getFullTextParser

[ckarthikin]

Hi, I digged into deep and got the results..thanks..

Is it possible extract the fulltextparser output then what is need to be changed in the below code. getFullTextParser s not working..

var parser = this.engine.getHeaderParser();
var biblioItem = new BiblioItem(this.lexicon);
var result = (string)parser.processing2(fileName, false, biblioItem).getLeft();

Best Regards,
karthikc

[boumenot]

I took a quick look at this, and this is not supported out of the box. I do not have any plans to implement this any time soon.

I think the following things need to be implemented for this work.

1. Sanitize the FullTextParser, and all of other engines that it instantiates. Ensure that dependencies flow through instead of FullTextParser reaching out for global state. The existing work should be a good guide for how to implement this.
2. Clean up any uses of GrobidProperties.

You will need to patch the my fork of Grobid, and then compile it with IKVM, and replace the existing Grobid.Java.dll.

[ckarthikin]

Hi thanks for the reply and guidance to proceed my work...i have one doubt regarding "You will need to patch the my fork of https://github.com/boumenot/grobid, and then compile it with IKVM, and replace the existing Grobid.Java.dll." i am not clear on this point. sorry to disturb u..plz explain...

[boumenot]

You need to take the resulting Java binary, and produces C# assembly. This is done using IKVM. I can't explain IKVM, but you should be able to pick it with a little bit of effort like I did.

There's a build script in my fork, as well as the command I called to transpile Java to C#. They will need to be adjusted to work on your machine, but they should be enough to get you started.

I re-built the Wapiti binary for official version of Grobid, and we've verified that it works on Windows. If you want the most complete functionality this is the place to start. My fork is incomplete.