views:

634

answers:

3

I am trying to write a web-app to manage references for my PhD thesis.

I used to manage this information inside a personal Confluence (fantastic tool! - http://www.atlassian.com/software/confluence/) instance however I'm fed-up with the opening of PDF's and cutting and pasting values into fields that I wish to record.

I have exposed a webservice that will return me images based on a PDF filename and a page number. The same webservice also exposes a method that will return the text inside of a provided rectangle (top left x-y coord, bottom right x-y coord).

I would like to be able to drag a rectangle over part of the PDF image and then call the webservice to give me the text (which I will then store on a EntityBean). I am looking at using the JBoss application stack (Application Server, Hibernate, Seam and Richfaces). Does anybody know how I could go about achieving this? I have seen the ability to draw custom images in other RIA toolkits (e.g. dojo), but I can't see a way of doing this inside of Richfaces.

Hopefully somebody out there could prove me wrong, or provide some idea about what I can do (as I am not a web developer - I'm mainly building this tool because the RIA frameworks available now have got me interested!)

I already have the code to extract the text, my problem is purely how can I get the user to draw a "selection rectangle" inside the web browser over the top of the image?

Many Thanks,

Aidos

+1  A: 

Try using the RichFaces Paint 2D tag

It exposes the Graphics2D package to the user interface.

Track user drag events on the image using javascript, then post the co-ordrdinates to the backing bean to re-render the image with a drawn on selection box.

Andy MacGilvery
A: 

Have you considered Mendeley ? It will try to parse and extract bibliographic information from your pdfs.

anders.norgaard
A: 

you can do it with itext (http://www.lowagie.com/iText/)

SomaSekhar