π Populate your database using the GeoNames service
The package is useful for applications that rely on the geo data.
By default, it provides 4 models: Continent, Country, Division, City and translations for them.
The translations are powered by the nevadskiy/laravel-translatable package.
The package also keeps the data up-to-date by fetching daily modifications provided by the GeoNames service and uses them to synchronize your own database.
composer require nevadskiy/laravel-geonamesIf you are going to use translations, you also need to install an additional package.
composer require nevadskiy/laravel-translatable- Laravel
8.0or newer - PHP
7.3or newer
Publish the package resources using the command:
php artisan vendor:publish --tag=geonames-migrations --tag=geonames-modelsBefore seeding, make sure you run the database migrations.
php artisan migrateThen, run the seeding process.
php artisan geonames:seedIt will download geonames resources and insert the dataset into your database.
Note that the seeding process may take some time. On average, it takes about 40 minutes (without downloading time) to seed the full dataset with translations.
If you have issues with memory leaks during seeding, check out this section.
Add the following code to the app/Console/Kernel.php file if you want to receive geonames daily updates.
Geonames daily updates are published at 3:00 in the UTC time zone, so to be sure that they are already available, it is recommended to run the command a bit later.
protected function schedule(Schedule $schedule)
{
$schedule->command('geonames:daily-update')->dailyAt('4:00');
}Note that time is specified for the
UTCtimezone.
If you missed some daily updates or just decided to change seeder filters, you can sync your database records according to the current geonames dataset.
php artisan geonames:syncThis command will create missing records, remove redundant ones, and updated modified ones according to the current dataset.
Note that the
geoname_idandalternate_name_idfields is required to synchronize data.
If you want to customize migrations or data that should be imported, you can simply do this by overriding the default seeders.
To do that, publish the package seeders using command:
php artisan vendor:publish --tag=geonames-seedersPublish the package config:
php artisan vendor:publish --tag=geonames-config Then, specify published seeders in the config/geonames.php file:
'seeders' => [
Database\Seeders\Geo\ContinentSeeder::class,
Database\Seeders\Geo\ContinentTranslationSeeder::class,
Database\Seeders\Geo\CountrySeeder::class,
Database\Seeders\Geo\CountryTranslationSeeder::class,
Database\Seeders\Geo\DivisionSeeder::class,
Database\Seeders\Geo\DivisionTranslationSeeder::class,
Database\Seeders\Geo\CitySeeder::class,
Database\Seeders\Geo\CityTranslationSeeder::class,
]To reduce the database size, you can set up filters for seeding only those geo data that you really need in your application.
For example, you can set the minimum population for the city. All cities with smaller population will not be imported.
To do that, override the $minPopulation property in the CitySeeder class.
To have full control over this behaviour, override the filter method of the seeder.
To add custom fields to the table, you also need to tell the seeder how to fill those fields using the mapAttributes method.
The mapAttributes method should return all attributes of the database record, including timestamps, because model events will not be fired during seeding process since the package uses a bulk insert strategy.
However, all model casts and mutators will be applied as usual.
For example, if you want to use UUIDs as primary keys, you can extend the original seeder as following:
<?php
namespace Database\Seeders\Geo;
use App\Models\Geo\City;
use Illuminate\Support\Str;use Nevadskiy\Geonames\Seeders\CitySeeder as Seeder;
class CitySeeder extends Seeder
{
/**
* {@inheritdoc}
*/
protected static $model = City::class;
/**
* {@inheritdoc}
*/
protected function mapAttributes(array $record): array
{
return array_merge(parent::mapAttributes($record), [
'id' => (string) Str::uuid(),
]);
}
}For a more significant change in the structure, you can add your own seeders or extend existing ones.
Each seeder must implement the Nevadskiy\Geonames\Seeders\Seeder interface.
All seeders that are specified in the geonames config file will be executed one by one in the specified order.
To use translations you need to install the nevadskiy/laravel-translatable package.
Read its documentation to learn more about how it works. You can also use it to handle translations of other models.
Otherwise, you still can use the package without translations, just simply remove the following:
- translation migrations
- translation seeders (from the
geonamesconfig file as well) - the
HasTranslationstrait andtranslatableprop from published models
One of the most popular issues associated with seeding large amounts of data is a memory leak.
This package reads files using PHP generators and lazy collections to avoid loading the entire file into memory.
However, there are packages that log database queries and model events during long-running commands to memory, which leads to memory leaks.
There are instructions on how to avoid memory leaks when working with the following packages:
Publish flare config using php artisan vendor:publish --tag=flare-config.
Set report_query_bindings to false in the flare.php config file as following:
'flare_middleware' => [
...
AddQueries::class => [
'maximum_number_of_collected_queries' => 200,
'report_query_bindings' => false,
],
...
]Update the telescope.php config file as following:
'ignore_commands' => [
'geonames:seed',
'geonames:daily-update',
'geonames:sync',
]Please see CHANGELOG for more information what has changed recently.
Please see CONTRIBUTING for more information.
The MIT License (MIT). Please see LICENSE for more information.