Field Data Types
- Capella Operational
- reference
You can assign a data type to a field to tell the Search Service how to analyze its data.
When you create a mapping for a field in Quick Mode or Advanced Mode, you need to set the field’s data type.
If you create a Search index and do not set a data type for a field, the Search Service automatically assigns a field data type. For example, if you created a dynamic type mapping, the Search Service automatically assigns data types to all fields in the type mapping.
The following field data types are available:
Field Data Type | Description |
---|---|
text |
The field contains a string. The string can contain numbers and special characters. |
number |
The field contains a number. It does not contain any alphabetic characters. |
datetime |
The field contains a date/time value that matches the format of a Date/Time Parser in the index. |
boolean |
The field contains a true or false value. |
disabled |
This field data type is deprecated. It’s included for compatibility only. |
The field contains geopoint (latitude and longitude) data, represented as either:
|
|
The field contains a GeoJSON object. A GeoJSON object describes a shape made of floating point coordinates with the following JSON properties:
For example: { "type": "LineString", "coordinates": [ [-2.753735609842721, 53.94860827535115], [-2.599898256093695,53.65007434185782] ] } This JSON object describes a For more information about GeoJSON queries, see the Query object. |
|
ip |
The field contains an IP address, formatted in IPv4 or IPv6 CIDR syntax. For example: { "ipv4": "4.7.44.162", "ipv6": "2001:4800:0000:0000:0000:0000:0000:0000" } |
The field contains an array of floating point numbers or an array of arrays that represent a vector embedding. Use the For more information about Vector Search, see Use Vector Search for AI Applications. |
|
(Server version 7.6.2 and later) vector_base64 |
The field contains an array of floating point numbers formatted as a base64 encoded string that represent a vector embedding. Use the For more information about Vector Search, see Use Vector Search for AI Applications. |