Map SQL Integer Types to Sequelize DataTypes

SQL Integer Types and Sequelize DataTypes

SQL databases support multiple integer sizes for storage optimization. Sequelize mirrors these with dedicated DataTypes constants, and the converter maps each one precisely.

Standard Mapping Table

SQL Type	Sequelize DataType	Storage	Range
TINYINT	DataTypes.TINYINT	1 byte	-128 to 127
SMALLINT	DataTypes.SMALLINT	2 bytes	-32,768 to 32,767
MEDIUMINT	DataTypes.MEDIUMINT	3 bytes	-8M to 8M
INT / INTEGER	DataTypes.INTEGER	4 bytes	-2B to 2B
BIGINT	DataTypes.BIGINT	8 bytes	Very large

Example Conversion

CREATE TABLE metrics (
  id BIGINT PRIMARY KEY,
  status_code SMALLINT NOT NULL,
  retry_count TINYINT DEFAULT 0,
  request_count INTEGER NOT NULL,
  byte_size BIGINT
);

Metric.init({
  id: { type: DataTypes.BIGINT, primaryKey: true, allowNull: false },
  status_code: { type: DataTypes.SMALLINT, allowNull: false },
  retry_count: { type: DataTypes.TINYINT, allowNull: true, defaultValue: 0 },
  request_count: { type: DataTypes.INTEGER, allowNull: false },
  byte_size: { type: DataTypes.BIGINT, allowNull: true },
}, { ... });

TINYINT(1) Boolean Pattern

MySQL uses TINYINT(1) as a de facto boolean type. The converter detects this specific pattern and maps it to DataTypes.BOOLEAN instead of DataTypes.TINYINT:

is_verified TINYINT(1) DEFAULT 0
-- becomes -->
is_verified: { type: DataTypes.BOOLEAN, defaultValue: false }

PostgreSQL Aliases

PostgreSQL provides INT2 (SMALLINT), INT4 (INTEGER), and INT8 (BIGINT) aliases. The converter recognizes all of these and maps them to the appropriate Sequelize DataType. SERIAL and BIGSERIAL are treated as auto-incrementing INTEGER and BIGINT respectively.