lucene-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Joseph Lewis <joehm...@gmail.com>
Subject Re: How to query a lucene index from JavaScript (no server or jvm)?
Date Wed, 14 Nov 2012 22:45:45 GMT
If you're just looking for something that is that small of a scale, and on
that small of a set of input files, you might to think about just writing
some kind of script (using something like beautiful soup to extract the
text from pages), that would then create a Lucene-like index, meaning your
script could look at some kind of table for the word the user input, and
get a list of documents back that had that list.

In JSON it might look something like this:

{
   "word1":["http://example.com/page1","http://example.com/page54"],
   "word2":["http://example.com/page54"],
   ...
}

This would have the benefit of the format already being parsable through
Javascript, although you'd probably want to cut down the size of the URLs
you kept by storing them in their own object with some kind of primary key.

If, for some other reason, you wanted to use Lucene, I believe the
"simpletext" codec will save the Lucene index in something that would be
easier for you to work with:
http://lucene.apache.org/core/4_0_0/codecs/org/apache/lucene/codecs/simpletext/package-summary.html

This is a newly introduced feature to 4.0 if I remember correctly, meant as
a debugging feature.


- Joseph | http://josephlewis.net


On Wed, Nov 14, 2012 at 8:38 AM, Carlos de Luna Saenz <
cdelunasaenz@yahoo.com.mx> wrote:

> JavaFX is your friend and WebView is the ally
>
>
>
>
>
> >________________________________
> > De: Paul Jungwirth <pj@illuminatedcomputing.com>
> >Para: general@lucene.apache.org
> >Enviado: Martes, 13 de noviembre, 2012 5:59 P.M.
> >Asunto: Re: How to query a lucene index from JavaScript (no server or
> jvm)?
> >
> >> I have been asked to "develop a HTML page with a Search textbox to
> search
> >a
> >collection of HTML files without using a Server"
> >
> >That sounds very tricky! This is client-side Javascript, right? I don't
> >know how JS running in someone's web browser could access your Lucene
> index
> >without a server making it available. If you want an easy server solution,
> >Solr and ElasticSearch are both good options that expose Lucene to remote
> >clients, with web-based APIs you could call directly from client-side
> >Javascript. Or you could send the JS calls to whatever you're using to
> >serve the HTML pages, and have that use Lucene/Solr/ES/etc. behind the
> >scenes.
> >
> >Good luck,
> >Paul
> >
> >
> >On Fri, Nov 9, 2012 at 6:23 AM, dcgoodwin <dennis.goodwin@sas.com> wrote:
> >
> >> I have been asked to "develop a HTML page with a Search textbox to
> search a
> >> collection of HTML files without using a Server". Lets say the number of
> >> HTML files are less than 500 and the file size is less than .5 Mbyte.
> >>
> >> I am creating a Lucene index from the set of HTML files from Java when I
> >> generate the HTML files from XML source files. I am not seeing how to
> query
> >> the Lucene index from JavaScript only (no server of jvm). I am thinking
> I
> >> need to create 1 or more XML files from the Lucene index and use
> JavaScript
> >> to load-up the XML files to perform a search.
> >>
> >> Please let me know if this sounds reasonable given the constraints and
> if
> >> folks have tried this before.
> >>
> >>
> >>
> >>
> >> --
> >> View this message in context:
> >>
> http://lucene.472066.n3.nabble.com/How-to-query-a-lucene-index-from-JavaScript-no-server-or-jvm-tp4019318.html
> >> Sent from the Lucene - General mailing list archive at Nabble.com.
> >>
> >
> >
> >
> >--
> >_________________________________
> >Pulchritudo splendor veritatis.
> >
> >
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message