How to compare 2 field in elasticsearch - elasticsearch

Question

Ok, I have example result on my data in elastic search :
"hits" : [
{
"_index" : "solutionpedia_data",
"_type" : "doc",
"_id" : "nyODP24BA840z5O6WguE",
"_score" : 46.63439,
"_source" : {
"ID" : "1",
"PRODUCT_NAME" : "ATM",
"UPDATEDATE" : "13-FEB-18",
"PROPOSAL" : [
{
}
],
"MARKETING_KIT" : [ ],
"VIDEO" : [ ]
}
},
{
"_index" : "classification",
"_type" : "doc",
"_id" : "5M-r5m4BNYha4zuWalJa",
"_score" : 39.25268,
"_source" : {
"productId" : "1",
"productName" : "ATM",
"productIconUrl" : "media/8ae0f0c3-1402-4559-901e-7ec9b874ce68-prod032.webp",
"type" : "nonconnectivity",
"businessLineId" : "",
"subsidiaries" : "",
"segment" : [],
"productType" : "Efisien",
"tariff" : null,
"tags" : [ ],
"contact" : [],
"mediaId" : [
"Med391"
],
"documentId" : [
"doc260",
"doc261"
],
"createdAt" : "2019-09-22T05:22:46.956Z",
"updatedAt" : "2019-09-22T05:22:46.956Z",
"totalClick" : 46
}
}
]
this is a result of my alias. can we search for the same data based on 2 different fields, the example above is the ID and productId fields. Can we make these 2 objects in one bucket or compare?
i was try with some aggregate but nothing :
{
"query": {
"match_all": {}
},
"size": 0,
"aggregations": {
"product catalog": {
"terms": {
"field": "productId.keyword",
"min_doc_count": 2,
"size": 100
},
"aggregations": {
"product solped": {
"terms": {
"field": "ID.keyword",
"min_doc_count": 2
}
}
}
}
}
}
result :
{
"took" : 9,
"timed_out" : false,
"_shards" : {
"total" : 10,
"successful" : 10,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : 1276,
"max_score" : 0.0,
"hits" : [ ]
},
"aggregations" : {
"product catalog" : {
"doc_count_error_upper_bound" : 0,
"sum_other_doc_count" : 0,
"buckets" : [ ]
}
}
}

Answer 1

You can achieve this with a Scripted Bucket Aggregation, using script logic to define your buckets (pseudo code: if field a exists value of field a, if field b exists value of field b).
Another (and better) way to achieve this is to change your data model and indexing logic on Elasticsearch side and store the information in a field of the same name.
You could also consider the alias data type to make fields with different names in different indices accessible under one common field name. This is also the approach Elastic takes with the Elastic Common Schema specification.