[wildfly-dev] RA subsystem: renaming <admin-object> "pool-name" attribute to "name"

[Brian Stansberry]

On Mon, Jul 30, 2018 at 3:32 AM, Tomas Hofman <thofman at redhat.com> wrote:

> Yes, that's what I believe should be done.
>
> Adding also Tomasz and Tom :).
>
> I'm yet to asses the impact on documentation, by my feeling from my
> initial investigation was that I didn't found very many documents dealing
> with this.
>

Please check with the Teiid folks, as Teiid heavily uses this subsystem.


>
> Am I correct in thinking that if the new parser is able to parse the old
> config version, no other migration work is needed?
>

Right, unless Stefano wants the parser for the new schema version to be
forgiving and allow the old attribute.


> Tomas
>
> On 19/07/18 22:38, Brian Stansberry wrote:
>
>> So basically the issue here is that the xml attribute should be called
>> 'name' instead of 'pool-name'?
>>
>> Sounds like a minor Enhancement not a major Bug. If there's much in the
>> way of docs out there that use pool-name then the cost of changing it
>> (wrong docs) may outweigh any benefit.
>>
>> On Thu, Jul 19, 2018 at 9:24 AM, Tomas Hofman <thofman at redhat.com
>> <mailto:thofman at redhat.com>> wrote:
>>
>>     The <admin-object> and <connection-definition> elements in resource
>> adapters
>>     subsystem have "pool-name" attribute that looks like it isn't used for
>>     anything, which is misleading for users.
>>
>>     It looks that "pool-name" attribute was intended for functionality
>> that wasn't
>>     implemented. The attributes are only present in XML, and do not exist
>> in
>>     management model.
>>
>>     During resource creation the values are passed into service value
>> objects
>>     (ModifiableAdminObject, ModifiableConnDef), but #getPoolName()
>> methods are not
>>     called from anywhere.
>>
>>     The attributes can't be simply removed because their values are used
>> for
>>     resource addressing, e.g.
>>
>>         /subsystem=resource-adapters/.../admin-objects=test-a-o:add(...)
>>
>>     will produce
>>
>>         <admin-object ... pool-name="test-a-o"/>
>>
>>     so some "name" attribute is still needed.
>>
>>     Unless you think that this is not worth having new schema version (or
>> the
>>     intended functionality that requires "pool-name" attrs is planned to
>> be
>>     implemented), I would create new XSD schema version with "pool-name"
>>     renamed to
>>     "name" and update the parser. I suppose the new XSD version should be
>> 6.0,
>>     rather than 5.1, no matter how small the change.
>>
>>     Also, AFAIK this change couldn't be backported to released product
>> streams.
>>
>>     The issue where this was raised is
>>     https://issues.jboss.org/browse/JBEAP-15023
>>     <https://issues.jboss.org/browse/JBEAP-15023>
>>
>>     --     Tomas Hofman
>>     Software Engineer, JBoss SET
>>     Red Hat
>>     _______________________________________________
>>     wildfly-dev mailing list
>>     wildfly-dev at lists.jboss.org <mailto:wildfly-dev at lists.jboss.org>
>>     https://lists.jboss.org/mailman/listinfo/wildfly-dev
>>     <https://lists.jboss.org/mailman/listinfo/wildfly-dev>
>>
>>
>>
>>
>> --
>> Brian Stansberry
>> Manager, Senior Principal Software Engineer
>> Red Hat
>>
>
> --
> Tomas Hofman
> Software Engineer, JBoss SET
> Red Hat
>



-- 
Brian Stansberry
Manager, Senior Principal Software Engineer
Red Hat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/wildfly-dev/attachments/20180730/f577d925/attachment.html