Nicolas Maisonneuve wrote:
> my version doesn't allow nested lucene field because lucene
> index is a flat
> structure of fields.
> The official versionof LuceneIndexTransformer is a fulltext
> indexation.. It
> indexes the content (and not the structure) in a general field named
> "content", there are also link field and text-attr but it's optionaly)
I see. So this is why you say your version is more "low-level", with fields that map
directly to the Lucene index.
> With my version you can choose which information you index in
> user specific
> fields. (much better to the relevance)
> so i'ts not a really structured indexation, but you can
> search a solution ,
> for example field name= xpath name
> <tag1>
> <tag2>coco</tag2>
> <tag3coco2</tag3>
> </tag1>
> with my luceneTransformer
>
> <field name="tag1.tag2">coco</field>
> <field name="tag1.tag3">coco2</field>
OK, so with the existing transformer you could directly index this document:
<tag1>
<tag2>coco</tag2>
<tag3>coco2</tag3>
</tag1>
You would get:
field value
tag1 coco coco2
tag2 coco
tag3 coco2
With your transformer you would need to transform it first into something like:
<field name="tag1">coco coco2</field>
<field name="tag2">coco</field>
<field name="tag3">coco2</field>
Is that right?
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
