Dev Principles

Web Development

Dev Principle #3: Version Your Database

July 27, 2016

Jake Litwicki

By Jake Litwicki

There are numerous benefits of having a standardized local development environment for your entire team to work on and share. This arrangement is essential for reducing bugs and interchanging team members amongst projects – a topic discussed at length in Vagrant Up Can Make Development Easier.

One of the major milestones to achieve in your team’s workflow is assuring everyone is working on the same version of your back-end. This is most often a database of some kind. Whether NoSQL, relational (MySQL, SQL Server), or serverless, we recommend standardizing and versioning your data to ensure the greatest efficiency. Versioning your database requires treating it like code.

Typical best practices require that each developer has his/her own copy of the database to work. This avoids imposing bugs or changes on other developers which causes unnecessary work, and removes unnecessary roadblocks and conflicts. It’s generally the best overall approach to an Agile workflow.

Migrations

Nowadays most development frameworks and platforms have some form of migrations. WordPress is a notable exception, although some hybrid solutions do exist via plugins like WP DB Migrate.

Wikipedia defines a migration as “the management of incremental, reversible changes to relational database schemas.”

In practice this means that regardless of which branch of code the developer is working on, a single command can be executed to create, build, and/or update the database so that work can begin immediately. This almost completely alleviates the need for copying/importing databases, fixing conflicts, or dealing with connecting to remote databases and the inevitable connectivity issues that arise.

For example, using Laravel the process would be one or two simple commands:

Copy

# Create the articles table
php artisan make:migration create_articles_table --create=articles
Copy

# Add to the existing articles table
php artisan make:migration add_votes_to_articles_table --table=articles

At this point, every branch of code you go to work on will have a fresh database. Any schema changes necessary to complete your work could be built into a fresh migration script. This can be executed on top of your existing schema.

When merged into your primary release branch, any future development will have your migration(s). As a result, the workflow continues uninterrupted.

Example Migration

Below is an example migration to set up a simple table using Eloquent (Laravel):

Copy

<?php

use Illuminate\Database\Schema\Blueprint;
use Illuminate\Database\Migrations\Migration;

class CreateArticlesTable extends Migration
{
    /**
     * Run the migrations.
     *
     * @return void
     */
    public function up()
    {
        Schema::create('articles', function (Blueprint $table) {
            $table->increments('id');
            $table->string('title');
            $table->string('body');
            $table->timestamps();
        });
    }

    /**
     * Reverse the migrations.
     *
     * @return void
     */
    public function down()
    {
        Schema::drop('articles');
    }
}

Seeding

Seeding is the process of populating data into a database. This can be “dummy” data or enum data (or both), and is very useful for testing and tuning your models; particularly, in object oriented frameworks.

A simple example that generates a single article:

Copy

<?php

use Illuminate\Database\Seeder;
use Illuminate\Database\Eloquent\Model;

class DatabaseSeeder extends Seeder
{
    /**
     * Run the database seeds.
     *
     * @return void
     */
    public function run()
    {
        DB::table('articles')->insert([
            'title' => str_random(10),
            'body' => str_random(1000)
        ]);
    }
}

For further reading on other Development Principles and Best Practices, check out Dev Principle #1: Choose Appropriate Variable Names and Dev Principle #2: Coding Layout and Style Matter.

Jake Litwicki

Jake Litwicki

Software Development Director

Jake is a software engineer and leader with over ten years of experience working with and/or managing teams of designers, developers, and artists. He loves working with projects involving Symfony, Drupal, Wordpress, Angular, and Laravel but most of all loves the challenge of a new technology. In his spare time Jake is an aspiring fisherman and woodworker while not spending time with his wife and two children.

Unless otherwise specified, source code in this post is licensed under a
Creative Commons Attribution 4.0 International license (CC BY 4.0).

You might also like...

20

Jun.

Ben Spencer

Unit/Automated Testing

Software testing is the process of discerning whether a product is acceptable to its end users, its target audience, its purchasers, and the stakeholders to whom it belongs. At its core, software testing examines code and its execution: does the code do what it is supposed to do, when it is supposed to do it, … Continued

12

May.

Travis Luong

Should I Use RxJS in My Project?

If you’re thinking about using RxJS, you might as well be asking, “Should I use a jump kick in a street fight?” Unless you are Bruce Lee, probably not. As a disclaimer, I am not an RxJS pro. I only started learning about it a couple of weeks ago. Consider this post a record of … Continued

9

Mar.

Steve Hulet

The Definition of Done

In development, we often split tasks up into smaller pieces and distribute them among multiple developers. We need to eventually bring this work back together, and this is easier when everyone agrees on what “done” means. The definition of “done” is important in Scrum for calculating velocity – “a measure of the amount of work … Continued