Re: LDAP issues (and some fixes)

[Pawel Salek <pawsa TheoChem kth se>]

On 2001-04-29 02:17 Thomas LeRoux wrote:
> COMPILATION PROBLEMS: First off, I'm using OpenLDAP 2.0.7, and it
> seems as if the Balsa configuration script does not include
> libresolv.so (-lresolv) when compiling its LDAP test program.  Not
> sure why, but I'm guessing this is a new requirement for OpenLDAP.  

I have just tested linking against openldap-2.0.7 on my RH Linux box
(ix86), and haven't experienced any problems. could it be it is
distribution depedent (you didn't mention if you use linux or other unix
system)? In principle, I don't see a problem with adding -lresolv, I
would just like to know clearly why it has to be done.

> DISPLAY NAME PROBLEMS: Once compiled, the Address book worked fairly
> well when querying my LDAP directory.  However, once I selected a name
> from the list, the display name on the email only shows "Lastname
> <foobar@host.com>"
> instead of "Firstname Lastname <foobar@host.com>".  I took a quick
> look, and it seems as if Balsa is not using the "cn" (Common Name)
> attribute for the display name.  It's using the "sn" (Surname) and
> "givenName" (First name) attributes concatenated.  However, since "cn"
> is a required attribute, and since it was always intended to be the
> first name, I feel this attribute is more correct (since I don't even
> use the "givenName" attribute, I was only getting last names).  I
> think someone else was having this problem as seen in the list
> archives.  This fix is included in the patch attached.

This sounds right. If I don't get any objections from Berend, I will
commit this part.

>   I believe this also resolves bug # 29254.

In bugzilla.gnome.org!? I could not find it.

> CACHE PROBLEMS: Ahhggg!!!  This one drove me NUTS until I (sortof)
> anyway, as any LDAP delays will not affect the app thread.  That is, a
> 60
> second LDAP query won't freeze the app.  Regardless, I've commented
> out the
> client side cache in my patch - feel free to ignore this change.

Actually, I recall that I had some problems with caching as well when I
tested this part of the code - but it was long time ago and I don't
remember much. But the symptoms seem familiar.

Thanks a lot for your comments and for the patch.

/Pawel
-- 
Pawel Salek (pawsa@theochem.kth.se) http://www.theochem.kth.se/~pawsa/
Theoretical Chemistry Division, KTH voice: +46 8 790-8202