VITEC-website/migrations/check_product_downloads.php

<?php

declare(strict_types=1);

/**
 * Comparison report: download records vs. the delivered product workbooks
 * (request Olli 2026-10-28). Per product: linked downloads
 * (tx_vitec_product_download_mm, uid_local = product), datasheet coverage
 * (download title suffix "datasheet" or a Datasheet category), the manual
 * Resources-tab selection (tx_vitec_product_datasheet_mm) - matched against
 * the workbook coverage from migrations/products_xlsx/manifest.json.
 *
 * READ-ONLY. Usage: php migrations/check_product_downloads.php
 */

error_reporting(E_ALL);
ini_set('display_errors', '1');

$root = null;
$dir = __DIR__;
for ($i = 0; $i < 5; $i++) {
    if (is_file($dir . '/config/system/settings.php')) { $root = $dir; break; }
    $dir = dirname($dir);
}
if ($root === null) { exit("no project root\n"); }
$settings = require $root . '/config/system/settings.php';
$db = $settings['DB']['Connections']['Default'];
$pdo = new PDO(
    sprintf('mysql:host=%s;port=%s;dbname=%s;charset=utf8mb4', $db['host'] ?? 'localhost', $db['port'] ?? 3306, $db['dbname'] ?? ''),
    $db['user'] ?? '', $db['password'] ?? '',
    [PDO::ATTR_ERRMODE => PDO::ERRMODE_EXCEPTION, PDO::ATTR_DEFAULT_FETCH_MODE => PDO::FETCH_ASSOC]
);

// --- workbook coverage from the manifest ------------------------------------
$manifest = json_decode((string)file_get_contents($root . '/migrations/products_xlsx/manifest.json'), true) ?: [];
$workbookSlugs = [];
foreach (($manifest['workbooks'] ?? []) as $file => $entry) {
    if (!empty($entry['product'])) {
        $workbookSlugs[(string)$entry['product']] = basename((string)$file);
    }
}

// --- data ---------------------------------------------------------------------
$products = $pdo->query('SELECT uid, title, slug, hidden, legacy, supportproduct, subproduct FROM tx_vitec_domain_model_product WHERE deleted = 0 ORDER BY title')->fetchAll();
$downloads = $pdo->query('SELECT uid, title, hidden FROM tx_vitec_domain_model_download WHERE deleted = 0')->fetchAll();
$downloadTitle = array_column($downloads, 'title', 'uid');

$linksByProduct = [];
$linkedDownloadUids = [];
foreach ($pdo->query('SELECT uid_local, uid_foreign FROM tx_vitec_product_download_mm') as $r) {
    $linksByProduct[(int)$r['uid_local']][] = (int)$r['uid_foreign'];
    $linkedDownloadUids[(int)$r['uid_foreign']] = true;
}
$sheetsByProduct = [];
foreach ($pdo->query('SELECT uid_local, COUNT(*) c FROM tx_vitec_product_datasheet_mm GROUP BY uid_local') as $r) {
    $sheetsByProduct[(int)$r['uid_local']] = (int)$r['c'];
}

$isDatasheet = static function (string $title): bool {
    return (bool)preg_match('/data\s*sheet\s*$/i', trim($title));
};
$docType = static function (string $title): string {
    foreach (['datasheet', 'data sheet', 'brochure', 'user guide', 'quick start guide', 'application note',
              'white paper', 'case study', 'success story', 'manual', 'firmware', 'software', 'flyer', 'guide'] as $suffix) {
        if (preg_match('/' . preg_quote($suffix, '/') . '\s*$/i', trim($title))) {
            return $suffix;
        }
    }
    return '(no type suffix)';
};

// --- per-product evaluation ------------------------------------------------------
$noDownloads = [];
$noDatasheet = [];
$okCount = 0;
$withoutWorkbook = [];
foreach ($products as $p) {
    $uid = (int)$p['uid'];
    $slug = (string)$p['slug'];
    $linked = $linksByProduct[$uid] ?? [];
    $hasSheet = false;
    foreach ($linked as $dUid) {
        if ($isDatasheet((string)($downloadTitle[$dUid] ?? ''))) { $hasSheet = true; break; }
    }
    $flag = ((int)$p['legacy'] ? ' [legacy]' : '') . ((int)$p['hidden'] ? ' [hidden]' : '')
        . ((int)$p['supportproduct'] ? ' [support]' : '') . ((int)$p['subproduct'] ? ' [sub]' : '');
    $line = sprintf('#%-4d %-45s dl=%-2d sheetsTab=%d%s', $uid, $p['title'], count($linked), $sheetsByProduct[$uid] ?? 0, $flag);
    if (!isset($workbookSlugs[$slug])) {
        $withoutWorkbook[] = $line;
    }
    if (isset($workbookSlugs[$slug])) {
        if ($linked === []) { $noDownloads[] = $line; }
        elseif (!$hasSheet) { $noDatasheet[] = $line; }
        else { $okCount++; }
    }
}

// --- orphan downloads ---------------------------------------------------------------
$orphans = [];
$orphanTypes = [];
foreach ($downloads as $d) {
    if (!isset($linkedDownloadUids[(int)$d['uid']])) {
        $type = $docType((string)$d['title']);
        $orphanTypes[$type] = ($orphanTypes[$type] ?? 0) + 1;
        $orphans[] = $d['title'];
    }
}

printf("Produkte: %d (davon %d mit Workbook) · Downloads: %d (davon %d verlinkt, %d verwaist)\n\n",
    count($products), count($workbookSlugs), count($downloads), count($linkedDownloadUids), count($orphans));

echo "=== A) Workbook-Produkte OHNE verknuepfte Downloads (" . count($noDownloads) . ") ===\n";
foreach ($noDownloads as $l) { echo "  $l\n"; }

echo "\n=== B) Workbook-Produkte mit Downloads, aber OHNE Datasheet (" . count($noDatasheet) . ") ===\n";
foreach ($noDatasheet as $l) { echo "  $l\n"; }

echo "\n=== C) Workbook-Produkte mit Datasheet-Abdeckung: $okCount ===\n";

echo "\n=== D) Produkte OHNE Workbook (" . count($withoutWorkbook) . ") - zur Einordnung ===\n";
foreach ($withoutWorkbook as $l) { echo "  $l\n"; }

arsort($orphanTypes);
echo "\n=== E) Verwaiste Downloads nach Typ ===\n";
foreach ($orphanTypes as $type => $n) { printf("  %-22s %d\n", $type, $n); }
echo "\n  Beispiele (max 30):\n";
foreach (array_slice($orphans, 0, 30) as $t) { echo "    - $t\n"; }