Should llUpdateKeyValue() succeed with no matching key?

[Qie Niangao]

The dataserver event triggered by llUpdateKeyValue() returns success even if no such key exists before nor after the Update.

I guess a database update of a non-existent row may return without raising an error, but it seems weird that llUpdateKeyValue() not only signals success, but returns the value supplied to the call, implying (to me) that the value was stored successfully.

I can deal with the current behavior, but I wonder if this is really how it was intended to work.

[Lucia Nightfire]

IMO, we're certainly not emulating SQL with KVP, but that depends on what DBA you ask, heh.

[Dolphin Linden]

That would be a bug

The intent is that an unchecked update will create keys which do not exist.