You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@uima.apache.org by Richard Eckart de Castilho <re...@apache.org> on 2017/11/30 22:22:23 UTC

Diverging JCas classes and type system in v3

Hi,

in v2, the following is possible:

* Assume you have JCas class "X" with a feature "a"
* Define a type system with the class "X" and two features, "a" and "b"
* initialize a CAS with this type system
* Create/access instances of X via the JCas wrapper and access "a" through getters and setters
* Create/access instances of X via the CAS API and access "b" through the CAS API

v3 now has a tighter coupling between JCas and CAS, right? Is the above still possible?

Best,

-- Richard

Re: Diverging JCas classes and type system in v3

Posted by Marshall Schor <ms...@schor.com>.

yes, this is still fully supported.

Should be no problem.

-Marshall


On 11/30/2017 5:22 PM, Richard Eckart de Castilho wrote:
> Hi,
>
> in v2, the following is possible:
>
> * Assume you have JCas class "X" with a feature "a"
> * Define a type system with the class "X" and two features, "a" and "b"
> * initialize a CAS with this type system
> * Create/access instances of X via the JCas wrapper and access "a" through getters and setters
> * Create/access instances of X via the CAS API and access "b" through the CAS API
>
> v3 now has a tighter coupling between JCas and CAS, right? Is the above still possible?
>
> Best,
>
> -- Richard