Simple node wrapper around sqldef for easy use with node.
This is a super-simple wrapper around sqldef that will download the appropriate (for your platform/arch) CLI from sqldef releases and run it. It works with mysql & postgres.
If it's unclear how this might be usful in your workflow, just checkout this live demo.
Install it in your project as a dev-dependency, or use it standalone in npx:
npm i -D sqldef
# or
npx sqldef --help
You can use it in your own code (like for migration scripts or whatever)
import sqldef from 'sqldef'
const main = async () => {
const output = await (sqldef({
type, // the type of your database ("mysql" or "postgres")
database, // the name of your database
user, // the username of your database
password, // the password of your database
host, // the hostname of your database
port, // the port of your database
socket, // the unix socket (for mysql)
file, // the schema file to read/write
dry: true, // dry run - don't do anything to the database
get: true // get the current definition from database
}))
}
main()
Use it in your npm scripts.
Usage: sqldef <command> [options]
Commands:
sqldef export Export your database to a file
sqldef import Import your database from a file
Options:
-?, --help Show help [boolean]
-f, --file The schema file to import/export [default: "schema.sql"]
-t, --type The type of the database [default: "mysql"]
-h, --host The host of the database [default: "localhost"]
-d, --database The name of the database [default: "test"]
-u, --user The user of the database [default: "root"]
-P, --password The password for the database [default: ""]
-p, --port The port of the database
-s, --socket The socket of the database (only for mysql)
-x, --no-confirm Don't confirm before running the import/export [boolean]
-v, --version Show version number [boolean]
Examples:
sqldef export --user=cool Save your current schema, from your
--password=mysecret mysql database, in schema.sql
--database=mydatabase
sqldef import --user=cool Save the structure from your currnet
--password=mysecret database in schema.sql
--database=mydatabase
This uses cosmiconfig, which gives us a lot of options for configuration, and your settings will be merged from all the things you setup.
- a
sqldef
property inpackage.json
- a
.sqldefrc
file in JSON or YAML format - a
.sqldefrc.json
file - a
.sqldefrc.yaml
,.sqldefrc.yml
, or.sqldefrc.js
file - a
sqldef.config.js
file exporting a JS object
Here are some cool ways to use this tool in your project.
development
Make a file called schema.sql
, and make a .sqldefrc
file with your settings. Make sure .sqldefrc
is in your .gitignore
, so you don't accidentally check-in your database secrets.
.sqldefrc
type: mysql
host: localhost
database: test
user: root
password: root
file: schema.sql
Maybe also make a .sqldefrc.example
to help your users set stuff up.
Now you can run npx sqldef import
to apply the schema in schema.sql
, and npx sqldef export
to put the current structure in the database from your schema.sql
.
CI
Let's say you want to run it in CI with the schema in schema.sql
. I recommend you use environment-variables to keep all the config in your CI settings:
DB_TYPE=mysql
DB_HOST=mysqlhost.com
DB_PWD=mysecretpassword
DB_USER=root
Now, install as a dev-dependency in your project with npm i -D sqldef
, and make a script in package.json
for doing schema migration:
{
"scripts": {
"migrate": "sqldef import --file=schema.sql --type=$DB_TYPE --host=$DB_HOST --password=$DB_PWD --user=$DB_USER --no-confirm",
"schema": "sqldef export --file=schema.sql --type=$DB_TYPE --host=$DB_HOST --password=$DB_PWD --user=$DB_USER"
}
}
Now, configure your CI to execute npm run migrate
when code gets checked in. You might want to keep the above .sqldefrc
method (from above) for development. The configuration you setup there will override the DB_
environment vars, so it should work pretty well together. If you are using environment variables for other things, you could also just use your environment, and set it all in once place (like .env
using dotenv.)