Alright, I finally managed to limit the resulting set of documents by adding
a size keyword :
{
"fields":[
],
"size":0,
"from":0,
"sort":[
"_score"
],
"query":{
"filtered":{
"query":{
"bool":{
"should":[
{
"multi_match":{
"query":"general",
"fields":[
"Titre.Valeur.original^4",
"Titre.Valeur.partial",
"Resume.Valeur.original^2",
"Resume.Valeur.partial",
"Corps.Valeur.original^2",
"Corps.Valeur.partial"
]
}
}
],
"minimum_should_match":1
}
}
}
},
"aggs" : {
"Count":{
"cardinality":{
"field":"IDDocument"
}
},
"doc_limit_agg" :{
"filter" : {
"script" : {
"script" : "_doc.score <0.05"
}
},
"aggs" : {
"docagg" : {
"terms" : {
"field" : "IDDocument",
"order" : { "version_max>score"
: "desc" }
,"size":4
},
"aggs" : {
"version_max" : {
"filter" : { "range" :
{ "NumeroVersion" : { "gt" : 0 } }},
"aggs" : {
"vMax" : {
"max" : { "field" : "NumeroVersion" }},
"score" : {
"max" : { "script":"_doc.score"}}
}
}
}
}
}
}
}
}
Pagination could then be addressed based on highest score order and its
filtered value (see script above).
Last question, and yet maybe something not supported by ES : is there a way
to dynamically calculate a column (like a ROW_NUMBER() in SQL) and use it
to sort and filter like above. Actually I'm seeking a better and more robust
way to filter/sort than the score column.
Again the SQL equivalent to illustrate what I'm aiming at :
with VersionMax as (
select d.IDDocument, max(d.NumeroVersion) as NumeroVersion, max(d.Score)
as ScoreDocument,
ROW_NUMBER() OVER(ORDER BY max(d.Score) DESC, d.IDDocument DESC) as
RowNumber
from @Document d
group by d.IDDocument
)
select top 2 *
from VersionMax v
where v.RowNumber < 4
order by v.RowNumber desc
--
View this message in context:
http://elasticsearch-users.115913.n3.nabble.com/ES-aggregation-and-pagination-tp4052774p4052906.html
Sent from the ElasticSearch Users mailing list archive at Nabble.com.
--
You received this message because you are subscribed to the Google Groups
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/1395929238409-4052906.post%40n3.nabble.com.
For more options, visit https://groups.google.com/d/optout.
