Define Field Mappings for a MongoDB Search Index

The following index definition:

• Enables dynamic mappings at the root level to automatically index fields in the collection based on the typeSet definition named only_strings, which indexes string fields in the collection with the positions index option and with store set to false. These configuration options save disk space by not storing the field values or term offsets for query highlighting.

{
  "mappings": {
    "dynamic": {
      "typeSet": "only_strings"
    }
  },
  "typeSets": [
    {
      "name": "only_strings",
      "types": [
        {
          "type": "string",
          "store": false,
          "indexOptions": "positions"
        }
      ]
    }
  ]
}

The following index definition:

• Enables dynamic mappings at the root level to dynamically index fields in the collection based on the typeSet named only_strings, which indexes string fields in the collection with the positions index option and with store set to false, which saves disk space by not storing the field values or term offsets for query highlighting.

• Indexes the awards field as the document type and uses the dynamic option with the only_numbers typeSet to dynamically index only the numeric sub-fields in the awards document. This means that the numeric sub-fields wins and nominations are indexed, but the string sub-field text is not indexed.

{
  "mappings": {
    "dynamic": {
      "typeSet": "only_strings"
    },
    "fields": {
      "awards": {
        "type": "document",
        "dynamic": {
          "typeSet": "only_numbers"
        }
      }
    }
  },
  "typeSets": [
    {
      "name": "only_numbers",
      "types": [
        {
          "type": "number"
        }
      ]
    },
    {
      "name": "only_strings",
      "types": [
        {
          "type": "string",
          "store": false,
          "indexOptions": "positions"
        }
      ]
    }
  ]
}

The following index definition:

• Configures dynamic mappings for specific field types using the typeSet named indexedTypes, which specifies the following behavior:

  • Automatically index string fields as the token type.

  • Automatically index numeric fields as the number type.

• Excludes the plot field from being indexed.

{
  "mappings": {
    "dynamic": {
      "typeSet": "indexedTypes"
    },
    "fields": {
      "plot": []
    }
  },
  "typeSets": [
    {
      "name": "indexedTypes",
      "types": [
        {
          "type": "token"
        },
        {
          "type": "number"
        }
      ]
    }
  ]
}

The following index definition:

• Specifies static field mappings at the root level (dynamic: false), which means that fields that are not specified in mappings.fields are not indexed. The index definition explicitly includes the following fields for indexing:

  • The awards field, which is of type document. It has three embedded sub-fields, wins, nominations and text.

    The wins and nominations sub-fields contain numeric data and are indexed as the type number. nominations is indexed with the int64 representation, which allows it to support larger integer values than the default number type.

    The text sub-field contains string data and is indexed as the string type using the lucene.english analyzer, which is optimized for English text. The text field also uses the ignoreAbove option to ignore any string of more than 255 characters in length.

  • The title field, which contains string data and is indexed as type string using the lucene.whitespace analyzer, which is optimized for text that contains a lot of whitespace, such as product names or titles. The title field uses the multi option to specify the lucene.french analyzer as a secondary analyzer. This allows you to specify { "value": "title", "multi": "frenchAnalyzer" } as the path in your $search queries to better target French query terms.

  • The genres field, which contains an array of strings and is indexed as the type string. MongoDB Search indexes the array elements using the lucene.standard analyzer by default. For indexing arrays, MongoDB Search only requires the data type of the array elements. You don't have to specify that the data is contained in an array in the index definition.

{
  "mappings": {
    "dynamic": false,
    "fields": {
      "awards": {
        "type": "document",
        "fields": {
          "wins": {
            "type": "number"
          },
          "nominations": {
            "type": "number",
            "representation": "int64"
          },
          "text": {
            "type": "string",
            "analyzer": "lucene.english",
            "ignoreAbove": 255
          }
        }
      },
      "title": {
        "type": "string",
        "analyzer": "lucene.whitespace",
        "multi": {
          "frenchAnalyzer": {
            "type": "string",
            "analyzer": "lucene.french"
          }
        }
      },
      "genres": {
        "type": "string"
      }
    }
  }
}

This index definition:

• Specifies static field mappings (dynamic: false), which means fields and field types that aren't explicitly mentioned aren't indexed. So, the index definition includes:

  • The title field, which contains string data and is indexed as the string type using the lucene.whitespace analyzer, which is optimized for text that contains a lot of whitespace, such as product names or titles.

  • The awards field, which is of type document and contains three sub-fields. The wins and nominations sub-fields contain numeric data, and the text sub-field contains string data. Instead of explicitly setting the data type for each nested field in the document, the index definition enables dynamic mapping for all the sub-fields in the document. This indexes wins and nominations as the number type, and text as the string type. Data in the text field is indexed using the lucene.standard analyzer by default.

  {
    "mappings": {
      "dynamic": false,
      "fields": {
        "title": {
          "type": "string",
          "analyzer": "lucene.whitespace"
        },
        "awards": {
          "type": "document",
          "dynamic": true
        }
      }
    }
  }

This index definition:

• Specifies static field mappings at the root level (dynamic: false), which means that fields that aren't specified in the mappings.fields option aren't indexed.

• Defines dynamic mappings for the awards field, which is of type document, using the custom typeSet named movieAwards. The awards field has three embedded sub-fields: wins, nominations, and text. Instead of automatically indexing the sub-fields by setting dynamic to true at the root level or by explicitly indexing each nested field using static mappings, the index definition configures dynamic mappings for the document field type using the typeSet definition named movieAwards. The movieAwards typeSet does the following:

  • Indexes string fields as the string type using the multi option to specify multiple analyzers: lucene.english and lucene.french.

  • Indexes numeric fields as type number using the default settings for the number type.

    {
      "mappings": {
        "dynamic": false,
        "fields": {
          "awards": {
            "type": "document",
            "dynamic": {
              "typeSet": "movieAwards"
            }
          }
        }
      },
      "typeSets": [
        {
          "name": "movieAwards",
          "types": [
            {
              "type": "string",
              "multi": {
                "english": {
                  "type": "string",
                  "analyzer": "lucene.english"
                },
                "french": {
                  "type": "string",
                  "analyzer": "lucene.french"
                }
              }
            },
            {
              "type": "number"
            }
          ]
        }
      ]
    }