I have documents of below structure
{
user_id: 'user_id',
tags: ['tag_1', 'tag_2', 'tag_3', 'tag_1'],
time: 123 // timestamp
}
I used below query to get all users which has a list of tags
{
"query": {
"bool": {
"must": [],
"filter": {
"terms": {
"tags": [
"tag_1",
"tag_6"
]
}
}
}
},
"aggs": {
"users": {
"terms": {
"field": "user_id",
"size": 100
}
}
}
}
This query return user_id with total document count as below
{
"aggregations": {
"users": {
"doc_count_error_upper_bound": 9882,
"sum_other_doc_count": 435400587,
"buckets": [
{
"key": 34611, // user_id
"doc_count": 1026
}
]
}
}
}
Actually I need to get sum of items actually occurred in the documents as below
{
"aggregations": {
"users": {
"doc_count_error_upper_bound": 9882,
"sum_other_doc_count": 435400587,
"buckets": [
{
"key": 34611, // user_id
"sum": 9897 // in total 9897 times tag_1 and tag_6 from 1026 documents
}
]
}
}
}
I can't find solution for this query. Can anyone help me in this.