respect-validation/src-dev/Commands/UpdatePostalCodesCommand.php
Henrique Moody 7892a7c902
Port Bash scripts to PHP
It makes more sense to use PHP to generate PHP code than to use Bash. I
love writing Bash scripts, but I know it's not for everyone, and they
can become quite complex. Porting them to PHP code also lowers the
barrier for people to change them.

While I was making those changes, I also noticed a problem with how we
save the domain suffixes. We're converting all of them to ASCII, so we
are not preserving languages such as Chinese, Thai, and Hebrew, which
use non-ASCII characters.
2026-01-06 10:06:22 +01:00

130 lines
3.8 KiB
PHP

<?php
/*
* Copyright (c) Alexandre Gomes Gaigalas <alganet@gmail.com>
* SPDX-License-Identifier: MIT
*/
declare(strict_types=1);
namespace Respect\Dev\Commands;
use Symfony\Component\Console\Attribute\AsCommand;
use Symfony\Component\Console\Command\Command;
use Symfony\Component\Console\Input\InputInterface;
use Symfony\Component\Console\Output\OutputInterface;
use Symfony\Component\Console\Style\SymfonyStyle;
use Symfony\Component\VarExporter\VarExporter;
use function basename;
use function count;
use function dirname;
use function explode;
use function file_get_contents;
use function file_put_contents;
use function implode;
use function ksort;
use function preg_replace;
use function preg_replace_callback;
use function sprintf;
use function str_contains;
use function str_starts_with;
use function strlen;
use function trim;
use const PHP_EOL;
#[AsCommand(
name: 'update:postal-codes',
description: 'Update the list of postal codes in the PostalCode validator',
)]
final class UpdatePostalCodesCommand extends Command
{
private const string LIST_URL = 'https://download.geonames.org/export/dump/countryInfo.txt';
protected function execute(InputInterface $input, OutputInterface $output): int
{
$io = new SymfonyStyle($input, $output);
$io->title('Updating postal codes list');
// Download the list
$io->section('Downloading list');
$io->text(sprintf('Fetching from: %s', self::LIST_URL));
$listContent = file_get_contents(self::LIST_URL);
if ($listContent === false) {
$io->error('Failed to download postal codes list');
return Command::FAILURE;
}
$io->success('Downloaded successfully');
// Process the list
$io->section('Processing postal codes');
$lines = explode("\n", $listContent);
$postalCodes = [];
foreach ($lines as $line) {
$line = trim($line);
// Skip comments and empty lines
if ($line === '' || str_starts_with($line, '#')) {
continue;
}
// Split by tab
$parts = explode("\t", $line);
if (count($parts) < 15) {
continue;
}
$countryCode = $parts[0];
$countryFormat = $parts[13];
$countryRegex = trim($parts[14]);
if ($countryRegex === '') {
continue;
}
// Convert format
$countryFormat = preg_replace_callback('/(#+|@+)/', static function ($matches) {
$length = strlen($matches[0]);
$regex = str_contains($matches[0], '#') ? '\d' : '\w';
if ($length > 1) {
$regex .= '{' . $length . '}';
}
return $regex;
}, $countryFormat);
$postalCodes[$countryCode] = ['/^' . $countryFormat . '$/', '/' . $countryRegex . '/'];
}
ksort($postalCodes);
// Create the data file
$dataFilename = dirname(__DIR__, 2) . '/data/domain/postal-code.php';
$fileContent = implode(PHP_EOL, [
'<?php declare(strict_types=1);',
'// Copyright (c) https://download.geonames.org/export/dump/countryInfo.txt',
'// SPDX-License-Identifier: CC-BY-4.0',
'return ' . preg_replace('/\\\([dws])/', '\\1', VarExporter::export($postalCodes)) . ';' . PHP_EOL,
]);
// Write the data file
if (file_put_contents($dataFilename, $fileContent) === false) {
$io->error('Failed to write data file');
return Command::FAILURE;
}
$io->success(sprintf('Updated %s successfully', basename($dataFilename)));
$io->text(sprintf('Total postal codes: %d', count($postalCodes)));
return Command::SUCCESS;
}
}