@sebspark/spanner-migrate

spanner-migrate is a CLI tool for managing schema migrations for Google Cloud Spanner. It simplifies schema evolution by allowing you to create, apply, rollback, and track migrations.

Installation

Install @sebspark/spanner-migrate as a global package:

yarn add -D @sebspark/spanner-migrate

CLI Commands

spanner-migrate provides several commands for managing database migrations in Google Spanner.

Initialize Configuration

spanner-migrate init

Initializes a .spanner-migrate.config.json file by prompting for:

• Spanner instance name
• One or more database configurations
• Optional Google Cloud project name

Create a Migration

spanner-migrate create <description ...> [--database <name>]
spanner-migrate create add users table
spanner-migrate create --database=mydb add users table

Creates a new migration file with the specified description.

• If --database (-d) is provided, it uses the specified database.
• If multiple databases exist and none is specified, the user is prompted to select one.
• The filename is generated from the description (<timestamp>_add_users_table.sql).

Apply Migrations

spanner-migrate up
spanner-migrate up --database <name>
spanner-migrate up --database <name> --max <n>

Applies pending migrations.

• If no --database and --max are provided, applies all migrations to all databases.
• If --database (-d) is provided, applies migrations only to that database.
• If --max (-m) is provided, limits the number of migrations applied (requires --database).
• --max must be an integer greater than 0.

Roll Back Last Migration

spanner-migrate down
spanner-migrate down --database <name>

Rolls back the last applied migration.

• If a single database exists, it is automatically selected.
• If multiple databases exist, --database is required.
• The specified --database must exist.

Show Migration Status

spanner-migrate status
spanner-migrate status --database <name>

Displays migration status.

• If --database is specified, shows status for that database.
• If no --database is provided, shows status for all configured databases.

Help

spanner-migrate --help
spanner-migrate <command> --help

Displays help for the CLI or a specific command.

Programmatic Usage

In addition to the CLI, spanner-migrate can be used as a Node.js module to manage migrations programmatically.

Importing

import { init, create, up, down, status } from '@sebspark/spanner-migrate'

Initializing Configuration

import { init, type Config } from '@sebspark/spanner-migrate'

const config: Config = {
  instance: {
    name: 'my-instance',
    databases: [
      { name: 'mydb', migrationsPath: './migrations' },
    ],
  },
  projectId: 'my-gcp-project',
}

await init(config, '.spanner-migrate.config.json')

Writes the given configuration to a .spanner-migrate.config.json file.

Creating a Migration

import { create, type DatabaseConfig } from '@sebspark/spanner-migrate'

const databaseConfig: DatabaseConfig = {
  name: 'mydb',
  migrationsPath: './migrations',
}

await create(databaseConfig, 'add users table')

Creates a new migration file for the specified database.

Applying Migrations

import { up, type Config, type DatabaseConfig } from '@sebspark/spanner-migrate'

// Load configuration
const config: Config = /* Load from file or define inline */

// Apply all migrations to all databases
await up(config)

// Apply all migrations to a specific database
const databaseConfig: DatabaseConfig = config.instance.databases[0]
await up(config, databaseConfig)

// Apply up to 5 migrations to a specific database
await up(config, databaseConfig, 5)

• Applies pending migrations.
• If a database is specified, only applies migrations to that database.
• If max is specified, applies at most max migrations.

Rolling Back Migrations

import { up, type Config, type DatabaseConfig } from '@sebspark/spanner-migrate'

const config: Config = /* Load from file */
const databaseConfig: DatabaseConfig = config.instance.databases[0]

// Roll back the last applied migration
await down(config, databaseConfig)

• Rolls back the last applied migration for the specified database.
• Requires a database to be specified.

Checking Migration Status

import { up, type Config, type DatabaseConfig } from '@sebspark/spanner-migrate'

const config: Config = /* Load from file */

// Check status for all databases
const migrationStatus = await status(config)
console.log(migrationStatus)

// Check status for a specific database
const databaseConfig = config.instance.databases[0]
const migrationStatusSingle = await status(config, [databaseConfig])
console.log(migrationStatusSingle)

• Displays applied and pending migrations for one or more databases.
• If a specific database is provided, only its status is shown.

Running on Spanner Emulator

If you want to test your migrations against a Spanner Emulator, you will need to set:

process.env.SPANNER_EMULATOR_HOST = 'localhost:<port>'

License

Apache-2.0