Reputation: 783
I've got an index like:
[
{
"Name": "Alex",
"LastName": "Ich",
"Department": 2
},
{
"Name": "Charlie",
"LastName": "Sheen",
"Department": 3
},
{
"Name": "Peter",
"LastName": "Petrelli",
"Department": 5
},
{
"Name": "Alan",
"LastName": "Harper",
"Department": 6
},
{
"Name": "Ann",
"LastName": "Bottle",
"Department": 3
},
]
And I want to get the results with distinct Department
, I don't care about order, just 1 result per Department
. I tried with aggregations but I could only manage to get the different Deppartments
with the doc_count associated. They query I tried is something like:
{
"aggs": {
"deppartments": {
"terms": {
"field": "Department"
}
}
},"size": 0
}
It returns:
"buckets": [
{
"key": 2,
"doc_count": 1
},
{
"key": 3,
"doc_count": 2
},
{
"key": 5,
"doc_count": 1
},
{
"key": 6,
"doc_count": 1
},
]
When I want something like:
[
{
"Name": "Alex",
"LastName": "Ich",
"Department": 2
},
{
"Name": "Charlie",
"LastName": "Sheen",
"Department": 3
},
{
"Name": "Peter",
"LastName": "Petrelli",
"Department": 5
},
{
"Name": "Alan",
"LastName": "Harper",
"Department": 6
}
]
Upvotes: 0
Views: 54
Reputation: 12672
You can use Top hits aggregation for this
{
"aggs": {
"departments": {
"terms": {
"field": "Department",
"size": 10
},
"aggs": {
"search_results": {
"top_hits": {
"size": 10 <--- you can change the size to 1 if you want
}
}
}
}
},
"size": 0
}
Does this help?
Upvotes: 1