[infinispan-dev] Retrieval operations with the IGNORE_RETURN_VALUES flag

[Dan Berindei]

I agree, eventually we should have separate getAndXxx methods like JSR-107.
The Jokre is cool, but running with an instrumentation agent is never going
to be ok for everyone.

I was thinking about this in the context of AtomicHashMap. It uses
withFlags(SKIP_REMOTE_LOOKUP, DELTA_WRITE, FORCE_WRITE_LOCK) for write
operations, but it still has to keep a reference to the original cache for
read operations. Of course, after more digging, I don't think the
SKIP_REMOTE_LOOKUP flag is that useful for writes either (since we already
have a copy of the map in the invocation context at that point). But it
seemed like an interesting idea in the general case.



On Mon, Jun 3, 2013 at 1:15 PM, Sanne Grinovero <sanne at infinispan.org>wrote:

> ha got it, good point.
> but I'm not persuaded: doesn't it get even more confusing for users? Imho
> it would be more helpful to throw an exception.
>
> these flags are confusing and ideally we should evolve the API, as the JSR
> did, or push on The Jokre.
> On 3 Jun 2013 11:08, "Dan Berindei" <dan.berindei at gmail.com> wrote:
>
>> Sanne, I'm only talking about get operations. I was thinking that if you
>> call cache.get(key), you want the value of that key, regardless of where it
>> is stored...
>>
>> Obviously, write operations would still behave as they do now.
>>
>>
>> On Mon, Jun 3, 2013 at 12:59 PM, Sanne Grinovero <sanne at infinispan.org>wrote:
>>
>>> Hi Dan,
>>> I'm not sure I understood this. How can I prevent it to return values
>>> if you have the flag ignored? Note that in some cases it makes a huge
>>> performance difference.
>>>
>>> Sanne
>>>
>>> On 3 June 2013 10:52, Dan Berindei <dan.berindei at gmail.com> wrote:
>>> > Hi guys
>>> >
>>> > CacheLoaderInterceptor and DistributionInterceptor both honour the
>>> > IGNORE_RETURN_VALUES flag for get commands, but I think it would be
>>> more
>>> > useful if they ignored it - just like they ignore it for conditional
>>> > commands.
>>> >
>>> > That would make it possible for users to only keep a reference to a
>>> > cache.getAdvancedCache().withFlags(IGNORE_RETURN_VALUES) and use it
>>> for both
>>> > read and write operations.
>>> >
>>> > What do you think?
>>> >
>>> > Cheers
>>> > Dan
>>> >
>>> >
>>> > _______________________________________________
>>> > 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
>>>
>>
>>
>> _______________________________________________
>> 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
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/infinispan-dev/attachments/20130603/93ba37b1/attachment.html