[Date Prev][Date Next]
[Chronological]
[Thread]
[Top]
[ldapext] RE: [XEDDEV] LDIFv2
>Andrew,
Hi Kurt! Thanks for your comments.
>
>I'm a bit concerned by the use of whitespace to indicate
>continuations in LDIFv2 XML values, especially in the last
>line of the value. I think this choice will make it
>especially hard on those who create LDIF by hand, often by
>example (instead of by reading the specification), to
>get it right.
>
I think you may be right and will incorporate this into my next draft.
Below is your suggestion:
dn: cn=Andrew Sciberras,o=Adacel,c=AU
objectClass: organizationalPerson
attr;transfer-rxer>:
><?xml version=3D'1.0'?>
> <value>
> <item>250 Bay Street</item>
> <item>Brighton Victoria 3186</item>
> <item>Australia</item>
> </value>
>sn: Sciberras
Incorporating your changes, this should actually look like this:
dn: cn=Andrew Sciberras,o=Adacel,c=AU
objectClass: organizationalPerson
attr;transfer-rxer>:
><?xml version='1.0'?>
> <value>
> <item>250 Bay Street</item>
> <item>Brighton Victoria 3186</item>
> <item>Australia</item>
> </value>
>
sn: Sciberras
The difference (ignoring the fact that I've removed the 3D text after
'version=' ) is that "sn: Sciberras" is on a new line. And that the value
ends with a line that contains '>' followed by a new line
character/sequence.
>I note as well that nothing in this extension is XML specific,
>the extension is useful for incorporation of values which are
>represented in any multi-line format.
This is correct, assuming that values can all be represented in UTF8
excluding the NULL character.
The intention of this draft is to specifically allow XML encoded information
to be displayed in its encoding, rather than BASE64, because XML encodings
are text based.
Allowing this draft to be open-ended in what it applies to (i.e. including
non text based encodings which meet the aforementioned requirements) will be
counter to my original intention. From my point of view, a text encoding
(i.e. XML) is more presentable than BASE64, which in turn is more
presentable than non-ASCII character encoding.
>
>Kurt
>
Cheers,
Andrew.
_______________________________________________
Ldapext mailing list
Ldapext@ietf.org
https://www1.ietf.org/mailman/listinfo/ldapext