote: > > Date: Thu, 30 May 2002 11:19:17 +0100 > > From: Patrick Welche <[EMAIL PROTECTED]> > > > > Why is that check there? > > So that your software distribution is portable to hosts that don'

Re: AC_FUNC_GETLOADAVG

On Thu, May 30, 2002 at 01:43:38PM +0100, Patrick Welche wrote: > On Thu, May 30, 2002 at 04:53:10AM -0700, Paul Eggert wrote: > > > Date: Thu, 30 May 2002 11:19:17 +0100 > > > From: Patrick Welche <[EMAIL PROTECTED]> > > > > > > Why is tha

make problem

With CVS of 14 Jun 14:49 GMT, I just got: ../tests/autom4te --language M4sh ./autoconf.as -o ./autoconf.in autom4te: cannot open /usr/src/local/autoconf/lib/autom4te.cfg: No such file or directory at /usr/src/local/autoconf/bin/autom4te line 428 *** Error code 1 Just in case it matters, one ma

Re: make problem

On Fri, Jun 14, 2002 at 05:42:19PM +0100, Patrick Welche wrote: > With CVS of 14 Jun 14:49 GMT, I just got: > > ../tests/autom4te --language M4sh ./autoconf.as -o ./autoconf.in > autom4te: cannot open /usr/src/local/autoconf/lib/autom4te.cfg: No such file or >directory >

Re: make problem

On Fri, Jun 14, 2002 at 10:56:03AM -0700, Paul Eggert wrote: > > That's a known bug in the current CVS, which I introduced and am > currently looking into. (Sorry about that.) :-) So you already know :-) I must be on the wrong list.. Is the other bit related? Using the old /usr/local/bin/autom4

autom4te and perl 5.8.0

I made a bit of a fool of myself on automake list.. I wonder if any of you have perl 5.8.0rc1 installed, and our recent autoconf to try out: % cat file.pl BEGIN { my $datadir = ($ENV{'autom4te_perllibdir'} || '/usr/local/share/autoconf'); unshift @INC, "$datadir"; } use Autom4te::General; us

Re: autom4te and perl 5.8.0

On Wed, Jun 19, 2002 at 11:53:24PM +0200, Gerrit P. Haase wrote: > Hallo Patrick, > > Am Mittwoch, 19. Juni 2002 um 17:07 schriebst du: > > > On Wed, Jun 19, 2002 at 10:28:23AM +0200, Alexandre Duret-Lutz wrote: > >> > >> To make sure that it's not the fault of Autoconf, can you try > >> this r

Re: configure.lineno leftover?

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

Re: configure.lineno leftover?

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

mkinstalldirs ?

Just noticed with today's cvs: # gmake install ... gmake[2]: Nothing to be done for `install-exec-am'. /bin/sh ./mkinstalldirs /autoconf /bin/sh: Can't open ./mkinstalldirs gmake[2]: *** [install-pkgdataDATA] Error 2 gmake[2]: Leaving directory `/usr/src/local/autoconf' mkinstalldirs lives in .

Re: mkinstalldirs ?

On Tue, Sep 24, 2002 at 05:58:47PM +0200, Akim Demaille wrote: > > | Just noticed with today's cvs: > | # gmake install > | ... > | gmake[2]: Nothing to be done for `install-exec-am'. > | /bin/sh ./mkinstalldirs /autoconf > | /bin/sh: Can't open ./mkinstalldirs > | gmake[2]: *** [install-pkgdataD

autopoint?

What is autopoint? % sh bootstrap Bootstrapping CVS Bison... running: autoreconf --verbose --install --force autoreconf: Entering directory `.' autoreconf: running: autopoint --force Can't exec "autopoint": No such file or directory at /usr/local/share/autoconf/Autom4te/General.pm line 498, line

Re: Compiling autoconf-2.56

On Fri, Nov 29, 2002 at 12:17:25PM +0200, Dani Mezher wrote: ... > I am building autoconf-2.56 on a FreeBSD-4.7 system. I > installed GNU M4 1.4 but the configure script crashes with an error GNU > M4 1.4 is required. U can see the output of the configure script. ... > checking for m4.

AC_CYGWIN

I had a go at compiling glib on a non-linux platform (!) and found configure.in:300: warning: AC_CANONICAL_HOST invoked multiple times Line 106 is AC_CANONICAL_HOST, and line 300 is AC_CYGWIN. I found the definition of AC_CYGWIN in lib/autoconf/specific.m4: # AC_CYGWIN # - # Check for

rpath

By the time configure & libtool have done there thing on a system, they know whether or not that system needs a -Wl,-R${libdir} or not. Is there a way of getting that information out simply? (In order to create a pkg-config file with a sensible --libs entry) Cheers, Patrick

style

In the Fine Manual: @node Coding Style @section Coding Style In order to highlight the recommended coding style, here is a macro written the old way: @example dnl Check for EMX on OS/2. dnl _AC_EMXOS2 AC_DEFUN(_AC_EMXOS2, ... and the new way: @example # _AC_EMXOS2 # -- # Check for EMX o

  1   2   >