[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: LDIF and structural objectclasses



Michael Ströder wrote:
Daniel Tiefnig wrote:
 > 1) Why are userpassword attributes always encoded in base64 in

I guess that's because userPassword is defined with SYNTAX 1.3.6.1.4.1.1466.115.121.1.40 (Octet String) in OpenLDAP's schema.

Hmm, good point, allthough I thought, that base64 coding is just used for output in LDIF syntax, and SYNTAX definition shouldn't interfere with LDIF in that way, should it? I have to think about that a little bit, i think. :o)


Note that it's up to the LDIF creator which notation to use. Any attribute MAY be base64-encoded and your LDIF parser has to handle that.

Well, that's okay for me, I was just wondering why it is this way..

If you wanna parse LDIF use an appropriate module for your favourite programming language which correctly handles all the stuff defined in RFC2849 (e.g. module ldif in package python-ldap...:-)

Ah, no. I'm happily using PerLDAP without any problems so far, it's just that I'm writing up a short introduction to LDAP (and OpenLDAP) to be held at Linuxdays[0] here in Graz, (Austria) and I found those mentioned things a little bit, uhm, confusing. :o)



lg, daniel

[0] http://linuxtage.at (german only)