[Standards] XEP-136 and XEP-59 implementation comments: threading

Alexander Tsvyashchenko lists at ndl.kiev.ua
Sat May 10 07:03:27 UTC 2008


Hello Peter,

>> 1) JIDs matching: the conclusion seemd to be to use 'exactMatch'
>> attribute (?), but as you've pointed out this is the problem not only
>> for XEP-136.
>
> I added that:
>
> http://www.xmpp.org/extensions/xep-0136.html#collections-with

Ah, sorry - it seems I missed that :-(

Some comments then ;-)

> If the client or server wishes to match an exact bare JID, the boolean
> 'exactmatch' attribute MUST be included and MUST be set yo "true" or "0" [7].

1) I think there is a typo, last phrase should read as "true" or "1".

2) I would not restrict it to bare JIDs: instead it seems to me more  
logical to write smth similar to "if the client or server wishes to  
match an exact bare JID or domain JID, the boolean 'exactmatch'  
attribute MUST be included and MUST be set yo "true" or "1", thus JIDs  
like example.com matching exactly example.com instead of  
*@example.com/*, and JIDs like user at example.com matching exactly  
user at example.com instead of user at example.com/*" - the latter one may  
be important as some collections may be recorded with bare JIDs like  
user at example.com, so we might need to access them somehow.

3) 'exactmatch' is currently present only in 'chat' element: however,  
it doesn't look like it belongs there, as 'chat' element is in fact  
not used for matching. Instead, it should belong to commands  
description (such as 'remove' or 'list' elements) and in preferences  
(such as 'item' element).

Maybe it should be specified in some separate sub-section that  
'exactmatch' attribute may be present in every element where JID is  
included and 'matching' makes sence?

4) XML Schema probably should be updated too?

Good luck!                                     Alexander

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.





More information about the Standards mailing list