Ok, that makes sense.
Right now, the Catalog does not revert anything if an exception is thrown.
I'll look into improving this behavior.

Torben

On Thu, Nov 17, 2016 at 5:53 PM, Justin Deoliveira <[email protected]>
wrote:

> Hey Torben
>
> Someone can correct me if I'm wrong and we never quite formalized this but
> here is my take/understanding. If the catalog throws an exception
> deliberately then it should be responsible for leaving/restoring
> modifications accordingly to the same state as before that method was
> called. If that doesn't happen I would say it's a bug.
>
> The catalog isn't however responsible for restoring any state "across
> methods". For instance if adding a resource, then a layer and the layer
> method throws an exception. Removing the resource would be the
> responsibility of the client.
>
> $0.02
>
> -Justin
>
>
> On Thu, Nov 17, 2016 at 12:29 PM Torben Barsballe <
> [email protected]> wrote:
>
>> I was looking into GEOS-4754, and discovered that the documentation
>> around "CatalogException" is rather unclear.
>>
>> Justin - you seem to be the original author of this code, so if you have
>> any insight it would be helpful.
>>
>> My main question is: Is the Catalog responsible for reverting any changes
>> it may have made when a CatalogException is thrown, or is this a
>> responsibility of the client?
>>
>> More details here: https://osgeo-org.atlassian.net/browse/GEOS-7865
>>
>> Thanks
>>
>
------------------------------------------------------------------------------
_______________________________________________
Geoserver-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

Reply via email to