From: | Bill Studenmund <wrstuden(at)netbsd(dot)org> |
---|---|
To: | Peter Eisentraut <peter_e(at)gmx(dot)net> |
Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, <pgsql-patches(at)postgresql(dot)org> |
Subject: | Re: Patch to add Heimdal kerberos support |
Date: | 2001-11-14 22:18:57 |
Message-ID: | Pine.NEB.4.33.0111141403550.29259-100000@vespasia.home-net.internetconnect.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-patches |
On Wed, 14 Nov 2001, Peter Eisentraut wrote:
> Tom Lane writes:
>
> > Couldn't we do this in a way that doesn't require a user configure switch?
>
> We've seen a Heimdal patch before that had the same source changes but
> solved the configury in some other ugly way. It's not trivial to write a
There is another test we can do. There is a variable which contains the
Heimdal version number, '__heimdal_version'. If that symbol is present,
then we have Heimdal, and if it's not, we have MIT.
So something like 'nm /usr/lib/libkrb5.a | grep __heimdal_version' is a
good test.
> macro that checks for struct members. However, that other patch didn't
> need to link in all these extra libraries as this patch tries; it only
> used -lkrb5. There is only one Heimdal I trust?
Yes, there is only one Heimdal. However some linkers are better with
inter-library dependencies than others. NetBSD 1.5, the OS I've used for
developing this, doesn't handle them automagically. So you have to
explicitly include them. I think -current and 1.5.2 have this fixed, I'm
just stuck in the stone age. ;-)
Something like --with-extralibs would suffice to cover this. But the extra
libs would need to be used for compiling the kerberos test cases. :-)
I would appreciate help with this. I think for now the difference in API
needs to be supported, but everything else about the config setup can be
changed as desired. :-)
Take care,
Bill
From | Date | Subject | |
---|---|---|---|
Next Message | Bill Studenmund | 2001-11-14 22:42:37 | Re: Patch to add Heimdal kerberos support |
Previous Message | Dave Cramer | 2001-11-14 20:01:07 | Patch to fix getIndex for forte's transparent persistence |