[Date Prev][Date Next]
Re: Accessing LDAP with v2 and v3 simultaneously (ITS#922)
At 11:46 AM 12/9/00 +0000, email@example.com wrote:
>Full_Name: Olaf Schlüter
>Submission from: (NULL) (220.127.116.11)
>If I upload a LDIF File containing a node with UTF8 formatted strings
>(like cn: Schlüter) I can retrieve it correctly for example using netscape,
>does recognize OpenLDAP as capable of LDAPv3 and does a correct UTF8 -> latin1
>conversion before displaying the search results. But I noticed if I query using
>"ldapsearch -P2 -x" that I receive the attributes still in UTF8. I believe they
>be recoded to T.61 according to the LDAP v2 spec.
Note that is not possible to transliterate all UTF-8 strings to
T.61 strings (and vice versa). Transliteration can only do this
for a subset of strings. OpenLDAP only does this for IA5, a common
subset of both UTF-8 and T.61. You'll note that some commercial
dual implementations take a similar stance on transliteration.
Though additional character set transliteration could be done, note
that this is only one difference between LDAPv2 and LDAPv3. There
are many others, as discussed in this I-D:
Basically, the bottom line is that complete dual implementations are very
difficult, if not impossible, to develop. There are no plans to fully
undertake such development as LDAPv2 is considered by many to be obsolete.
In fact, the IETF is currently undertaking work which, if completed as
planned, will move the LDAPv2 specification to "historical" (and hence
obsolete) status during the next year.