Laravel : Migrations & Seeding for production data
My application needs a pre registered data set to work. So i need to insert them in the database when i set up the application.
Laravel propose two mechanisms :
- Database migrations : "They allow a team to modify the database schema and stay up to date on the current schema state."
- Database seeding : "Laravel also includes a simple way to seed your database with test data using seed classes."
When I read this description, none of these solutions seems to be adapted.
A similar question has been asked on stackoverflow and answered. The answer proposes to use the a database seeder to populate the database by detecting the current environment :
<?php
class DatabaseSeeder extends Seeder {
public function run()
{
Eloquent::unguard();
if (App::environment() === 'production')
{
$this->call('ProductionSeeder');
}
else
{
$this->call('StagingSeeder');
}
}
}
Of course, this solution works. But i am not sure that it is the right way to do this, because by inserting data using seeders you are losing all the advantages provided by the migration mechanism (database upgrate, rollback...)
I want to know what is the best practice in this case.
Laravel development is about freedom. So, if you need to seed your production database and think DatabaseSeeder is the best place to do so, why not?
Okay, seeder is mainly to be used with test data, but you'll see some folks using it as you are.
I see this important kind of seed as part of my migration, since this is something that cannot be out of my database tables and artisan migrate
is ran everytime I deploy a new version of my application, so I just do
php artisan migrate:make seed_models_table
And create my seedind stuff in it:
public function up()
{
$models = array(
array('name' => '...'),
);
DB::table('models')->insert($models);
}
I've often found myself wondering what the right answer to this is. Personally, I'd steer clear of using seeding to populate required rows in the database as you'll have to put a load of conditional logic in to ensure that you don't attempt to populate something that's already there. (Deleting and recreating the data is very inadvisable as you could end up with key mismatches and if you're using cascading deletes you may accidentally wipe a load of your database my mistake! ;-)
I put the 'seeding' of rows into the migration script as the chances are, the data will need to be there as part of the rollout process.
It's worth noting that you should use the DB class instead of Eloquent models to populate this data as your class structure could change over time which will then prevent you from re-creating the database from scratch (without rewriting history and changing you migration files, which I'm sure is a bad thing.)
I'd tend to go with something like this:
public function up()
{
DB::beginTransaction();
Schema::create(
'town',
function (Blueprint $table) {
$table->increments('id');
$table->string('name');
$table->timestamps();
}
);
DB::table('town')
->insert(
array(
array('London'),
array('Paris'),
array('New York')
)
);
Schema::create(
'location',
function (Blueprint $table) {
$table->increments('id');
$table->integer('town_id')->unsigned()->index();
$table->float('lat');
$table->float('long');
$table->timestamps();
$table->foreign('town_id')->references('id')->on('town')->onDelete('cascade');
}
);
DB::commit();
}
This then allows me to 'seed' the town table easily when I first create it, and wont interfere with any additions made to it at run time.
This is what I use in production.
Since I run migration on each deployment
artisan migrate
I create a seeder (just to keep seeding data out of migration for easy access later) and then run that seeder along with the migration
class YourTable extends Migration
{
/**
* Run the migrations.
*
* @return void
*/
public function up()
{
//migrate your table // Example
Schema::create('test_table', function(Blueprint $table)
{
$table->increments('id');
$table->timestamps();
$table->softDeletes();
});
//seed this table
$seeder = new YourTableSeeder();
$seeder->run();
}
/**
* Reverse the migrations.
*
* @return void
*/
public function down()
{
Schema::drop('test_table');
}
}
I do not add this seed call to seeds/DatabaseSeeder.php to avoid running it twice on a new installation.
The Artisan Command Solution
-
Create a new artisan command
php artisan make:command UpsertConfigurationTables
-
Paste this into the newly generated file:
UpsertConfigurationTables.php
<?php namespace App\Console\Commands; use Exception; use Illuminate\Console\Command; class UpsertConfigurationTables extends Command { /** * The name and signature of the console command. * * @var string */ protected $signature = 'upsert:configuration'; /** * The console command description. * * @var string */ protected $description = 'Upserts the configuration tables.'; /** * The models we want to upsert configuration data for * * @var array */ private $_models = [ 'App\ExampleModel' ]; /** * Create a new command instance. * * @return void */ public function __construct() { parent::__construct(); } /** * Execute the console command. * * @return mixed */ public function handle() { foreach ($this->_models as $model) { // check that class exists if (!class_exists($model)) { throw new Exception('Configuration seed failed. Model does not exist.'); } // check that seed data exists if (!defined($model . '::CONFIGURATION_DATA')) { throw new Exception('Configuration seed failed. Data does not exist.'); } /** * seed each record */ foreach ($model::CONFIGURATION_DATA as $row) { $record = $this->_getRecord($model, $row['id']); foreach ($row as $key => $value) { $this->_upsertRecord($record, $row); } } } } /** * _fetchRecord - fetches a record if it exists, otherwise instantiates a new model * * @param string $model - the model * @param integer $id - the model ID * * @return object - model instantiation */ private function _getRecord ($model, $id) { if ($this->_isSoftDeletable($model)) { $record = $model::withTrashed()->find($id); } else { $record = $model::find($id); } return $record ? $record : new $model; } /** * _upsertRecord - upsert a database record * * @param object $record - the record * @param array $row - the row of update data * * @return object */ private function _upsertRecord ($record, $row) { foreach ($row as $key => $value) { if ($key === 'deleted_at' && $this->_isSoftDeletable($record)) { if ($record->trashed() && !$value) { $record->restore(); } else if (!$record->trashed() && $value) { $record->delete(); } } else { $record->$key = $value; } } return $record->save(); } /** * _isSoftDeletable - Determines if a model is soft-deletable * * @param string $model - the model in question * * @return boolean */ private function _isSoftDeletable ($model) { $uses = array_merge(class_uses($model), class_uses(get_parent_class($model))); return in_array('Illuminate\Database\Eloquent\SoftDeletes', $uses); } }
Populate
$_models
with the Eloquent models you want to seed.-
Define the seed rows in the model:
const CONFIGURATION_DATA
<?php namespace App; use Illuminate\Database\Eloquent\Model; use Illuminate\Database\Eloquent\SoftDeletes; class ExampleModel extends Model { use SoftDeletes; const CONFIG_VALUE_ONE = 1; const CONFIG_VALUE_TWO = 2; const CONFIGURATION_DATA = [ [ 'id' => self::CONFIG_VALUE_ONE, 'col1' => 'val1', 'col2' => 'val2', 'deleted_at' => false ], [ 'id' => self::CONFIG_VALUE_TWO, 'col1' => 'val1', 'col2' => 'val2', 'deleted_at' => true ], ]; }
Add the command to your Laravel Forge deployment script (or any other CI deployment script):
php artisan upsert:configuration
Other noteworthy things:
- Upsert Functionality: If you ever want to alter any of the seeded rows, simply update them in your model and it was update your database values next time you deploy. It will never create duplicate rows.
-
Soft-Deletable Models: Note that you define deletions by setting
deleted_at
totrue
orfalse
. The Artisan command will handle calling the correct method to delete or recover your record.
Problems With Other Mentioned Solutions:
- Seeder: Running seeders in production is an abuse of the seeders. My concern would be that an engineer in the future would alter the seeders thinking that it's harmless since the documentation states that they are designed to seed test data.
- Migrations: Seeding data in a migration is strange and a an abuse of the purpose of the migration. It also doesn't let you update these values once your migration has been run.