Marco
2006/6/5, Peter Sparkes <[EMAIL PROTECTED]>:
Question 1
Rank 0 means it is the hit with the highest score
Rank 1 means it is the hit with the second highest score
etc
If there is only one hit it is always rank 0
Question 2
Have a look at the cocoon 2.1.9 Luvene sample.
If you want to search for "foo" in "title" your search string should be "title:foo"
Question 3
Try
<map:match pattern="search.htm">
<map:generate type="search">
<map:parameter name="index" value="luceneindex"/>
</map:generate>
<map:transform type="log"/>
<map:serialize type="xml"/> <!--Look at the resulting xml and write your lucene.xslt to give you what you want-->
<map:transform src=""> <map:serialize/>
</map:match>
Peter
Hello again,
3 further questions
- is there any documentation on how Lucene calculates the rank score?
- 'rank' always starts with 0. How can I make it start with 1 ?
- let's say, my input document is like that:
<lucene:document url="" href="http://google.com" target="_blank" >http://google.com">
<title lucene:store="true" >Linktext</title>
</lucene:document>
how can I afterwards access the value of 'title'?
<xsl:template match="search:hit" >
<a target="_blank" href="" PROTECTED]">
<xsl:value-of select="uri"/> <--instead, I want the value of the stored title tag, in this case 'Linktext'
</a>
</xsl:template>
The doc (http://cocoon.apache.org/2.1/userdocs/concepts/xmlsearching.html ) says
<xsl:value-of select="search:[EMAIL PROTECTED]:name='title']"/> , but that doesn't work.
Any ideas?
Thank you very much
Marco
2006/6/2, [EMAIL PROTECTED] < [EMAIL PROTECTED] >:Thank you very much Peter. The problem is solved
2006/6/2, Peter Sparkes <[EMAIL PROTECTED]>:Hi,
Suggest use the LuceneIndexTransformer as follows:
<map:match pattern="buildIndex">
<map:generate src=""> <map:transform type="index"/>
<map:serialize type="xml"/>
</map:match>
<map:match pattern="karte.xml">
Generate a single file that contains all the data that you want
indexed as in section "Input document for the LuceneIndexTransformer" in
http://wiki.apache.org/cocoon/LuceneIndexTransformer
<map:serialize type="xml"/>
</map:match>
Peter
> Hi,
>
> I'd like to use the Lucene search in my page, but reading the docs at
> http://cocoon.apache.org/2.1/userdocs/concepts/xmlsearching.html and
> http://wiki.apache.org/cocoon/LuceneIndexTransformer does not really
> enlighten me.
>
> My case:
> The content of my pages is generated dynamically, getting the data
> from a mysql database. Something like that,
>
> <map:match pattern="karte">
> <map:generate type="file" src=""> > <map:transform type="sql">
> <map:parameter name="use-connection" value="mysql-pool"/>
> <map:parameter name="lfdnr" value="{request-param:lfdnr}"/>
> </map:transform>
> <map:transform src=""> > <map:serialize type="html"/>
> </map:match>
>
> So calling e.g. the address
> http://localhost:8080/cocoon/karte?lfdnr=412 shows me the results of
> the page with the sql primary key 412 transformed via style.xsl.
>
> Lucene should be able to index the columns 'lfd_nr', 'venue' and
> 'description' (without indexing the other 6 columns of the table) and
> should give me the results, if those fields match my search query
> obviously, like shown above (
> http://localhost:8080/cocoon/karte?lfdnr='pk'
> <http://localhost:8080/cocoon/karte?lfdnr=%27pk%27 >)
>
> Declaring the LuceneIndexTransformer of my sitemap is the only thing I
> have completely understood from the wiki. ;)
> A little guidance would be appreciated very much. (+ what to write in
> which files, since that does not always emanate from the wiki)
>
> Thanks in advance
> Marco
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
