Files
wms-app/scripts/normalize_document_dates.php
2026-08-17 13:12:07 +07:00

100 lines
5.1 KiB
PHP
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<?php
declare(strict_types=1);
$root = __DIR__ . '/../sdlc';
$english = [
'January'=>'01','February'=>'02','March'=>'03','April'=>'04','May'=>'05','June'=>'06',
'July'=>'07','August'=>'08','September'=>'09','October'=>'10','November'=>'11','December'=>'12',
'Jan'=>'01','Feb'=>'02','Mar'=>'03','Apr'=>'04','Jun'=>'06','Jul'=>'07','Aug'=>'08','Sep'=>'09','Oct'=>'10','Nov'=>'11','Dec'=>'12',
];
$thai = ['มกราคม'=>'01','กุมภาพันธ์'=>'02','มีนาคม'=>'03','เมษายน'=>'04','พฤษภาคม'=>'05','มิถุนายน'=>'06','กรกฎาคม'=>'07','สิงหาคม'=>'08','กันยายน'=>'09','ตุลาคม'=>'10','พฤศจิกายน'=>'11','ธันวาคม'=>'12'];
function normalDate(int|string $day, string $month, int|string $year, array $english, array $thai): string
{
$monthNo = $english[$month] ?? $thai[$month] ?? null;
if ($monthNo === null) return (string)$day . ' ' . $month . ' ' . (string)$year;
$yearNo = (int)$year;
if ($yearNo >= 2400) $yearNo -= 543;
return sprintf('%02d/%s/%02d', (int)$day, $monthNo, $yearNo % 100);
}
function normalizeDates(string $text, array $english, array $thai): string
{
$months = implode('|', array_map(static fn($v) => preg_quote($v, '/'), array_keys($english + $thai)));
// Compact Buddhist document/report dates, e.g. 25690817.
$text = preg_replace_callback('/\b(25\d{2})(\d{2})(\d{2})\b/', static function ($m) {
return sprintf('%02d/%02d/%02d', (int)$m[3], (int)$m[2], ((int)$m[1] - 543) % 100);
}, $text) ?? $text;
// Shared-month range: 11–13 May 2026.
$text = preg_replace_callback('/\b(\d{1,2})\s*[–-]\s*(\d{1,2})\s+('.$months.')\s+(20\d{2}|25\d{2})\b/u', static function ($m) use ($english,$thai) {
return normalDate($m[1],$m[3],$m[4],$english,$thai).'–'.normalDate($m[2],$m[3],$m[4],$english,$thai);
}, $text) ?? $text;
// Two-month range with a shared year: 5 January–14 August 2026.
$text = preg_replace_callback('/\b(\d{1,2})\s+('.$months.')\s*[–-]\s*(\d{1,2})\s+('.$months.')\s+(20\d{2}|25\d{2})\b/u', static function ($m) use ($english,$thai) {
return normalDate($m[1],$m[2],$m[5],$english,$thai).'–'.normalDate($m[3],$m[4],$m[5],$english,$thai);
}, $text) ?? $text;
// Hyphenated two-month range: 19-Feb–29-May-2026.
$text = preg_replace_callback('/\b(\d{1,2})-('.$months.')\s*[–-]\s*(\d{1,2})-('.$months.')-(20\d{2}|25\d{2})\b/u', static function ($m) use ($english,$thai) {
return normalDate($m[1],$m[2],$m[5],$english,$thai).'–'.normalDate($m[3],$m[4],$m[5],$english,$thai);
}, $text) ?? $text;
// A full single date using a month name or abbreviation.
$text = preg_replace_callback('/\b(\d{1,2})(?:-|\s+)('.$months.')(?:-|\s+)(20\d{2}|25\d{2})\b/u', static function ($m) use ($english,$thai) {
return normalDate($m[1],$m[2],$m[3],$english,$thai);
}, $text) ?? $text;
// Period descriptions without a repeated year, all known to be in 2026.
$text = preg_replace_callback('/\b(\d{1,2})\s*[–-]\s*(\d{1,2})\s+('.$months.')(?=\s+(?:commits|accounting|May|April|January|February|March|June|July|August))/u', static function ($m) use ($english,$thai) {
return normalDate($m[1],$m[3],2026,$english,$thai).'–'.normalDate($m[2],$m[3],2026,$english,$thai);
}, $text) ?? $text;
// Commit-period shorthand: 30-Apr through 08-May.
$text = preg_replace_callback('/\b(\d{1,2})-('.$months.')\s+through\s+(\d{1,2})-('.$months.')\b/u', static function ($m) use ($english,$thai) {
return normalDate($m[1],$m[2],2026,$english,$thai).'–'.normalDate($m[3],$m[4],2026,$english,$thai);
}, $text) ?? $text;
return $text;
}
$iterator = new RecursiveIteratorIterator(new RecursiveDirectoryIterator($root, FilesystemIterator::SKIP_DOTS));
$updated = 0;
foreach ($iterator as $file) {
if (!$file->isFile() || !str_starts_with($file->getFilename(), '200-WMS-')) continue;
$path = $file->getPathname();
$extension = strtolower($file->getExtension());
if ($extension === 'md') {
$before = file_get_contents($path);
if ($before === false) continue;
$after = normalizeDates($before, $english, $thai);
if ($after !== $before) {
file_put_contents($path, $after);
$updated++;
}
continue;
}
if (!in_array($extension, ['docx','xlsx'], true)) continue;
$zip = new ZipArchive();
if ($zip->open($path) !== true) continue;
$changed = false;
for ($i=0; $i<$zip->numFiles; $i++) {
$name = $zip->getNameIndex($i);
if (!str_ends_with($name, '.xml')) continue;
$before = $zip->getFromIndex($i);
$after = normalizeDates($before, $english, $thai);
if ($after === $before) continue;
$dom = new DOMDocument();
if (!$dom->loadXML($after)) throw new RuntimeException("Invalid XML after date conversion: {$path}::{$name}");
$zip->addFromString($name, $after);
$changed = true;
}
$zip->close();
if ($changed) $updated++;
}
echo "Updated {$updated} BRN WMS document(s).\n";