I am learning Laravel, but it causes me conflict to understand some things about the subject.
As I understood the migrations are a kind of control of versions of the database with OOP
Seeds are used to create test data.
And the factories are supersets for the test data.
However, I do not understand the concepts very clearly, and I do not know very well how to apply them in Laravel, or an exercise.
Let's start the explanation
Migrations are the intermediate layer between your app and the database manager; I explain, when you traditionally develop a software you start to create your relational model in tools such as MySQL's Work Bench or from the console you create each of the tables that will compose your database, however Laravel as a framework provides the ability to write each table in our database in the form of migration; so that in this way you can generate a version of it.
The structure of a migration is very similar to this one that I show you however it depends on what you are trying to solve
public function up()
{
Schema::create('flights', function (Blueprint $table) {
$table->increments('id');
$table->string('name');
$table->string('airline');
$table->timestamps();
});
}
Each migration represents an entity / table in your database so both the name that you place them must be representative of each one of them and not placing more than one at a time
Seeds are a Laravel method that will allow you to populate your database with test information; therefore as notes to be able to use this feature of Laravel you must have your database created before (Clarification so that this quality works does not depend on you having made migrations), that is, you can perfectly create your database of all the life and later use this seed tool to also fill your data tables
To be able to start with the seeders, you must create them from the console as in the following example
php artisan make:seeder UsersTableSeeder
As you can see, the seeder is named with the name that the table will have but in the singular.
What then do the seeders have?
Sample data sample your database for testing purposes, it remains Of course it should not be the production database but one of test; will simplify this process because instead of being filling in manually each table you can quietly generate multiple and make tests
Likewise I mention the Seeders work with the query builder Fluent therefore you can see in the example that instead of inclosing the model, they work directly invoking the name of the table
Example
public function run()
{
DB::table('users')->insert([
'name' => str_random(10),
'email' => str_random(10).'@gmail.com',
'password' => bcrypt('secret'),
]);
}
They are a tool to massively populate our database and even help us write tests for the information entered, likewise the Model Factories require to invoke the model that is linked to any of your tables in your database. data
Example
$factory->define(App\User::class, function ($faker) {
return [
'name' => $faker->name,
'email' => $faker->email,
'password' => str_random(10),
'remember_token' => str_random(10),
];
});