Delete nested array in elasticsearch

问题

I have documents containing nested tables in the following format :

{
"dataId": "dataIdentifier",
"versionId": "versionIdentifier",
"items": [
{
"obj1": "value1",
"obj2": "value2",
"obj3": "value3",
 },
 {
"obj1": "value4",
"obj2": "value5",
"obj3": "value6"
}, 
{
"obj1": "value7",
"obj2": "value8",
"obj3": "value9" 
}        
]
}

With "items" being of nested type. I want to delete one of the elements inside the "items" given than one of the object inside has a certain value. For instance it would be :

if (obj1 == "value4" & dataId == "dataIdentifier" & versionId == versionIdentifier) :
    DELETE TABLE ENTRY

In this case I want to delete the second entry of "items". I tried to do it with update_by_query, and my attempt was :

q = {
   "query": {
    "match_all": {}
    }
},
"script": {
    "lang" : "painless",
    "inline" : {if (ctx._source.dataId == item.dataId && ctx._source.versionId == item.versionId && ctx._source.items.obj1 == item.obj1) {ctx._source.items.remove()}}",
    "params" : {
        'dataFrame': [{
            "dataId" : 'myDataIdList',
            "versionId" : 'myVersionId',
            "obj1" : 'myValue'
            } ]
    }
 }    
}

es.update_by_query(index=myindex, body=q)

But I do not know how to designate the concerned entry in the "ctx._source.items.remove()" argument. I took a look at the Elasticsearch documentation but could not find what I was looking for. Any help would be greatly appreciated.

回答1:

With update by query you can't remove "a field", you can replace it, for that, you can update it with others two values you want to mantain or delete the document and index a new with that two values

来源：https://stackoverflow.com/questions/54148466/delete-nested-array-in-elasticsearch

标签

python

ElasticSearch

elasticsearch-py