On 10/30/16 10:50 AM, Jindřich Mynarz wrote:
>
> the behaviour you describe is what I'd like to be able to rely on. (In
> fact, it's the opposite of the Anytime query.) Updates should either
> be executed completely or fail transparently, so that the incomplete
> execution may be handled.
>
> I wil
Hi,
the behaviour you describe is what I'd like to be able to rely on. (In
fact, it's the opposite of the Anytime query.) Updates should either be
executed completely or fail transparently, so that the incomplete execution
may be handled.
I will have to test this properly to ensure that partial u
Hi Jindřich,
As Lorenz indicates an update should either be successful or fail especially if
it is to be part of a transaction (or even distributed transaction) where
commit or rollback is required if any part fails.
Do you have a specific use case for a partial SPARQL update ?
Best Regards
Hu
Hello Jindřich,
to my understanding, the assumption is that any SPARQL update will be
executed completely, otherwise it fails with some error messages or the
like.
Cheers,
Lorenz
> Hi Hugh,
>
> thanks, that supports my assumption. Are you aware of any other
> mechanism to tell if a SPARQL upda
Hi Hugh,
thanks, that supports my assumption. Are you aware of any other mechanism
to tell if a SPARQL update was executed completely? For some types of
updates, it is possible to rephrase the WHERE clause of the update to ASK
query expected to return false. However, many updates do not work that
