libtool-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH 366] Libtoolize now advises AC_CONFIG_MACRO_DIR use where app


From: Ralf Wildenhues
Subject: Re: [PATCH 366] Libtoolize now advises AC_CONFIG_MACRO_DIR use where appropriate.
Date: Thu, 24 Apr 2008 08:09:19 +0200
User-agent: Mutt/1.5.17+20080114 (2008-01-14)

* Gary V. Vaughan wrote on Thu, Apr 24, 2008 at 05:34:39AM CEST:
> On 23 Apr 2008, at 17:26, Ralf Wildenhues wrote:
>>
>> It seems that one of your patches causes this test failure:
>> --- expout      2008-04-23 22:41:22.000000000 +0200
>> +++ /home/ralf/libtool/write/build/tests/testsuite.dir/at-stdout    
>> 2008-04-23 22:41:22.000000000 +0200
>> @@ -1,3 +1,6 @@
>> +libtoolize: You should add the contents of the following files to  
>> `aclocal.m4':
>> +libtoolize:   `/home/ralf/libtool/write/build/../libtool/libltdl/m4/ 
>> libtool.m4'
>> +libtoolize:   `/home/ralf/libtool/write/build/../libtool/libltdl/m4/ 
>> lt~obsolete.m4'
>> libtoolize: Consider adding `AC_CONFIG_MACRO_DIR([m4])' to  
>> configure.ac and
>> libtoolize: rerunning libtoolize, to keep the correct libtool macros  
>> in-tree.
>> libtoolize: Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
>> 10. libtoolize.at:651: 10. verbatim aclocal.m4 w/o AC_CONFIG_MACRO_DIR 
>> (libtoolize.at:651): FAILED (libtoolize.at:759)

> Can you try again with a fresh checkout?  I can't reproduce here :-(

Still fails.  Here's a hint: when you run aclocal in libtoolize.at:692,
my aclocal.m4 file contains

-- snip --
# This should need upgrading:
# serial 25 LT_INIT
AC_DEFUN([LT_INIT],
[blah])

# This is newer than the upgrade version:
# serial 99999 ltoptions.m4

# This is older than the upgrade version:
# serial 1 ltversion.m4
-- snip --

both before and after running aclocal.  How is that sed script that runs
later going to change serials for libtool.m4 if aclocal.m4 doesn't
contain it at all?

> Maybe a difference in our installed automake and/or libtool versions?
> I have 1.10.1 and 2.2.3a master HEAD respectively.

Ahh.  When I use master automake instead of 1.10.1, I can reproduce the
succeeding of the test.  But say, why are you even running aclocal when
aclocal.m4 is hand generated?

Still doesn't explain though how you see this with 1.10.1.

Cheers,
Ralf




reply via email to

[Prev in Thread] Current Thread [Next in Thread]