View schemas persist the relational structure the driver infers for Elasticsearch types and queries. To provide an example of how custom schemas work, we will use the below mapping (where 'insured' is the name of the table).

{
  "insured": {
    "properties": {
      "name": { "type":"string" },
      "address": {
        "street": { "type":"string" },
        "city": { "type":"string" },
        "state": { "type":"string" }
      },
      "insured_ages": { "type": "integer" },
      "vehicles": {
        "type": "nested",
        "properties": {
          "year": { "type":"integer" },
          "make": { "type":"string" },
          "model": { "type":"string" },
          "body_style" { "type": "string" }
        }
      }
    }
  }
}

Also, consider the following example data for the above mapping:

{
  "_source": {
    "name": "John Smith",
    "address": {
      "street": "Main Street",
      "city": "Chapel Hill",
      "state": "NC"
    },
    "insured_ages": [ 17, 43, 45 ], 
    "vehicles": [
      {
        "year": 2015,
        "make": "Dodge",
        "model": "RAM 1500",
        "body_style": "TK"
      },
      {
        "year": 2015,
        "make": "Suzuki",
        "model": "V-Strom 650 XT",
        "body_style": "MC"
      },
      {
        "year": 2012,
        "make": "Honda",
        "model": "Accord",
        "body_style": "4D"
      }
    ]
  }
}

Defining a Custom Schema

Schemas persisted when GenerateSchemaFiles is set are placed into the folder specified by the Location property. For example, set GenerateSchemaFiles to "OnUse" and execute a SELECT query:

SELECT * FROM insured

You can then change column behavior in the resulting schema. The following schema uses the other:xPath property to define where the data for a particular column should be retrieved from. Using this model you can flatten arbitrary levels of hierarchy.

The es_index and es_type attributes specify the Elasticsearch index and type to retrieve. The es_index and es_type attributes give you the flexibility to use multiple schemas for the same type. If es_type is not specified, the filename determines the collection that is parsed.

Below is an example is an example of the column behavior markup. You can find a complete schema in Custom Schema Example.

  <rsb:script xmlns:rsb="http://www.rssbus.com/ns/rsbscript/2">  
  
    <rsb:info title="StaticInsured" description="Custom Schema for the Elasticsearch insured data set.">  
      <!-- Column definitions -->
      <attr name="_id"                         xs:type="string"  other:xPath="_id"                                                         other:sourceField="_id"                 other:analyzed="true"  />
      <attr name="_score"                      xs:type="double"  other:xPath="_score"                                                      other:sourceField="_score"              other:analyzed="true"  />
      <attr name="name"                        xs:type="string"  other:xPath="_source/name"                                                other:sourceField="name"                other:analyzed="true"  />
      <attr name="address.street"              xs:type="string"  other:xPath="_source/address/street"                                      other:sourceField="address.street"      other:analyzed="true"  />
      <attr name="address.city"                xs:type="string"  other:xPath="_source/address/city"                                        other:sourceField="address.city"        other:analyzed="true"  />
      <attr name="address.state"               xs:type="string"  other:xPath="_source/address/state"                                       other:sourceField="address.state"       other:analyzed="true"  />
      <attr name="insured_ages"                xs:type="string"  other:xPath="_source/insured_ages"          other:valueFormat="aggregate" other:sourceField="insured_ages"        other:analyzed="false" />
      <attr name="insured_ages.0"              xs:type="integer" other:xPath="_source/insured_ages[0]"                                     other:sourceField="insured_ages"        other:analyzed="false" />
      <attr name="vehicles"                    xs:type="string"  other:xPath="_source/vehicles"              other:valueFormat="aggregate" other:sourceField="vehicles"            other:analyzed="true"  />
      <attr name="vehicles.0.year"             xs:type="integer" other:xPath="_source/vehicles[0]/year"                                    other:sourceField="vehicles.year"       other:analyzed="true"  />
      <attr name="vehicles.0.make"             xs:type="string"  other:xPath="_source/vehicles[0]/make"                                    other:sourceField="vehicles.make"       other:analyzed="true"  />
      <attr name="vehicles.0.model"            xs:type="string"  other:xPath="_source/vehicles[0]/model"                                   other:sourceField="vehicles.model"      other:analyzed="true"  />
      <attr name="vehicles.0.body_style"       xs:type="string"  other:xPath="_source/vehicles[0]/body_style"                              other:sourceField="vehicles.body_style" other:analyzed="true"  />
  
      <input name="rows@next" desc="Internal attribute used for paging through data."  />
    </rsb:info>  
  
  
    <rsb:set attr="es_index" value="auto"/>
    <rsb:set attr="es_type"  value="insured"/>
  
  </rsb:script>