[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
Re: ldap v2/v3 interoperability
Hi Roland,
Roland Hedberg wrote:
>
> Hi Patrick !
>
> To emphasize Haralds point about character sets and real world
> implementations;
> in the Tisdag project I have found myself forced to add a code segment to the
> ldapserver that tries to figure out what characterset a v2 client uses and
> expects
> to get the answer in. Clearly this is not actually doable only a educated
> guess
> can be attempted.
>
> -- Roland
In fact , as Helmut Volpers already said on this list, we also built in
our V3 Server a flexibility allowing to configure which character set
will be used within a V2 session. The server then expects the requests
in this character set, and converts the results in this character set
too. A kind of static negociation between client and server. But, can it
be the rule that servers have to support all clients available on the
market, speaking all possible or impossible languages within a V2
session?
Note that, to comment Chris´s first question, the way the server stores
the data should be independent of the character set used in the add
request if possible. It has not to be utf8 because the add request was a
V3-one.
Patrick
--
--------------------------------------------------------------------
Patrick Fantou Tel: +49 89 636 41203
Siemens Nixdorf Informationssysteme AG Fax: +49 89 636 45860
SNI ITN COM
Otto-Hahn-Ring 6, D-81739 Munich, Germany
e-mail: Patrick.Fantou@mch.sni.de
--------------------------------------------------------------------