Luke/LIMO - how to "surf" query results

[Kai_testing Middleton <ka...@yahoo.com>]

I'm interested in playing with varying use-cases in my corpus: I want to throw a lot of queries at it.  I have used nutch to gather about 132000 pages (as reported by Luke) and want to throw queries against these documents using Luke (or possibly LIMO - haven't tried that one yet).  Unfortunately for me, Luke's UI is not optimized for quick browser viewing of document results.  For example, when I perform a query in Luke I can't just double click on a resulting row to have it come up in the browser -- Luke is built as a general Lucene index tool and doesn't assume the documents represent web pages.  I have to double click the result row to get the detail view, click the URL field, click the button to copy that to the clipboard, paste it into the URL field of the browser, delete lots of extraneous characters from the URL line, then press enter.  This makes it more cumbersome to try lots of URLs quickly, the way one might do with a google search.  

Does anyone know of a better way?  Would LIMO be quicker for this since it's browser-based?




      ____________________________________________________________________________________
Shape Yahoo! in your own image.  Join our Network Research Panel today!   http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7 

Re: Luke/LIMO - how to "surf" query results

[Renaud Richardet <re...@apache.org>]

hi Kai,

I don't understand why you can't use the web-interface of Nutch (or 
opensearch interface), can you explain why?

Alternatively, Luke has extensions -- maybe you can work it out this way...

- Renaud


Kai_testing Middleton wrote:
> I'm interested in playing with varying use-cases in my corpus: I want to throw a lot of queries at it.  I have used nutch to gather about 132000 pages (as reported by Luke) and want to throw queries against these documents using Luke (or possibly LIMO - haven't tried that one yet).  Unfortunately for me, Luke's UI is not optimized for quick browser viewing of document results.  For example, when I perform a query in Luke I can't just double click on a resulting row to have it come up in the browser -- Luke is built as a general Lucene index tool and doesn't assume the documents represent web pages.  I have to double click the result row to get the detail view, click the URL field, click the button to copy that to the clipboard, paste it into the URL field of the browser, delete lots of extraneous characters from the URL line, then press enter.  This makes it more cumbersome to try lots of URLs quickly, the way one might do with a google search.  
>
> Does anyone know of a better way?  Would LIMO be quicker for this since it's browser-based?
>
>