Thanks for your reply Alexandre and Mikhail.
If you are indexing the parent/child as a block, then you know number
of children when you are indexing parent and can store that as a
field. That would be most efficient.
- I can achieve that when I do my re-indexing.
When searching for children you can assign ^=1 per every hit, then
you can aggregate these hits with {!parent ... score=total}, then
you can cut off parents exceeding number of child limit with {!frange}.
- Just to check, is this done during query time?
Regards,
Edwin
On 30 August 2016 at 14:51, Mikhail Khludnev <m...@apache.org> wrote:
> Hello,
> When searching for children you can assign ^=1 per every hit, then
> you can aggregate these hits with {!parent ... score=total}, then
> you can cut off parents exceeding number of child limit with {!frange}.
>
>
> On Tue, Aug 30, 2016 at 5:38 AM, Zheng Lin Edwin Yeo <edwinye...@gmail.com
> >
> wrote:
>
> > Hi,
> >
> > Would like to check, in the parent/child nested document, is it possible
> to
> > return only records with a certain number of child?
> > For example, I want to return all the parent's records which has more
> than
> > 100 child.
> > Is it possible to be done?
> >
> > I'm using Solr 6.1.0
> >
> > Regards,
> > Edwin
> >
>
>
>
> --
> Sincerely yours
> Mikhail Khludnev
>
