[infinispan-dev] BUILD FAILED

[Sanne Grinovero]

On 21 May 2012 09:34, Galder Zamarreño <galder at redhat.com> wrote:
>
> On May 18, 2012, at 2:22 PM, Sanne Grinovero wrote:
>
>> It seems I'm unable to use any configuration to configure a CacheLoader.
>
> ^ Trust me, that was the most painful thing of migrating inner configuration to new config API.

Yes I can see that, great work. Still it wasn't ready, this should not
have been integrated yet as it's wasted a lot of my time and the
counter is still going.

>
> It is doable though, but you need to somehow mimmic what you'd do with XML, here's an example:
> https://github.com/galderz/infinispan/blob/master/core/src/test/java/org/infinispan/distribution/DistSyncTxCacheStoreSharedTest.java#L68
>
> So, add a cache loader, and then set individual properties via setProperty. That works for all cache stores.

Clearly Jenkins isn't on the same page..

>
> This should improve once each module can provide its own config parser, and we can finally provide proper programmatic and XML configuration for cache loaders…etc:
> https://issues.jboss.org/browse/ISPN-1978
>
> That, alongside proper fluent config for cache stores:
> https://issues.jboss.org/browse/ISPN-1524

Yes that would all be awesome, still until it's done this locked me
out from trying my new implementations.

>
>> I'm having better luck if I do all of this:
>> - force my CacheLoader to implement CacheStore (didn't want to do that yet)
>
> ^ Shouldn't be necessary, we have other cache loader only impls out there, such as ClusterCacheLoader

I agree it *shouldn't* but I'm getting a ClassCast exception as the
code is assuming all CacheLoaders are CacheStore as well.

I'm stuck.

>
>> - use the programmatic configuration, and *legacy* deprecated code only
>> - avoid to configure any option - I'm having to hardcode what I need
>> it do to in the CacheLoader implementation
>
> ^ setProperty should work.

Only for CacheStore.

>
>> Only then I'm able to test my things; in fact most of the new failures
>> are related to CacheLoader tests.
>>
>> On 17 May 2012 14:50, Galder Zamarreño <galder at redhat.com> wrote:
>>> Some of this could be result of migrating to the new configuration in master, I'll have a look over next few days.
>>>
>>> On May 17, 2012, at 12:56 AM, Sanne Grinovero wrote:
>>>
>>>> Build 198 [1] introduced 25 failures, nearly doubling the amount to
>>>> 43. Ok, business as usual.. In fact, no worries I have recently
>>>> configured a mail filter "from jenkins -> SPAM" as it was too painful.
>>>>
>>>> But this time it broke the build of the Lucene Directory!
>>>> I'll have my revenge, très-bien froide.
>>>>
>>>> 1 - https://infinispan.ci.cloudbees.com/job/Infinispan-master-JDK6-tcp-NON_PARALLEL/198/#showFailuresLink
>>>>
>>>> _______________________________________________
>>>> infinispan-dev mailing list
>>>> infinispan-dev at lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/infinispan-dev
>>>
>>> --
>>> Galder Zamarreño
>>> Sr. Software Engineer
>>> Infinispan, JBoss Cache
>>>
>>>
>>> _______________________________________________
>>> infinispan-dev mailing list
>>> infinispan-dev at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/infinispan-dev
>>
>> _______________________________________________
>> infinispan-dev mailing list
>> infinispan-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/infinispan-dev
>
> --
> Galder Zamarreño
> Sr. Software Engineer
> Infinispan, JBoss Cache
>
>
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/infinispan-dev