On Mon, Jun 9, 2008 at 5:41 PM, Stefan de Konink <[EMAIL PROTECTED]> wrote:
> Tom Hughes schreef:
>> In message <[EMAIL PROTECTED]>
>>           Stefan de Konink <[EMAIL PROTECTED]> wrote:
>>
>>>>>>> - Assures the client always get the id back
>>>>>> That is already assured by the current API definition.
>>>>> For 'updates' too? According to spec only an id is send after a create.
>>>> Well you don't need it on an update as you already know the ID.
>>> Not if you create a changeset with updates and creates. Should the
>>> client maintain which object was a create and which one was an update?
>>> Now that sounds like making it difficult for the clients, instead of
>>> just returning all id's in the diff/changeset.
>>
>> Unsurprising the changeset spec addresses these issues.
>
> You mean this?
>
> When downloading data the client will receive a version number. This
> number must be stored and provided when updating (or deleting); if it
> does not match the server's current verision number at the time the
> update/delete is made, the operation will fail.
>
>
> I don't really see how changesets are going to address all changes at
> the same time + returning version numbers. Since it is not an atomic
> operation, how can one reliably getting these id's before a /close ?


Now you're confusing a changeset with the diff upload call. The diff
upload call /is/ atomic. You can have multiple diff upload calls per
changeset.
The changeset spec in the 0.6 API includes all of this information.

>
>
> Anyway it gets of the original point, which I would like to have solved
> first.
>


I think it has been solved: just implement the API the way it is.

Dave

_______________________________________________
dev mailing list
dev@openstreetmap.org
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/dev

Reply via email to