Supported Types
Data types​
The data types are derived from the types defined in the JSON schema specification.
The type and format properties in schemas are used to determine the
data type of the field. The type property indicates the type of the field.
The format property provides additional information about the underlying type.
Fields will always have a type property, but some may also have a format
property. The JSON schema
specification already defines
a set of common formats. We support these formats and define others as well.
For optimal performance and efficient data layout, we also impose some restrictions on what data types can be used for primary key fields.
The full list of supported type and format are listed below.
Note that the client libraries that we provide use language-idiomatic types for the types and formats below. For example, while a 64-bit integer is represented as type integer in JSON requests and responses, but our Java client library uses the Java long type.
| Type | Format | Description | Supported for Key Fields | Supported for autoGenerate |
|---|---|---|---|---|
| integer | A 64-bit integral number. Optionally, format can be specified as int64. | Yes | Yes | |
| integer | int32 | A 32-bit integral number. | Yes | Yes |
| number | A double-precision 64-bit IEEE 754 floating point. Optionally, format can be specified as double. | No | No | |
| number | float | A single-precision 32-bit IEEE 754 floating point. | No | No |
| string | An arbitrary string. It may contain Unicode characters. | Yes | Yes | |
| string | byte | Binary data in an undifferentiated byte stream. | Yes | Yes |
| string | uuid | Universally unique identifiers (UUIDs). UUIDs are 16-byte numbers used to uniquely identify records. | Yes | Yes |
| string | date-time | An RFC3339 timestamp in UTC time. This in the format of yyyy-MM-ddTHH:mm:ss.SSSZ. The milliseconds portion (".SSS") is optional. | Yes | Yes |
| boolean | A boolean value, either "true" or "false". | No | No | |
| array | An array of values. The items property indicates the schema for the array values. | No | No | |
| object | A container type that stores other fields. The properties key defines the schema for the object, but is optional. To store semi-structured data where schema may not be known you can omit properties. | No | No |
int64​
Representing 64-bit integral numbers.
{
"age": {
"type": "integer"
}
}
int32​
Representing 32-bit integral numbers.
{
"age": {
"type": "integer",
"format": "int32"
}
}
double​
Representing double-precision 64-bit IEEE 754 floating point values.
{
"balance": {
"type": "number"
}
}
float​
Representing single-precision 32-bit IEEE 754 floating point values.
{
"balance": {
"type": "number",
"format": "float"
}
}
byte​
Representing binary data in an undifferentiated byte stream.
{
"data": {
"type": "string",
"format": "byte"
}
}
string​
Representing strings of text that may contain Unicode characters.
{
"name": {
"type": "string"
}
}
uuid​
Representing universally unique identifiers (UUIDs). UUIDs are 16-byte numbers used to uniquely identify records.
{
"cart_id": {
"type": "string",
"format": "uuid"
}
}
date-time​
Representing an RFC3339 timestamp in UTC time.
{
"order_date": {
"type": "string",
"format": "date-time"
}
}
boolean​
Representing a boolean value, either "true" or "false".
{
"is_active": {
"type": "boolean"
}
}
array​
Representing an array of values. The items property indicates the data type
for the array values.
{
"languages": {
"type": "array",
"items": {
"type": "string"
}
}
}
object​
Representing a container type that stores other fields. The properties key
defines the schema for the object.
{
"address": {
"type": "object",
"properties": {
"street": {
"type": "string"
},
"city": {
"type": "string"
},
"state": {
"type": "string"
},
"zip": {
"type": "integer"
}
}
}
}
Tigris also supports storing semi-structured data for objects where the schema is not known. This is done by omitting
the properties key which is optional for the object data type.
{
"event_data": {
"type": "object"
}
}
Nested data​
This section demonstrates schema with nested data.
Array​
Data of type array requires the items keyword to be expressed as JSON key. This defines the type of items that will be in the array.
The example below extends the schema for the user collection by adding the
field languages
{
"languages": {
"type": "array",
"items": {
"type": "string"
}
}
}
Object​
Data of type object requires the properties keyword to be expressed as JSON key. This is similar to the top-level properties key and specifies the fields that make up the object.
The example below extends the schema for the user collection by adding the field
address
{
"address": {
"type": "object",
"properties": {
"street": {
"type": "string"
},
"city": {
"type": "string"
},
"state": {
"type": "string"
},
"zip": {
"type": "integer"
}
}
}
}
Below is what the complete schema looks like with the addition of the two new fields languages and address
{
"title": "users",
"properties": {
"id": {
"type": "integer"
},
"name": {
"type": "string",
"maxLength": 100
},
"balance": {
"type": "number"
},
"languages": {
"type": "array",
"items": {
"type": "string"
}
},
"address": {
"type": "object",
"properties": {
"street": {
"type": "string"
},
"city": {
"type": "string"
},
"state": {
"type": "string"
},
"zip": {
"type": "integer"
}
}
}
},
"primary_key": ["id"]
}
Null Values​
null is an allowed value that can be used with any of the fields in the
schema except for the ones that are part of the primary key definition. If a
field is set to the null value, then the document stored in the database
will have that field set to the null value. If you are looking for the
behavior where you would like fields without any values to not be stored as
part of the document, then simply skip setting them to any value.