Repository files navigation

An awesome SurrealDB migration tool, with a user-friendly CLI and a versatile Rust library that enables seamless integration into any project.

WarningThis project is not production-ready, use at your own risk.

This project can be used:

• as a Rust library

cargo add surrealdb-migrations

• or as a CLI

cargo install surrealdb-migrations

• or inside a CI workflow, using aGitHub Action

Checksurrealdb-migrations GitHub Action in the marketplace

• or even as aninitContainer via aDocker image

docker pull dbottiau/surrealdb-migrations:latest

The SurrealDB Migrations project aims to simplify the creation of a SurrealDB database schema and the evolution of the database through migrations. A typical SurrealDB migration project is divided into 3 categories: schema, event and migration.

A schema file represents no more than one SurrealDB table. The list of schemas can be seen as the Query model (in a CQRS pattern). Theschemas folder can be seen as a view of the current data model.

An event file represents no more than one SurrealDB event and the underlying table. The list of events can be seen as the Command model (in a CQRS pattern). Theevents folder can be seen as a view of the different ways to update the data model.

A migration file represents a change in SurrealDB data. It can be a change in the point of time between two schema changes. Examples are: when a column is renamed or dropped, when a table is renamed or dropped, when a new data is required (with default value), etc...

stateDiagram-v2    scaffold : Scaffold a project    changeSchema : Change schema/event    createMigration: Create migration (data changes)    apply : Apply to your database    state fork_state <<fork>>        [*] --> scaffold        scaffold --> fork_state        fork_state --> changeSchema        fork_state --> createMigration    state join_state <<join>>        changeSchema --> join_state        createMigration --> join_state        join_state --> apply        apply --> fork_state

You can start a migration project by scaffolding a new project using the following command line:

surrealdb-migrations scaffold template empty

This will create the necessary folders and files in order to perform migrations. Theempty template should look like this:

• /schemas
  • script_migration.surql
• /events
• /migrations

There are a number of pre-defined templates so you can play around and get started quickly.

Once you have created your migration project, you can start writing your own model. Based on the folders you saw earlier, you can create schema files, event files and migration files.

You can create strict schema files that represent tables stored in SurrealDB.

surrealdb-migrations create schema post --fields title,content,author,created_at,status

This will create a schemaless table with predefined fields:

DEFINE TABLE post SCHEMALESS;DEFINE FIELD title ON post;DEFINE FIELD content ON post;DEFINE FIELD author ON post;DEFINE FIELD created_at ON post;DEFINE FIELD status ON post;

You can also create events in the same way.

surrealdb-migrations create event publish_post --fields post_id,created_at

This will define a table event with predefined fields:

DEFINE TABLE publish_post SCHEMALESS;DEFINE FIELD post_id ON publish_post;DEFINE FIELD created_at ON publish_post;DEFINE EVENT publish_post ON TABLE publish_post WHEN $event == "CREATE" THEN (    # TODO);

And when updating data, you can create migration files this way:

surrealdb-migrations create AddAdminUser

This will create a new file using the current date & time of the day, like20230317_153201_AddAdminUser.surql for example. All migrations files should be listed in a temporal order.

Finally, when you are ready, you can apply your schema and migrations to the database using the following command line:

surrealdb-migrations apply

Or directly inside your Rust project using the following code:

use surrealdb_migrations::MigrationRunner;use surrealdb::engine::any::connect;use surrealdb::opt::auth::Root;#[tokio::main]asyncfnmain() ->Result<()>{let db =connect("ws://localhost:8000").await?;// Signin as a namespace, database, or root user    db.signin(Root{username:"root",password:"root",}).await?;// Select a specific namespace / database    db.use_ns("namespace").use_db("database").await?;// Apply all migrationsMigrationRunner::new(&db).up().await.expect("Failed to apply migrations");Ok(())}

Repeat the process from step 2. Change schema and/or create data change migrations.

To help you get started quickly, there is a list of predefined templates you can use:

You can scaffold a project using any of these templates using the following command line:

surrealdb-migrations scaffold template <TEMPLATE>

You can create a.surrealdb configuration file at the root of your project. This way you won't have to set the same configuration values every time.

[core]path ="./tests-files"schema ="less"[db]address ="ws://localhost:8000"username ="root"password ="root"ns ="test"db ="test"

In thecore section, you can define the path to your schema/migration files, if it is not the current folder.

In thedb section, you can define the values used to access your SurrealDB database. It can be theurl,username,password, the namespacens or the name of the databasedb.

Here is the definition of the.surrealdb configuration file:

[core]# Optional# Type: String# Description: Path to the folder that contains your migration project (root folder by default)# Default: "."path# Optional# Type: "less" | "full"# Description: Define SCHEMALESS or SCHEMAFULL option by default when creating new table/event file# Default: "less"schema[db]# Optional# Type: String# Description: Address of the surrealdb instance# Default: "ws://localhost:8000"address# Optional# Type: String# Description: Username used to authenticate to the surrealdb instance# Default: "root"username# Optional# Type: String# Description: Password used to authenticate to the surrealdb instance# Default: "root"password# Optional# Type: String# Description: Namespace to use inside the surrealdb instance# Default: "test"ns# Optional# Type: String# Description: Name of the database to use inside the surrealdb instance# Default: "test"db[filters]# Optional# Type: Array<String># Description: Tags used to filter schema/migration files# Default: [ "*" ]tags# Optional# Type: Array<String># Description: Tags used to exclude schema/migration files# Default: [ "old" ]exclude_tags

Here is the list of all environment variables that you can use:

• SURREAL_MIG_PATH - Path to the folder that contains your migration project
• SURREAL_MIG_SCHEMA - Define SCHEMALESS or SCHEMAFULL option by default when creating new table/event file
• SURREAL_MIG_ADDRESS - Address of the surrealdb instance
• SURREAL_MIG_USER - Username used to authenticate to the surrealdb instance
• SURREAL_MIG_PASS - Password used to authenticate to the surrealdb instance
• SURREAL_MIG_NS - Namespace to use inside the surrealdb instance
• SURREAL_MIG_DB - Name of the database to use inside the surrealdb instance
• SURREAL_MIG_TAGS - Tags used to filter schema/migration files
• SURREAL_MIG_EXCLUDE_TAGS - Tags used to exclude schema/migration files

The execution context describes how configuration settings are applied and prioritized when a command is executed.

The order of execution goes like this:

Environment variables -> Configuration file -> CLI arguments

Environment variables: Default settings can be via environment variables. These are typically set up in the system or shell environment.

Configuration file: If additional customization is needed, a config file can override environment variables, offering fine-tuned settings. See the section above.

CLI arguments: These take precedence over both environment variables and the configuration file. CLI arguments are explicitly provided at runtime, ensuring immediate and specific customization for the specified command.

By default, migrations are forward-only. However, it can be interesting to revert a migration in order to undo a mistake. You will find backward migrations in two places:

1. Inside the/migrations/down folder with the same name as your forward migration
2. Inside the/migrations but with the.down.surql extension next to the forward migration

So, a migration project with backward migrations might look like this:

• /schemas
  • script_migration.surql
• /events
• /migrations
  • 20231605_205201_AddProduct.surql
  • /down
    • 20231605_205201_AddProduct.surql

Or like this:

• /schemas
  • script_migration.surql
• /events
• /migrations
  • 20231605_205201_AddProduct.surql
  • 20231605_205201_AddProduct.down.surql

If you want to create a DOWN migration file when creating the migration file, use this command:

surrealdb-migrations create AddProduct --down

If you need to, you can revert all migrations back to the one you specified.

surrealdb-migrations apply --down 20231605_205201_AddProduct

And if you need to undo all your migrations, use this command:

surrealdb-migrations apply --down 0

Database branching is a similar concept to version control system like Git where you manage code repositories with branches.

With database branching, you can create a separate copy or branch of the main database to perform various tasks such as testing new features, implementing changes, or running experiments. This allows developers or teams to work independently on different branches without interfering with the stability and integrity of the original database.

You can make make schemas changes, apply new migrations and/or change data on a separate branch. These changes are isolated from the main database until they are merged back, allowing for better control and organization of database changes.

In a development workflow, you have a primary/main database that contains the latest features on your project. You often work on multiple features or you want to try the work of your colleagues but it messes up your development database, whether you are using migrations or not. Database branching allows you to create a fork of the main database, work on a new feature, apply schema or data changes and then merge your new changes to the main database when your feature is ready.

stateDiagram-v2    main : Main branch    createBranch : Create branch "feature-1"    makeChanges : Make changes on "feature-1"    merge : Merge "feature-1" on main    remove : Remove branch "feature-1"    [*] --> main    main --> [*]    main --> createBranch    createBranch --> makeChanges    makeChanges --> remove    makeChanges --> merge    remove --> [*]    merge --> [*]

You start by creating a new branch using the following command line:

surrealdb-migrations branch new --address http://localhost:8000

You will then receive a message like this:

You can now use the branch with the following configuration:ns: branchesdb: bright-fold-1617

You can now make your changes on the newly generated database usingns anddb properties.When you are done with your changes, you can merge your branch to the origin branch using the following command line:

surrealdb-migrations branch merge bright-fold-1617 --mode all --address http://localhost:8000

There are 3 merge modes, each with its own interest:

TBD

This feature requires 3 namespaces:

• features
• branches
• branches/origin

It is strongly recommended to avoid using one of these namespaces in your SurrealDB instance.

# create new branch from current branch with a random name, from default ns/dbsurrealdb-migrations branch new# create new branch from current branch with a name, from default ns/dbsurrealdb-migrations branch new <BRANCH_NAME># create new branch, from a specific ns/dbsurrealdb-migrations branch new --ns <NS> --db <DB># review diffs between original branch and the new branchsurrealdb-migrations branch diff <BRANCH_NAME># commit and merge branch changes to the original branchsurrealdb-migrations branch merge <BRANCH_NAME># remove branch (ie. rollback)surrealdb-migrations branch remove <BRANCH_NAME># list all existing branchessurrealdb-migrations branch list# display infos of a branchsurrealdb-migrations branch status <BRANCH_NAME>surrealdb-migrations branch <BRANCH_NAME>

This project contains sample apps that demontrates how to use thesurrealdb-migrations given certain contexts. Here is a list of existing samples:

Thanks goes to these wonderful people (emoji key):

Amar Sood 💻	Rom's 💻🤔🚇	Pranay Pratyush 🐛	Zafar Ansari 🤔	Tim 🤔	Matt Jackson 🐛	Lucas 🤔🐛
Lucas 🤔	Stephan Gerhard 🐛🤔	Yt 🐛	Dave Warnock 🐛	Johan Smits 💻🐛	Gabriel Hansson 💻	Liam Woodleigh-Hardinge 💻
Gerard Ruiz 🤔💻	Marko Lazić 🐛	Rhea Jara 🐛	Chiru B 🤔	Marco Mihai Condrache 💻	Tosic.Killer 💻	Christian Kaps 💻

This project follows theall-contributors specification. Contributions of any kind welcome!

Inspired by awesome projects:

• Entity Framework
• Fluent Migrator
• kards-social by Micha de Vries@kearfy