Can I do this with Lucene ?

[fmarchioni <fm...@libero.it>]

Dear Lucene Users,
I'd like to implement a search engine and I'd like to know if it's possible to
do it with Lucene.
I have many text files containing faqs, every faq has a text file.
Inside the file questions are listed with a delimiter (I use # but it can by any)

#Question1
bla bla bla bla

#Question2
bla bla bla bla

...and so on....

Now I'd like something that search for words inside the faqs and RETURNS THE BLOCK
where it was found (#Question1 bla bla bla bla)....in other words I want to retrieve
a single faq using a search engine.

I didn't see anything like this in the docs....just something that return the number of hits
of a search.....can Lucene do it for me easily or it's better to write it on my own using
standard Java I/O Api ?
Thanks a lot
Francesco




---------------------------------------------------------------------
To unsubscribe, e-mail: lucene-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: lucene-user-help@jakarta.apache.org

Re: Can I do this with Lucene ?

[Eric Jain <Er...@isb-sib.ch>]

> I want to retrieve
> a single faq using a search engine.

You could index FAQs instead of files.

  for file in listFiles():
    for faq in extractFaqs(file):
      doc = Document()
      doc.add(Field.Keyword('key', faq.id))
      doc.add(Field.Keyword('file', file))
      doc.add(Field.Unstored('content', faq.content))
      writer.addDocument(doc)


> I didn't see anything like this in the docs....just something that
> return the number of hits of a search

One solution would be to create a custom HitCollector that simply counts
hits and then 'searcher.search(query, hitCollector)'.

--
Eric Jain


---------------------------------------------------------------------
To unsubscribe, e-mail: lucene-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: lucene-user-help@jakarta.apache.org

Re: Can I do this with Lucene ?

[Morus Walter <mo...@tanto-xipolis.de>]

Hi Francesco,

> I'd like to implement a search engine and I'd like to know if it's possible to
> do it with Lucene.
> I have many text files containing faqs, every faq has a text file.
> Inside the file questions are listed with a delimiter (I use # but it can by any)
> 
> #Question1
> bla bla bla bla
> 
> #Question2
> bla bla bla bla
> 
> ...and so on....
> 
> Now I'd like something that search for words inside the faqs and RETURNS THE BLOCK
> where it was found (#Question1 bla bla bla bla)....in other words I want to retrieve
> a single faq using a search engine.
> 
> I didn't see anything like this in the docs....just something that return the number of hits
> of a search.....can Lucene do it for me easily or it's better to write it on my own using
> standard Java I/O Api ?

Well lucene indexes units of text as you provide them.
So if you make each question a document in lucenes sense you will get 
the question and any info you stored with it (eg. faq-file and question
number).
Seems straight forward to me.

There is no need that files and documents in lucene have a 1:1 relation.

HTH
	Morus

---------------------------------------------------------------------
To unsubscribe, e-mail: lucene-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: lucene-user-help@jakarta.apache.org