Re: About a bug of the camel-mongodb component...

2014-04-16 Thread Babak Vahdat
Hi Ràul, Now the other ticket (CAMEL-7370) has been fixed as well. And regarding the reason why I went for ReadPreference#valueOf() instead of ReadPreference#[primary/secondary/nearest/etc.] (that's the +0.5 points you gave about it): - The logic to be implemented in this way is much easier, as

Re: About a bug of the camel-mongodb component...

2014-04-15 Thread Babak Vahdat
Well, I don't see any extra advantage when using ReadPreference#[primary/secondary/nearest/etc.] utilities compared to ReadPreference#valueOf() as: - ReadPreference#valueOf() is also part of the public API, the same as ReadPreference#[primary/secondary/nearest/etc.] API - ReadPreference#[primary/s

Re: About a bug of the camel-mongodb component...

2014-04-15 Thread Raul Kripalani
Hi Babak, You're right. On revisiting this logic it seems like this functionality was indeed unfinished. Even in the case where the Read Preference could be resolved, we would still throw an exception. I don't think there's much discussion here. Backwards-compatibility is not beneficial here becau

Re: About a bug of the camel-mongodb component...

2014-04-15 Thread Babak Vahdat
Hi Raúl, Thanks a lot for your feedback but it's still not clear to me how this URI option used to work properly in the previous versions as from very first day (Camel 2.10.0) the preexisting logic used to throw an IllegalArgumentException *unconditionally* in all cases at the end of the method, s

Re: About a bug of the camel-mongodb component...

2014-04-15 Thread Raul Kripalani
Hello Babak, Thanks for pointing this out! Just some background... The ReadPreference class in the MongoDB Java API has changed substantially between the driver versions: * The initial contribution of camel-mongodb was done against driver version 2.7.3 [1]. * Currently we are on driver version 2