automake
[Top][All Lists]
Advanced

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

Re: FYI: same library installed several times in same directory (PR/350)


From: Harlan Stenn
Subject: Re: FYI: same library installed several times in same directory (PR/350)
Date: Sun, 20 Oct 2002 18:50:02 -0400
User-agent: EMH/1.10.0 SEMI/1.13.7 (Awazu) FLIM/1.13.2 (Kasanui) XEmacs/21.1 (patch 12) (Channel Islands) (i386--freebsd)

> Please write to the lists, not to me.

Sorry - I thought I did.

>  Harlan> This bug still bites me in 1.7 and 1.7.1 .
> 
> :(
> 
>  Harlan> I notice that the automake tests for this bug are
>  Harlan> SKIPped when I run "make check"; what do I need to do
>  Harlan> to enable these tests?
> 
> libtool5.test and libtool6.test?  
> They require libtoolize to be in your PATH.

It is...

> Try to run the test manually,
> 
>   cd tests; ./libtool5.test; echo $?
> 
> that may help to understand why it is skipped (SKIP means the script
> exits with $? = 77)

I think I found the problem.

At the site where I have this problem I install the auto* tools using mpkg,
which lets me have multiple versions of the auto* tools.

This means I am still using the following patch which has not been applied
to the main tree:

--- aclocal.in~     Fri Feb  1 20:40:48 2002
+++ aclocal.in        Wed Jun 12 22:24:33 2002
@@ -215,6 +215,9 @@
      exit 0;
     }
 
+    # Add directories from ACLOCAL_PATH
+    push @dirlist, split ':', $ENV{ACLOCAL_PATH};
+
     # Search the versioned directory near the end, and then the
     # unversioned directory last.  Only do this if the user didn't
     # override acdir.

When I "select" the version of libtool I want to use it sticks the path to
its share/aclocal in ACLOCAL_PATH.

I (believe I) need to use this mechanism instead of the recently-implemented
dirlist solution because dirlist changes behavior system-wide, and I need
per-shell control over the versions of software that will be used.

I'll hack around this problem locally.

> It'd be nice to modify one of these tests to exhibit your issue.

I'm working on it now.

H




reply via email to

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