Elasticsearch Copy_to data need to copied self subdocument - elasticsearch

Question

Thanks in advance for helping.
I have created ES mapping as :
{"mappings": {
"policy": {
"properties": {
"name": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"tags": {
"properties": {
"scope": {
"type": "text",
"store": "true",
"copy_to": [
"tags.tag_scope"
]
},
"tag": {
"type": "text",
"store": "true",
"copy_to": [
"tags.tag_scope"
]
},
"tag_scope": {
"type": "text",
"store": "true"
}
}
}
}
}
}
}
When i index policy document all tag and scope value from different tags document copied to tag_scope property.
For Example I added a document as to elastic search:
{
"name": "policy1",
"tags": [
{
"tag": "pepsi",
"scope": "prod"
},
{
"tag": "coke",
"scope": "dev"
}
]
}
It is storing all 4 values as in tag_scope documents as:
"tags.tag_scope": [
"pepsi",
"test",
"coke",
"dev"
]
My Exceptions was, it should store like :
{
"name": "policy1",
"tags": [
{
"tag": "pepsi",
"scope": "prod",
"tag_scope" : ["pepsi","prod"]
},
{
"tag": "coke",
"scope": "dev",
"tag_scope" : ["coke","dev"]
}
]
}
Could you please help me to do correct mapping for same?

Answer 1

What you are looking for is Nested Datatype. Change your mapping to the below:
PUT <your_index_name>
{
"mappings":{
"policy":{
"properties":{
"name":{
"type":"text",
"fields":{
"keyword":{
"type":"keyword",
"ignore_above":256
}
}
},
"tags":{
"type": "nested",
"properties":{
"scope":{
"type":"text",
"store":"true",
"copy_to":[
"tags.tag_scope"
]
},
"tag":{
"type":"text",
"store":"true",
"copy_to":[
"tags.tag_scope"
]
},
"tag_scope":{
"type":"text",
"store":"true",
"fields": { <---- Added this
"keyword": {
"type": "keyword"
}
}
}
}
}
}
}
}
}
Notice how I've made tags as nested type. This would allow the below to be stored as individual document itself, which in your case tags basically has two nested documents.
{
"tag":"coke",
"scope":"dev"
}
Now your tags.tag_scope should be what you are expecting it to be.
Now when it comes to querying for what you are looking for, the below is how a Nested Query should be.
Nested Query:
POST <your_index_name>/_search
{
"query": {
"bool": {
"must": [
{
"nested": {
"path": "tags",
"query": {
"bool": {
"must": [
{
"match": {
"tags.tag_scope": "pepsi"
}
},
{
"match": {
"tags.tag_scope": "prod"
}
}
]
}
}
}
}
]
}
}
}
As to return the list of unique tags.tag_scope value you would need to return aggregation query. Notice that I've mentioned size:0 which means I only want to see aggregation result and not normal query results.
Aggregation Query:
POST <your_index_name>/_search
{
"size":0,
"query":{
"bool":{
"must":[
{
"nested":{
"path":"tags",
"query":{
"bool":{
"must":[
{
"match":{
"tags.tag_scope":"pepsi"
}
},
{
"match":{
"tags.tag_scope":"prod"
}
}
]
}
}
}
}
]
}
},
"aggs":{ <----- Aggregation Query Starts Here
"myscope":{
"nested":{
"path":"tags"
},
"aggs":{
"uniqui_scope":{
"terms":{
"field":"tags.tag_scope.keyword",
"size":10
}
}
}
}
}
}
Aggregation Response:
{
"took": 53,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 1,
"max_score": 0,
"hits": []
},
"aggregations": {
"myscope": {
"doc_count": 2,
"uniqui_scope": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": "coke",
"doc_count": 1
},
{
"key": "dev",
"doc_count": 1
},
{
"key": "pepsi",
"doc_count": 1
},
{
"key": "prod",
"doc_count": 1
}
]
}
}
}
}
Hope this helps.