Add a column with a migration

Task: products need a stock count, and the database already exists.

# Recipe: change the schema with a new migration, never by editing an old one.
paths:
  components: ./components
  migrations: ./migrations

datasources:
  db:
    driver: sqlite
    database: ./data/shop.db

The first migration made the table:

CREATE TABLE products (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    name TEXT NOT NULL
);

INSERT INTO products (name) VALUES ('Mug'), ('Sticker');

The change is a second file. Migrations run in order, each once, each in its own transaction; an applied migration is never edited:

-- Rows that already exist get the default.
ALTER TABLE products ADD COLUMN stock INTEGER NOT NULL DEFAULT 0;

UPDATE products SET stock = 12 WHERE name = 'Mug';

quantum migrate up

<q:component name="Products">
  <q:query name="products" datasource="db">
    SELECT name, stock FROM products ORDER BY name
  </q:query>

  <ui:window title="Products">
    <ui:table source="{products}">
      <ui:column key="name" label="Name" />
      <ui:column key="stock" label="In stock" />
    </ui:table>
  </ui:window>
</q:component>

Each test starts from a database built by the migrations, so the test sees the schema a new install gets:

<q:test name="both migrations ran, in order" page="/">
  <test:visit />
  <test:expect text="In stock" />
  <test:expect table="products" count="1" where="name = 'Mug' AND stock = 12" />
  <test:expect table="products" count="1" where="name = 'Sticker' AND stock = 0" />
</q:test>

tests/stock.test.q
  PASS  both migrations ran, in order
1 passed, 0 failed

Rather write the schema you want and let Quantum write the migration? See quantum migrate plan in Project Structure. See DB-6 and DB-8.