TermsStatsFacets

Elastic Stack Elasticsearch
1

Hi,

the filter is not recognized in the TermsStatsFacets ** this is a bug
I guess.

One more question: when using "doc.score" the result for my data is a
bit disappointing. E.g. when no query string is specified all docs
have score == 1 ... so, is it possible to access the document position
in the result and use it within the script?

I would like to give values which comes 'more earlier' in the result a
greater influence and sort the facets against that. This way when the
user sorts the result also the facets will be reordered and show
probably more meaningful values.

Regards,
Peter.

**

2

On Wednesday, February 23, 2011 at 6:04 PM, Karussell wrote:
Hi,

the filter is not recognized in the TermsStatsFacets ** this is a bug
I guess.
Pushed a fix, it was not added when using the Java API.

One more question: when using "doc.score" the result for my data is a
bit disappointing. E.g. when no query string is specified all docs
have score == 1 ... so, is it possible to access the document position
in the result and use it within the script?
Document position does not mean much with the way Lucene works. You can access a timestamp and work with it if you have it.

I would like to give values which comes 'more earlier' in the result a
greater influence and sort the facets against that. This way when the
user sorts the result also the facets will be reordered and show
probably more meaningful values.

Regards,
Peter.

**
TermsStatsFacetsShouldAllowFilter.java · GitHub

3

thanks * 2 :slight_smile:

timestamp is a good hint

On 23 Feb., 18:39, Shay Banon shay.ba...@elasticsearch.com wrote:

On Wednesday, February 23, 2011 at 6:04 PM, Karussell wrote:

Hi,

the filter is not recognized in the TermsStatsFacets ** this is a bug
I guess.

Pushed a fix, it was not added when using the Java API.

One more question: when using "doc.score" the result for my data is a
bit disappointing. E.g. when no query string is specified all docs
have score == 1 ... so, is it possible to access the document position
in the result and use it within the script?

Document position does not mean much with the way Lucene works. You can access a timestamp and work with it if you have it.