CODEWITHSUNDEEP
elasticsearch
Aysennoussi
Aysennoussi

Reputation: 3860

How does elasticsearch work?

How does Elasticsearch handle indices? I have a field called Coordinates which has subfields lat and lng in a collection called users. I want to search inside this collection but by indexing the coordinate field. In my MongoDB configuration the coordinate field is already a 2D index. How can I try to tell ElasticSearch to search indexing that field? Does it know that coordinates is an index ? or should I tell it by this script?

curl -XPUT "localhost:9200/_river/artist/_meta" -d'
{
  "type": "mongodb",
    "mongodb": {
      "db": "mydb", 
      "collection": "users"
    },
    "index": {
      "name": "coordin", 
      "type": "coordinates"
    }
}'

Upvotes: 6

Views: 9969

Answers (2)

Lucas Gazire
Lucas Gazire

Reputation: 311

Elasticsearch works by your index configuration, assuming that you already configured everything and you are ready to receive some files you need to edit the file locate in /etc/elasticsearch/templates/ in my example below is a log4net and I added the geospatial configuration that I found here:

{
  "log4net": {
    "template": "log4net*",
    "mappings": {
      "fluentd": {
        "_ttl": {
          "enabled": true,
          "default": "62d"
        },
        "properties": {
          "hostname": {
            "type": "string",
            "index": "not_analyzed"
          },
          "level": {
            "type": "string",
            "index": "not_analyzed"
          },
          "enviroment": {
            "type": "string",
            "index": "not_analyzed"
          },
          "site": {
            "type": "string",
            "index": "not_analyzed"
          },
          "username": {
            "type": "string",
            "index": "not_analyzed"
          },
          "logger": {
            "type": "string",
            "index": "not_analyzed"
          },
          "thread": {
            "type": "string",
            "index": "not_analyzed"
          },
          "systemname": {
            "type": "string",
            "index": "not_analyzed"
          },
          "message": {
            "type": "string",
            "index": "not_analyzed"
          },
          "location": {
            "type": "geo_point",
            "fielddata": {
              "format": "compressed",
              "precision": "1cm"
            }
          }
        }
      }
    }
  }
}

Upvotes: 1

NikoNyrh
NikoNyrh

Reputation: 4128

I assume that here by "index" you mean like the ones at MongoDB and SQL server. At elasticsearch context it is a collection of types and documents, more like a database is a collection of tables and rows. By default all fields in elasticsearch are stored into a Lucene data structure from which it can be efficiently be queried.

Elasticsearch does support indexed geospatial data, documentation can be found from here.

Upvotes: 1

Related Questions