2011/11/6 Paolo Castagna <[email protected]>: > Hi, > I have never used Apache Stanbol and I'd like to try it with a small > HelloWorld Java example to learn how to interact/use Stanbol > programmatically rather than via REST / HTTP. Is it possible? > > I found a list of interesting "Enhancement Engines" [1], can someone > help me (or point me at if there is one) to write a small HellWorld > Java example with Stanbol? > > Let's say I have some text and I want to try the "Named Entity > Extraction Enhancement Engine" or the "Named Entity Tagging Engine" > how do I do this with a Java program (not via REST / HTTP)?
Hi Paolo, There is no such documentation yet unfortunately. In order to achieve this you will have to startup a felix runtime (or any other OSGi runtime such as Eclipse Equinox for instance) and deploy the InMemoryJobManager, the engines you mention, the entityhub and the dbpedia referenced site configuration and their dependencies (you have the exact list of bundles in the list.xml file of the stable launcher), then wrap your text into a content item and use the job manager. We should definitely write an "embedding Stanbol in a Java application" how to but this has not been done yet. > PS: > I had a quick look at the sources, there are many modules [2]... is > there a map? > By the way, I checked Stanbol out and I was able to install it (via > mvn clean install... 24 minutes :-)), but mvn package failed for me. What kind of error did you get? Which version of mvn are you using? -- Olivier http://twitter.com/ogrisel - http://github.com/ogrisel
