Saving to PostgreSQL
TypeormDatabase
context store provides a wrapper over the TypeORM EntityManager
optimized for batch saving. It currently supports only Postgres-compatible databases and seamlessly integrates with entity classes generated from the schema file.
Check out this section of the reference page to learn how to specify the database connection parameters.
Usage
import { Store, TypeormDatabase } from '@subsquid/typeorm-store'
import { FooEntity } from './model'
const dbOptions = {/* ... constructor options ... */}
processor.run(new TypeormDatabase(dbOptions), async ctx => {
// ...
await ctx.store.upsert([new FooEntity({ id: '1'}), new FooEntity({ id: '2'})])
})
Here,
FooEntity
represents a TypeORM entity class. In squids, these are typically generated from schema files with squid-typeorm-codegen.TypeormDatabase
constructor options govern the behavior of the class at the highest level. They are described in this section of the reference.ctx.store
is of typeStore
. See theStore
interface section of the reference page for more details.
Database migrations
The Squid SDK manages the database schema with TypeORM-based database migrations by means of the squid-typeorm-migration(1)
tool.
The tool auto-generates the schema migrations from the TypeORM entities created by codegen, so that custom migration scripts are rarely needed.
Here are some useful commands:
npx squid-typeorm-migration apply # Apply the DB migrations
npx squid-typeorm-migration generate # Generate a DB migration matching the TypeORM entities
rm -r db/migrations # Clean the migrations folder
Inspect the full list of available options with
npx squid-typeorm-migration --help
To generate or update the migrations after a schema change, follow the steps below.
Updating after schema changes
In most cases the simplest way to update the schema is to drop the database and regenerate the migrations from scratch.
1. Update schema.graphql
2. Regenerate the TypeORM entity classes
# generate entites code from the schema file
npx squid-typeorm-codegen
3. Recreate the database
# drop the database
docker compose down
# start a blank database
docker compose up -d
Note that without dropping the database the next step will generate a migration only for the schema difference.
4. Build the squid code
npm run build
5. Recreate the database migration
rm -r db/migrations
npx squid-typeorm-migration generate
6. Apply the database migration
npx squid-typeorm-migration apply
Updating a deployed squid schema
In some rare cases it is possible to update the schema without dropping the database and restarting the squid from a blank state. The most important case is adding an index to an entity field. More complex changes are usually not feasible.
Updating a running squid requires that you add an incremental migration.
1. Ensure that your local database is running and has the same schema as the database of your Cloud squid
In most situations re-creating the database container and applying existing migrations should be enough:
docker compose down
docker compose up -d
npx squid-typeorm-migration apply
2. Update schema.graphql
For example, add an index
3. Regenerate the model classes and build the code
npx squid-typeorm-codegen
npm run build
4. Add a new database migration
npx squid-typeorm-migration generate
This will create a new file in db/migrations
. You may want to examine it before proceeding to the next step.
5. Update the squid in Cloud
If the squid is deployed to SQD Cloud, update the deployed version.
If you're self-hosting it, update your remote codebase and run
npx squid-typeorm-migration apply
SQD Cloud deployment
By default, the TypeORM migrations are automatically applied by Cloud with the command npx squid-typeorm-migration apply
before the squid services are started. For custom behavior, one can override the migration script using the optional migrate:
section of squid.yaml.
To force Cloud to reset the database and start with a blank state after a schema change, use the --hard-reset
flag of sqd deploy.