Re: attributes tag

[Jon Roberts <jon@jonanddeb.net>]

Peter Marschall wrote:
> I was tempted to using tags for some attributes to store simplified names
> e.g. (This is for demonstration purposes withput the required encodings)
>
> sn: Müller
> sn;x-ascii: Mueller
>
> which would have helped us very much in our multi-language environment.
> (It is hard to key in Müller if you do not have the umlaut-u on your keyboard)
>
> Unfortunately I found no way to get the value of the base attribute alone
> (Müller in the example above) which is necessary for displaying the name 
> correctly.
> Querying for sn I always got the two values back without any possibility
> to distinguish which value beongs to the bae attribute.

This might depend on your client or client libraries. I use the language 
tags with JLDAP, and I am able to distinguish the values appropriately 
with the correct label. For example with an entry like:

dn: gospid=GVMattC01V01, ou=Gospel, ou=Expressions, o=mentata.com
objectclass: top
objectclass: gosp
gospid: GVMattC01V01
gospqualifier: 400101
gospowner: cn=Matthew, ou=Bible, ou=People, o=mentata.com
gospname: Matthew 1:1
gosptype: verse
gospdescription: The book of the generation of Jesus Christ, the son of 
David, the son of Abraham.
gospdescription;lang-en: An account of the genealogy of Jesus the 
Messiah, the son of David, the son of Abraham.
gospdescription;lang-la: liber generationis Iesu Christi filii David 
filii Abraham
gospdescription;lang-af: Die geslagsregister van Jesus Christus, die 
seun van Dawid, die seun van Abraham:
gospdescription;lang-de: Dies ist das Buch von der Geburt Jesu Christi, 
der da ist ein Sohn Davids, des Sohnes Abrahams.
gospdescription;lang-el: ?????? ???????? ????? ??????? ???? ????? ???? 
??????
gospdescription;lang-es: LIBRO de la generación de Jesucristo, hijo de 
David, hijo de Abraham.
gospdescription;lang-sq: Libri i gjenealogjisë së Jezu Krishtit, birit 
të Davidit, birit të Abrahamit.
gospdescription;lang-sv: Detta är Jesu Kristi, Davids sons, Abrahams 
sons, släkttavla.
gospdescription;lang-sw: Yesu Kristo alikuwa mzawa wa Daudi, mzawa wa 
Abrahamu. Hii ndiyo orodha ya ukoo wake:
gospdescription;lang-zh: ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 
? ? ? ? ? . ? ?
 ? ? ? ? ? ? ? ? ? ? ? ?

If I ask for the values of attribute gospdescription;lang-de, I get a 
single element array with:

Dies ist das Buch von der Geburt Jesu Christi, der da ist ein Sohn 
Davids, des Sohnes Abrahams.

If I ask for the values of attribute gospdescription (ie. no tag) I get 
as single element array with:

The book of the generation of Jesus Christ, the son of David, the son of 
Abraham.

In other words, if I request the attribute gospdescription in a search, 
while I may be retrieving the attribute in all languages over the wire 
(I believe I am), the JLDAP libraries can parse them out with their 
explicit names.

One thing I will say is that I can't search on gospdescription and only 
gospdescription. I can search the other language attributes exclusively, 
but when I try search on gospdescription exclusively, the directory 
finds matches in other languages. This isn't a complaint, because I 
believe this works as designed, but it's an interesting gotcha that may 
inform your design decisions.

Jon Roberts
www.mentata.com