--On Sunday, July 08, 2012 7:22 PM +0200 Philip Brusten philip.brusten@icts.kuleuven.be wrote:

Anyone?

On 27/06/2012 16:17, Philip Brusten wrote:

...

Hi,

Our OpenLDAP information is stored in UTF-8 character encoding, like any other default installation I guess. Some of the records have special characters in their name, like ë, é, è, ç, etc.

We want to make a webapplication that can do lookups on these names, without forcing the user to enter the correct characters. E.g. if they search for "Francois", it should match all givenNames "François" in LDAP. However if people do search for "François" if should also match all givenNames "François".

I've looked for certain matching rules that could do this, but I haven't found what I'm looking for so far.

Can anybody help me out?

Approx indexing perhaps, which uses soundex?

Alternatively, you could try adding multiple values pairs, like

givenName: Francois givenName;fr: François

etc.

--Quanah

--

Quanah Gibson-Mount Sr. Member of Technical Staff Zimbra, Inc A Division of VMware, Inc. -------------------- Zimbra :: the leader in open source messaging and collaboration

On 8/07/2012 20:44, Quanah Gibson-Mount wrote:

Approx indexing perhaps, which uses soundex?

We've tried this, but then '(givenName~=Francois)' matches a lot of other names like: Véronique, Franklin, Frank, Veronica, Franco, Faranak, etc. So this doesn't seem useful for us.

Alternatively, you could try adding multiple values pairs, like

givenName: Francois givenName;fr: François

Adding a tagging option for every character encoding independent attribute seems a bit drastic and redundant. Is there really no other way?

Kr,

Philip