Sorry, it's not fun at all actually. I was looking at the wrong
definitions, and the space is just the space that was really left
before the comment :(
> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
Lars> BTW, why does srcdir default to .. when this happens? It seems
Lars> like an ugly hack implemented to help ill-configured software
Lars> packages to configure from my point of view...
Could you detail what you are referring to :-?
In Autoconf, the words which include _A?_, A?_, _m4_, m4_ are
forbidden, in order to diagnose unexpanded macros. There is
(currently) no escape to this rule.
But then we have a conflict for Libtool for:
~/src/libtool % ace nostromo 15:58
configure:5
| Akim,
| I just updated my autoconf and rebuilt it. This modification still
| did not help.
|
| Akim Demaille wrote:
| > I think this is fixed now. It looks like another incarnation of
| >
| > 2000-11-03 Akim Demaille <[EMAIL PROTECTED]>
| >
| > AC_REQUIRE an
> "Lars" == Lars Hecking <[EMAIL PROTECTED]> writes:
Lars> Have you tried the native m4, too? See
Lars> http://sources.redhat.com/ml/autoconf/2000-03/msg00213.html
It won't work: we depend on the tracing options, and Marc has not
imported this in his M4.
ROVIDE): New test.
2000-11-03 Akim Demaille <[EMAIL PROTECTED]>
Index: acgeneral.m4
===
RCS file: /cvs/autoconf/acgeneral.m4,v
retrieving revision 1.616
diff -u -u -r1.616 acgeneral.m4
--- acgeneral.m4 2000/11/03 13:41:40 1.6
| LOL... Whose the guy who revamped AC_PROVIDE recently? He is
| dangerous... :)
s/Whose/Who's/, sorry :(
> "Bernard" == Bernard Dautrevaux <[EMAIL PROTECTED]> writes:
Bernard> Whar is AS_xxx used for in autoconf?
Autoshell, see m4sh.m4.
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> Hello, Akim!
>> In Autoconf, the words which include _A?_, A?_, _m4_, m4_ are
>> forbidden, in order to diagnose unexpanded macros. There is
>> (currently) no escape to this rule.
Pavel> Not _m4_.
Then we should, shouldn't we? N
> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
>> I'd rather take almost the full range and except AR if we need to.
Alexandre> I'd prefer that the maintainers of autoconf weren't so
Alexandre> greedy about prefixes :-)
In fact my position (let's take the full range) is also b
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> Autoconf could also check configure with "sh -n" - it may be
Pavel> more effective than scanning for macros, although some
Pavel> constructs will not be caught.
No really, `sh -n' is way too weak, it can only catch very obvious
mis
> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Nov 6, 2000, Earnie Boyd <[EMAIL PROTECTED]> wrote:
>> Why not just prefix the reserved autoconf variables with a `_'
>> character?
Alexandre> autoconf used to use just AC_/ac_. Now, it's claiming
Alexandre> ownershi
OG_CC will complain because AC_PROG_CPP was run
beforehand.
The following patch fixes this issue. There was another obvious
implementation possibility: having AC_LANG_COMPILER(C) provide
AC_PROG_CC, but I think this approach is better.
Index: ChangeLog
from Akim Demaille <[EMAIL P
upport .." >&6(for ZLIB support ...)
|
| Is there a fix for this?
Thanks, it's an actual stupid bug... I'm applying this:
2000-11-07 Akim Demaille <[EMAIL PROTECTED]>
* acgeneral.m4 (AC_CHECKING): Is not an alias of AC_MSG_NOTICE,
hence use AU
| AFAIS, apparent cause are the --?* clauses in autoconf.sh:
|
| > --autoconf-dir | --a* | -A )
| >test $# = 1 && eval "$exit_missing_arg"
| >shift
Yep, I'm in favor of removing all the --a* etc. it makes it uselessly
ambiguous, and short options are a better means to abbrev
| First, I've installed m4 version 1.4.
| Then i've configured autoconf (run configure tool)
| Then i've launched make tha issues the following error :
|
| "Error: Autoconf requires GNU m4 1.1 or later
| make: *** [autoconf.m4f] Error 1"
|
| I've made sure the m4 used is the right one (that's t
| Hi,
| With autoconf-2.13, I have been using constructs like the one below
| to detect whether gcc is used:
| [Similar constructs are used in libiberty, libgloss, newlib and
| Cygwin's winsup]
|
| if test $ac_cv_prog_gcc = yes; then
| # do something
| fi
|
| This test relies on the value of ac
>>>>> "Akim" == Akim Demaille <[EMAIL PROTECTED]> writes:
> $ac_cv_c_compiler_gnu() = yes;
Akim> Heck. Thanks for noticing :(
I'm checking in the fix, thanks!
* aclang.m4 (ac_cv_prog_gcc, ac_cv_prog_gxx, ac_cv_prog_g77):
Be AU_DEFUN'd, not AU_ALIAS'd.
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> Why do we do it? Whe want to protect the user from
Pavel> overquoting. When I was suggesting to protect from underquoting
Pavel> nobody seemed to be enthusiastic about it although it is
Pavel> underquoting that results in truly weir
| AC_INIT(whatever)
| my_ac_builddir_save="`pwd`"
UQA
| cd "$srcdir"
| srcdir="`pwd`"
UQA
| cd "$my_ac_builddir_save"
Useless Quotes Award :)
> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
Lars> Weird. Doesn't autoconf try to disable (unset) $CDPATH?
It does, and I believe it's knowledge of the various particular cases
is good enough so that you can use `cd "$foo" && pwd`.
Thanks for tracking this down. In fact the problem is that I'm not
sure what should be done out of the PROLOGUE in the case of
AC_LANG_PROGRAM(Fortran 77). Maybe you can answer me.
For C for instance, we have:
# AC_LANG_PROGRAM(C)([PROLOGUE], [BODY])
# --
m
Actually in the current case, GNUness should be tested in the
prologue, but in the body, which solves the current issue.
Still, what should we do out of the prologue for
AC_LANG_PROGRAM(Fortran 77)? Still ignore it but warn when it's
nonempty?
> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
Lars> Sure, but useless quotes are better than missing quotes. I
Lars> haven't studied the Bourne shell syntax that hard, and given all
Lars> the variations over how the so-called Bourne shell compatible
Lars> shells behave, I've taken the b
| On Tue, Nov 07, 2000 at 06:09:45PM +0100, Akim Demaille wrote:
| : absolute="`cd "$relative" && pwd`"
| :
| : is not portable, and nor is
| :
| : absolute="`cd \"$relative\" && pwd`"
|
| That the first line is not portable comes as n
>>>>> "Thomas" == Thomas E Dickey <[EMAIL PROTECTED]> writes:
Thomas> On 7 Nov 2000, Akim Demaille wrote:
>> >>>>> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
>>
Lars> Weird. Doesn't autoconf tr
> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
>> Yep, I'm in favor of removing all the --a* etc. it makes it
>> uselessly ambiguous, and short options are a better means to
>> abbreviate. And trying to support exact abbreviations, as is done
>> in configure, is an additional t
ICAL_TARGET([])
|
| I understand that the role of autoupdate is to create a script conforming
| with the today's guidelines. I don't think that giving an empty argument
| to AC_CANONICAL_TARGET is what we recommend users to do.
Just some information you might have missed:
2000-11-03 Akim Dem
| On Nov 7, 2000, Pavel Roskin <[EMAIL PROTECTED]> wrote:
| > Hello, Akim!
| >> > $ac_cv_c_compiler_gnu() = yes;
| >>
| Akim> Heck. Thanks for noticing :(
| >>
| >> I'm checking in the fix, thanks!
| >>
| >> * aclang.m4 (ac_cv_prog_gcc, ac_cv_prog_gxx, ac_cv_prog_g77):
| >> Be AU_DEFUN'd, no
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> Hello, Akim!
>> The one and only approach is the current one. Let's stop
>> reinventing the wheel.
Pavel> That's probably too stronly worded :-)
I agree, and I apologize.
>> But what's the point???
Pavel> The point is uniformi
| Hello, Akim!
| > For instance if for some reason the user had
| >
| > AC_CANONICAL_SYSTEM([This macro computes $host])(echo "$host")
| >
| > then autoupdate with parens (the current one) will produce
| >
| > AC_CANONICAL_TARGET([])(echo "$host")
| >
| > which is correct, while without paren
Hi David,
Sorry for the delays. I find it hard to understand how broken your
expr is. just to be sure, this refers to QNX 4.2.5 right?
| > expr 'a' : '\(a\)' || echo failure
|
| Gives:
| a
| failure
Here, $? is wrong, but the output is correct.
| > expr 'a' : '\(b\)' && echo failure
|
| G
> "Earnie" == Earnie Boyd <[EMAIL PROTECTED]> writes:
Earnie> But, neither of you answered my question. Why not use `_' to
Earnie> prefix A?_* names? You can then declare that the user is
Earnie> incorrect as _A?_* is a vendor variable! I as a user should
Earnie> be allowed to declare A?_
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> It doesn't mean that we have eliminate all common code right
Pavel> now, but it would be nice to start the conversion process now,
Pavel> e.g. with the option parcer.
We have to finish M4sh first, and extract an autoscript or somet
> "Paul" == Paul Eggert <[EMAIL PROTECTED]> writes:
Paul> How about if we add an AC_PROG_EXPR that searches for a working
Paul> expr and sets EXPR to its value, or to "false" if it can't find
Paul> one. Then AS_DIRNAME_EXPR can AC_REQUIRE([AC_PROG_EXPR]) and
Paul> use "$EXPR" instead of "exp
| The POSIX standard says:
|
|Usually, the matching operator shall return a string representing
|the number of characters matched ('0' on failure). Alternatively,
|if the pattern contains at least one regular expression
|subexpression "[\(...\)]", the string corresponding to "\1"
| Hello, Ralf!
| > checking how to run the C preprocessor... cc -E
| > checking for sh-rtems-gcc... sh-rtems-gcc
|
| It's a separate problem that needs to be addressed before 2.50.
| AC_PROG_CPP should require AC_PROG_CC.
Right.
| Maybe we should go as far as to make both AC_PROG_CPP and AC_PR
hi Alexandre,
I'm sorry to insist, but we've met this bug several times, and we
never had a clear definition of it. I'd like to fix Autoconf but
understanding why it's fixed :)
The problem with Ultrix is probably that somehow it changes the 8th
bit of the variable values. But `when' is the qu
>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Nov 10, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>> My bet is that IFS splitting is not performed because `:
>> ${French="$french"}' sets the 8
> "Bruce" == Bruce Korb <[EMAIL PROTECTED]> writes:
Bruce> Johan Danielsson wrote:
>> Autoheader does `sed "s/[^/]*$//"', which doesn't work with Cray
>> sed, and Cray claims this is in compliance with POSIX. Is this
>> correct?
Bruce> It is irrelevant. If true and Cray changes, there will
| > From: Akim Demaille <[EMAIL PROTECTED]>
| > Date: 15 Nov 2000 16:32:18 +0100
| >
| > Do you mean `s,[^/]*$,,' would work here? Man, how can someone write
| > something that broken :(.
|
| Just for the record, Cray is correct: POSIX does not allow
| `s/[^/]*$//&#
> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
Lars> I'm using AC_CHECK_SIZEOF on Cygwin and have found out that the
Lars> `conftest' file contains a \r\n after the size. The \r
Lars> character follows the variable into config.status (the
Lars> undefs.sed scripts) and ruins the sed scrip
| > From: Akim Demaille <[EMAIL PROTECTED]>
| > Date: 10 Nov 2000 10:23:25 +0100
| >
| > But our troubles could be worse than this, since `expr' is not used
| > only in AS_DIRNAME. For instance any CVS configure contains
| >
| > ac
| I think I found it - checking now:
|
| Index: aclang.m4
| ===
| RCS file: /cvs/autoconf/aclang.m4,v
| retrieving revision 1.89
| diff -u -r1.89 aclang.m4
| --- aclang.m4 2000/11/14 11:06:37 1.89
| +++ aclang.m4 2000/11/16
> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
Lars> : so how come we waited till today to discover this issue?
Lars> Running configure on the w32 platform hasn't really been common,
Lars> nor is it common yet either...
Is this the only failure of the test suite?
| On Thu, Nov 16, 2000 at 12:51:05PM +0100, Akim Demaille wrote:
| :
| : | -[FILE *f = fopen ("conftestval", "w");
| : | +[FILE *f = fopen ("conftestval", "wb");
| :
| : Portable?
|
| Not really my cup of tea, but (from fopen(3) on Linux):
|
| # The
> "Lars" == Lars J Aas <[EMAIL PROTECTED]> writes:
Lars> Sorry about that - autoconf.m4f was missing, but warnings were
Lars> supressed in the testsuite... The testsuite seems to be running
Lars> ok now that that is taken care of (got some failures in
Lars> semantics).
Thanks for the logs!
> "Earnie" == Earnie Boyd <[EMAIL PROTECTED]> writes:
Earnie> Lars, just to satisfy my curiosity please send me the output
Earnie> of `cygcheck -s -r -v'.
What's this? I'm looking for all the possible uname and the like
which we could include in the config.log. Currently, you have:
This
Would someone have an idea of why there is an exit right after the
./conftest in AC_TRY_RUN/AC_RUN_IFELSE? I found nothing interesting
in the ChangeLog.
| # _AC_RUN_IFELSE(PROGRAM, [ACTION-IF-TRUE], [ACTION-IF-FALSE])
| #
| # Compile,
>>>>> "Gary" == Gary V Vaughan <[EMAIL PROTECTED]> writes:
Gary> On Wed, Nov 08, 2000 at 10:43:38AM +0100, Akim Demaille wrote:
>> I can't wait for Gary's M4
Gary> Me either =)O|
Gary> Unfortunately, Rene has been too busy to reply to t
> "Mo" == Mo DeJong <[EMAIL PROTECTED]> writes:
Mo> Whats the deal?
That's the result of our hesitations about this issues, and the fact
that we couldn't even come up with a precise definition of what EXEEXT
and OBJEXT are.
For a start: are they related to host, or to build?
Once things ar
| I'm trying to build autoconf on UNICOS 10, besides the sed issue, I
| get these errors when running make check:
|
| ./semantics.m4:118: testing...
| --- - Tue Nov 14 18:17:47 2000
| +++ stdout Tue Nov 14 18:17:47 2000
| @@ -1,6 +1,6 @@
| #define SIZEOF_CHAR 1
| -#define SIZEOF_CHARCHAR
> "Mirar" == Mirar <[EMAIL PROTECTED]> writes:
Mirar> autoconf (or autoheader?) also complains on the use of AF_UNIX
Mirar> and AF_INET inside tested C code in configure.in, but it's kind
Mirar> of hard to test socket code without using those #define's.
This will be solved very soon. Sorry
| The following autoconf code:
| AC_CHECK_FUNC(gettext, [AC_DEFINE(HAVE_GETTEXT)],
| AC_CHECK_LIB(intl, gettext, [LIBS="$LIBS -lintl"
| AC_DEFINE(HAVE_GETTEXT)],
| INTLIBS="" ))
Improperly quoted.
AC_CHECK_FUNC(gettext,
> "Johan" == Johan Danielsson <[EMAIL PROTECTED]> writes:
Johan> xyz-xY
Johan> I guess this means this it won't work.
It means the test should have been ignored. I don't know why it has
not been :(
Could you try to see what happened? Rerun the corresponding debug.sh,
and run `../autoupda
> "phane" == phane Genaud writes:
phane> Hi, on a Irix 6.5 system, using automake 1.4, autoconf 2.12
Please, consider upgrading at least of 2.13. If the trouble persists,
give a try to CVS Autoconf (See http://sources.redhat.com/autoconf/
for details), and report bugs against this CVS Auto
> "Peter" == Peter Eisentraut <[EMAIL PROTECTED]> writes:
Peter> * Do not print an end-of-line at all.
!!! Of course! Thanks!
But the question remains open :)
| When I try to build the latest CVS autoconf I get:
| ...
| Processing ../autoconf.m4 produced output:
| ../autoconf.m4: define(AC_ACVERSION, 2.49b)
|
| gmake[1]: *** [autoconf.m4f] Error 1
|
| What did I miss?
Arg. That's a sanity check I recently introduced in ./Makefile.am:
when proce
> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Nov 27, 2000, "Town, Brad" <[EMAIL PROTECTED]> wrote:
>> I tried that as well (I didn't list everything I'd tried). If
>> there's no way to do it, I'd like to know so I don't keep trying
>> the impossible...
Alexandr
| Hi,
| given this configure.in (a useless case, but it'll serve as an
| example):
|
| == [snip]
|
| AC_INIT(configure.in)
| AC_CONFIG_HEADER(config.h)
|
| if false; then
| AC_PATH_XTRA
| else
| AC_PATH_XTRA
| fi
|
| AC_OUTPUT
|
|
| == [snip] ===
| But I got this warning:
|
| | configure.in:32: warning: AC_CONFIG_SUBDIRS: you should use literals
|
| But the code actually scans the subdirs:
|
| | for a in `(cd $srcdir ; echo *)`
| | do
| | if test "$a" != "CVS" -a "$a" != "RCS" ; then
| | if test -d "$srcdir/$a" ; then
| |
| Akim:
Hi!
| I'm curious as to how you plan to use this information. I got the
| impression that you were suggesting to use something like:
|
| expr foo : '\(bar\)' \| a : 'a\(\)
Nope, not exactly. It is true there are two issues being debated, and
the thread is quite confusing.
The first
> "Paul" == Paul Eggert <[EMAIL PROTECTED]> writes:
>> expr ''
Paul> It returns the empty string on both Solaris 2.5.1 sparc
Paul> (/usr/bin/expr) and Solaris 8 sparc (both /usr/bin/expr and
Paul> /usr/xpg4/bin/expr). For what it's worth, the standard requires
Paul> this behavior.
Thanks P
I tried to summarize the thread, and here is my proposal. Comments
are welcome, suggestions are most welcome :)
`expr'
No `expr' keyword starts with `x', so use `expr x"WORD" : 'REGEX''
to keep `expr' from misinterpreting WORD.
Don't use `length', `substr', `match' and `index'
>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Nov 17, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>> Because I believe the definition I gave of it implied it was two
>> chars wide. But maybe I'm making
Lars> binary mode (which may or may not cause trouble on some obscure
Lars> systems) again.
Yep, I'm also in favor of this solution.
Index: ChangeLog
from Akim Demaille <[EMAIL PROTECTED]>
When using Cygwin, in spite of all their efforts, it may happen
that `con
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> My guess is that your acversion.m4 has not been updated for
Pavel> some reason and still contains "define" instead of "m4_define"
s/some reason/CVS/ :)
ng obscure :)
Index: ChangeLog
from Akim Demaille <[EMAIL PROTECTED]>
QNX 4.2.5's expr always exits 1 when `:' is used with parens.
* doc/autoconf.texi (Limitations of Usual Tools) : More
information, thanks to Paul Berrevoets, Paul Eggert and David
Daniele, I'm applying the patch you need to be able to use CVS
Autoconf. May I ask you if you could fetch CVS Autoconf and check
that the test suite succeeds?
Thanks a lot!
Akim
Index: ChangeLog
from Akim Demaille <[EMAIL PROTECTED]>
* acgeneral.m4 (_AC_OUT
nostromo 18:44
ac_cv_tata=${ac_cv_tata=tata}
test "${ac_cv_toto+set}" = set || ac_cv_toto='${foo}'
I'm applying the following patch, again so that David can check the
fix is effective, but modification requirements are most welcome!
Index: C
I changed the status of Autoconf in BUGS.
-*- outline -*-
This file lists the bugs you must be aware ofBe sure to check this
file before using Autoconf, and especially CVS versions of Autoconf.
Many other bugs are registered on the GNATS server:
http://sources.redhat.com/cgi-bin/gnatsweb
I think we addressed the most serious bugs. I propose we plan 2.49b
this Friday. Let's work on the remaining serious bugs, but even if
there are some left, let's snapshot, too many people are waiting for a
better snapshot than 2.49a.
So please, if you happen to use a funny environment (which i
Err, I think I should stop working on Autoconf right now:
| for ac_file in `ls conftest.o conftest.obj conftest.* 2>/dev/null`; do
| case $ac_file in
| *.o | *.obj ) ac_cv_objext=`expr "$ac_file" : '.*\.\(.*\)'`;;
I really wrote thisline to catch .o and .obj before any other
ex
> "David" == David Morgan <[EMAIL PROTECTED]> writes:
David> Akim, It is unchanged at this time.
Actually Autoconf was changed, but its ./configure was not.
David> I can wait till tomorrow.
Good morning! :)
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
>> > b=`false` || echo Ok > > It should print "Ok". If it doesn't, we
>> are in trouble.
>>
>> We are in trouble - no response.
Pavel> :-(
I concur.
Pavel> Well, one consequence is that some errors will not be
Pavel> detected. For exa
> "Lars" == Lars Hecking <[EMAIL PROTECTED]> writes:
Lars> The test suite takes hours on some systems ...
More than before?
Does anybody knows how to profile a shell script?
Lars> Solaris 7, IRIX 6.3, DU 4.0F are fine.
Good news!
Lars> HP-UX 9.05 says "All 174 tests were successful (2
| acversion.m4:
| # This file is part of Autoconf. -*- Autoconf -*-
| # Version of Autoconf.
| # Copyright 1999, 2000 Free Software Foundation, Inc.
|
| m4_define([AC_ACVERSION], [2.49b])
|
| acversion.m4.in:
|
| # This file is part of Autoconf.
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
>> 67. ./syntax.at:27 FAILED near `syntax.at:27'
Pavel> AC_F77_WRAPPERS
I suppose you have a Fortran compiler on this machine, but not GNU.
Autoconf has indeed weaknesses wrt Fortran support.
>> 80. ./syntax.at:40 FAILED near `syntax.at
| On Wed, Nov 29, 2000 at 07:59:57PM +0100, Akim Demaille wrote:
| : Err, I think I should stop working on Autoconf right now:
| :
| : | for ac_file in `ls conftest.o conftest.obj conftest.* 2>/dev/null`; do
| : | case $ac_file in
| : | *.o | *.obj ) ac_cv_objext=`e
| How about doing something along the same lines for exeext:
|
| for ac_file in `ls conftest.exe conftest conftest.* 2>/dev/null`; do
| case $ac_file in
| conftest )ac_cv_exeext=;;
| *.$ac_cv_object ... ) ;;
| *)ac_cv_exeext=`expr "$ac_file
, and actually better, but... ?
Index: ChangeLog
from Akim Demaille <[EMAIL PROTECTED]>
* acspecific.m4 (_AC_EXEEXT, _AC_OBJEXT): Make the two macros more
alike: a loop over a sorted list of possible files.
Don't cleanup on errors, the trap will do it.
C
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> Hello, David!
>> I haven't checked the code yet but the behavior seems to be that
>> the torture tests are requiring me to hit ctrl-D as the old suite
>> required on every test.
Pavel> I don't think I have anything to do with this
| =
| Testing suite for autoconf, version 2.49b
| =
| testing AC_FUNC_MKTIME
| 80. ./syntax.at:40...
| ./syntax.at:40: autoconf --autoconf-dir .. -l $at_srcdir
| ./syntax.at:40: autoheader --autoconf-dir .. -l $a
| testing AC_FUNC_MKTIME
| 80. ./syntax.at:40...
| ./syntax.at:40: autoconf --autoconf-dir .. -l $at_srcdir
| ./syntax.at:40: autoheader --autoconf-dir .. -l $at_srcdir
| ./syntax.at:40: top_srcdir=$top_srcdir ./configure
| --- empty Wed Nov 29 19:34:07 2000
| +++ stderr Wed Nov 2
> "Daniele" == Daniele Arena <[EMAIL PROTECTED]> writes:
Daniele> Hi Akim, Beware! These are two different machines and
Daniele> OSs. The gzipped file I sent you is for Solaris8/SPARC, while
Daniele> the plaintext test 80 is the *only* failed test for BSDI3.1
Daniele> . Please tell me if you
> "Harlan" == Harlan Stenn <[EMAIL PROTECTED]> writes:
Harlan> I suspect that AC_EGREP_CPP is using K&R mode instead of ANSI
Harlan> mode under HP-UX.
Harlan> I have a test that attempts to determine how the C compiler
Harlan> will catify tokens and it's botching with CC=cc under HP-UX.
Har
> "Daniele" == Daniele Arena <[EMAIL PROTECTED]> writes:
Daniele> Ehm... Sorry, but now there's something I don't
Daniele> get. Really. Is it in any way possible that the tests give
Daniele> different results if run with different shells?...
Sure! That would reveal bugs in Autotest. But it
| This test will report `' on Cygwin caused by some Cygwin magic. Even though
| the file is named `conftest.exe', doing `ls conftest' will return `conftest',
| and doing `test -f conftest' will return true. Doing a plain `ls' will not
| return that virtual file, though. You therefore need to "
> "Daniele" == Daniele Arena <[EMAIL PROTECTED]> writes:
Daniele> What exactly is to be fetched by the PATH? (Just out of
Daniele> curiosity).
Just everything :) :) sed, expr, awk, cc and the like.
>>>>> "Alexandre" == Alexandre Oliva <[EMAIL PROTECTED]> writes:
Alexandre> On Nov 30, 2000, Akim Demaille <[EMAIL PROTECTED]> wrote:
>> | +./configure: 24794 Alarm call
>> This is why it failed: the shell complains of the signal it
Here is what I'm applying.
Index: ChangeLog
from Akim Demaille <[EMAIL PROTECTED]>
* doc/autoconf.texi (Systemology): New section.
Some about QNX 4.
2000-11-30 Akim Demaille <[EMAIL PROTECTED]>
Index:
> "Earnie" == Earnie Boyd <[EMAIL PROTECTED]> writes:
Earnie> HTH,
It does, thanks. But then, you are referring to EXEEXT as a *build*
feature, not a host feature.
Frankly, I think we will never end having problems until we introduce
the two sets. I can't see why EXEEXT and OBJEXT should
I've applied what Lars suggested.
> "Jim" == Jim Meyering <[EMAIL PROTECTED]> writes:
Jim> But that might be a little too involved a change to make this
Jim> close to a test release. Especially considering that this
Jim> happens only on systems with a broken mktime function. At worst,
Jim> all it does is cause a little conf
> "Daniele" == Daniele Arena <[EMAIL PROTECTED]> writes:
Daniele> OK, I give up. I reran all the tests with plain Bourne shell,
Daniele> and I obtained Yet Another Different Set of failed checks.
Good news :(
Daniele> If you really want, I can send you the tests with sh, bash
Daniele> and t
> "Pavel" == Pavel Roskin <[EMAIL PROTECTED]> writes:
Pavel> I don't know what "trap" you are talking about,
In the case we AC_MSG_ERROR.
Pavel> but the testsuite fails after this patch because "conftest" is
Pavel> not removed e.g. by AC_PROG_CC
Arg, I forgot to adjust my patch, I had not
| I think you can safely change the as_expr test into
|
| if expr a : '\(a\)' >dev/null; then
| as_expr=expr
| ekse
| as_expr=false
| fi
Just being applied, you'll be able to give it a try within a minute.
Please, could you try this:
(exit); echo $?
Have you tried running twice the check with twice the same shell?
Does it give consistent results?
Still on the same Solaris machine:
(false; exit); echo $?
It seems that some of your troubles are due to a bad propagation of
config.status' failures to configure.
801 - 900 of 1987 matches
Mail list logo
