Hi I am trying to get a query based on the following documents.
document 1
{
"date" : "2021-05-19",
"items" : [
{
"keyA" : "ValueA"
}
]
}
document 2
{
"date" : "2021-05-19",
"items" : [
{
"keyB" : "ValueB"
}
]
}
Output I expect
{
"date" : "2021-05-19",
"items" : [
{
"keyA" : "ValueA"
},
{
"keyB" : "ValueB"
}
]
}
I don't want to update this structure in ES. Only when I query, I want the result to be in this format.
Is it possible? or should I handle it after receiving the result?
You can use a combination of terms and top_hits aggregation
{
"size": 0,
"aggs": {
"dateagg": {
"terms": {
"field": "date"
},
"aggs": {
"top_keys": {
"top_hits": {
"_source": {
"includes": [
"items.*"
]
}
}
}
}
}
}
}
Search Result will be
"aggregations": {
"dateagg": {
"doc_count_error_upper_bound": 0,
"sum_other_doc_count": 0,
"buckets": [
{
"key": 1621382400000,
"key_as_string": "2021-05-19T00:00:00.000Z", // note this
"doc_count": 2,
"top_keys": {
"hits": {
"total": {
"value": 2,
"relation": "eq"
},
"max_score": 1.0,
"hits": [
{
"_index": "67598919",
"_type": "_doc",
"_id": "1",
"_score": 1.0,
"_source": {
"items": [
{
"keyA": "ValueA" // note this
}
]
}
},
{
"_index": "67598919",
"_type": "_doc",
"_id": "2",
"_score": 1.0,
"_source": {
"items": [
{
"keyB": "ValueB" // note this
}
]
}
}
]
}
}
}
]
}
}
Related
I have SQL query:
WHERE A = 1 AND B = 2 AND C REGEXP (eee|fff|ggg)
How can I write this query in elasticsearch ?
Not tested but something like this QUERY-DSL you need for your case with filter and regexp,
GET /_search
{
"size": 10,
"query": {
"filter" : {
"bool" : {
"must" : [
{ "term" : {"A" : 1}},
{ "term" : {"B" : 2}}
]
}
},
"regexp": {
"C": {
"value": "eee|fff|ggg"
}
}
}
}
OR
GET /_search
{
"size": 10,
"query": {
"filter": {
"bool": {
"must": [
{
"term": {
"A": 1
}
},
{
"term": {
"B": 2
}
},
{
"regexp": {
"C": {
"value": "eee|fff|ggg"
}
}
}
]
}
}
}
}
My sample JSON file for postman runner:
[ { "name": "runner", "hitler_id": "4006abc", "year": "2017", "boolean": "false", "expected": 717962 } ]
Pre request script:
var member = data.name; var booking = data.boolean; var fyyear = data.year; var sid = data.hitler_id;
console.log(data.name); console.log(data.boolean); console.log(data.year); console.log(data.hitler_id);
Body with parameters:
{ "size": 0, "query": { "bool": { "filter": [ { "terms": { "name": [ "{{name}}" ] } }, { "terms": { "salesman_id": [ "{{sid}}" ] } }, { "terms": { "fyyear": [ "{{fyyear}}" ] } }, { "terms": { "boolean": [ "{{boolean}}" ] } } ] } }, "aggs": { "year": { "terms": { "field": "year" }, "aggs": { "value": { "sum": { "field": "value" } } } } } }
For only string variables are accepted - name and boolean fields are working and the value is populated
for the other two, the variable values are not passed.
The variables are not used in your request body that way.
Either you have to store them in environment oder global variables via
pm.globals.set("variable_key", variable_value)
pm.environment.set("variable_key", "variable_value");
or just skip the pre-request script if you just want to use your data and reference the fields directly in your body:
{
"size": 0,
"query": {
"bool": {
"filter": [
{
"terms": {
"name": [
"{{name}}"
]
}
},
{
"terms": {
"salesman_id": [
"{{hitler_id}}"
]
}
},
{
"terms": {
"fyyear": [
{{year}}
]
}
},
{
"terms": {
"boolean": [
{{boolean}}
]
}
}
]
}
},
"aggs": {
"year": {
"terms": {
"field": "year"
},
"aggs": {
"value": {
"sum": {
"field": "value"
}
}
}
}
}
}
However take care you're storing the values in your data file. You stored the bool and the year as strings". But they should be represented as you already did for the "expected" var.
I have this index:
{
"mappings": {
"records" : {
"properties" : {
"suggest" : {
"type" : "completion",
"contexts": [
{
"name": "year",
"type": "category",
"path": "year"
}
]
}
}
}
}
}
I put some records:
POST http://localhost:9200/example/records
{
"suggest": {
"input": "foo123" ,
"contexts": {
"year": "1999"
}
}
}
POST http://localhost:9200/example/records
{
"suggest": {
"input": "some123" ,
"contexts": {
"year": "1999"
}
}
}
POST http://localhost:9200/example/records
{
"suggest": {
"input": "thing123" ,
"contexts": {
"year": "2000"
}
}
}
Now I would do this query (sql like):
SELECT * FROM example WHERE SUGGEST LIKE %123% AND YEAR=1999
How can I do in Elastic Search?
I type:
POST http://localhost:9200/example/records/_search?pretty
{
"query": {
"bool": {
"must": [
{ "wildcard" : { "suggest" : "*123*" } }
],
"filter":[
{ "term" : { "year" : "1999" } }
]
}
}
}
I have returned this response with blank results:
{
"took": 1,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"skipped": 0,
"failed": 0
},
"hits": {
"total": 0,
"max_score": null,
"hits": []
}
}
I am expecting to have returned this records:
foo123, year 1999
some123, year 1999
How can I do?
You need to use bool query with must if you care about score:
{
"query": {
"bool": {
"must": [
{ "wildcard" : { "name" : "*foo*" } },
{ "term" : { "year" : "1999" } }
]
}
}
}
or with filter if you just want to filter values and possibly cache the filter:
{
"query": {
"filter": {
"must": [
{ "wildcard" : { "name" : "*foo*" } },
{ "term" : { "year" : "1999" } }
]
}
}
}
Someone have any idea of what tokenizer to use and how to enable rule for the below,
Input : ["test1-data.example.com", "test2-new.example.com", "new1-test.example.com"]
Output (expected ) :
test1-data.example.com test2-new.example.com new1-test.exampl.com
It's not obvious whether it solves your problem or not, but here's one way you can do what it sounds like you're asking:
DELETE /test_index
PUT /test_index
{
"settings": {
"number_of_shards": 1
},
"mappings": {
"doc": {
"_all": {
"enabled": true,
"store": true,
"index": "not_analyzed"
},
"properties": {
"text_field": {
"type": "string",
"include_in_all": true
}
}
}
}
}
PUT /test_index/doc/1
{
"text_field": ["test1-data.example.com", "test2-new.example.com", "new1-test.example.com"]
}
POST /test_index/_search
{
"fields": [
"_all"
]
}
...
{
"took": 3,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"failed": 0
},
"hits": {
"total": 1,
"max_score": 1,
"hits": [
{
"_index": "test_index",
"_type": "doc",
"_id": "1",
"_score": 1,
"fields": {
"_all": "test1-data.example.com test2-new.example.com new1-test.example.com "
}
}
]
}
}
Here's the code in Sense:
http://sense.qbox.io/gist/45200711a41268634439b669e18541e68042ac8a
Type mapping
{
"pois-en": {
"mappings": {
"poi": {
"properties": {
"address": {
"type": "string",
"analyzer": "portuguese"
},
"city": {
"type": "integer"
},
(...)
"type": {
"type": "integer"
}
}
}
}
}
}
Query all:
GET pois-en/_search
{
"query":{
"match_all":{}
},
"fields": ["city"]
}
returns:
"hits": [
{
"_index": "pois-en",
"_type": "pois_poi",
"_id": "491",
"_score": 1,
"fields": {
"city": [
91
]
}
},
(...)
But when i filter using:
GET pois-en/_search
{
"query" : {
"filtered" : {
"query" : {
"match_all" : {}
},
"filter" : {
"term" : {
"city" : 91
}
}
}
}
}
Its returns nothing!
I can't figure out what i'm doing wrong.
To Django and Elasticsearch communication i'm Elasticutils (https://github.com/mozilla/elasticutils) but i'm using Sense now to make those queries.
Thanks in advance
The type name isn't consistent in your post (poi and pois_poi) - the returned document doesn't match your mapping.