[ 
https://issues.apache.org/jira/browse/CAMEL-17481?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Claus Ibsen updated CAMEL-17481:
--------------------------------
    Priority: Minor  (was: Trivial)

> camel-caffeine: Various improvements
> ------------------------------------
>
>                 Key: CAMEL-17481
>                 URL: https://issues.apache.org/jira/browse/CAMEL-17481
>             Project: Camel
>          Issue Type: Improvement
>          Components: camel-caffeine
>    Affects Versions: 3.11.5
>            Reporter: Andreas Klug
>            Priority: Minor
>
> Can you pls. comment on a pull request I'm going to create for the 
> camel-caffeine component regarding the following changes?
> +InvalidateAll+
> The caffeine library provides two "invalidateAll" methods, one with no 
> arguments and one taking a key set. Currently the camel-caffeine component 
> maps only the invalidateAll method providing a key set, defaulting to an 
> empty set if no key set is provided explictly. This essentially means that 
> there is no way to invalidate all keys of the cache, as providing an empty 
> key set to that very invalidateAll method leaves the cache unchanged. It 
> might make sense to call the invalidateAll method without parameters in case 
> no key set is provided, i. e. the appropriate header field is not set.
> +AsMap+
> The caffeine library provides a asMap method, which isn't yet mapped by 
> camel-caffeine. This essentially means that there is no way to get the list 
> of all keys present in the cache.
> +Cache lookup+
> I've got difficulties to understand how the cache reuse is supposed to work. 
> I'm only able to get the caffeine component to work using the .toF() 
> notation, as indicated by the unit test "testTo()", which I added to 
> CaffeineCacheFromScratchProducerTest. This unit test basically takes the 
> example from the camel-caffeine documentation using "to()" and it fails using 
> the current implementation. I added a 
> getCamelContext().getRegistry().bind(...) statement to the endpoint in case a 
> new cache is generated, which allows the unit test to succeed. Is that the 
> way it should be? Am I missing something?
> +Create cache if not exist+
> Somehow related to the cache lookup topic above, I wasn't able to find a 
> reference to the createCacheIfNotExist configuration property. I added a 
> check if that property is set. If it's set to true, then the cache is created 
> as already implemented. If not, an exception is thrown.
>  
>  
>  



--
This message was sent by Atlassian Jira
(v8.20.1#820001)

Reply via email to