evus.acvisa.com/app/home/controller/Seo.php

<?php
declare(strict_types=1);

namespace app\home\controller;

use api\Httpcurl;
use app\service\SeoMetaService;
use app\service\SitemapService;
use think\facade\Log;
use Throwable;

final class Seo
{
    public function sitemap()
    {
        $seoConfig = (array) config('seo');
        $sitemapConfig = (array) ($seoConfig['sitemap'] ?? []);
        $baseUrl = (new SeoMetaService($seoConfig))->baseUrl();
        if ($baseUrl === '') {
            return response(
                'SEO_BASE_URL is not configured.',
                503,
                ['Content-Type' => 'text/plain; charset=UTF-8']
            );
        }

        $cacheSeconds = max(60, (int) ($sitemapConfig['cache_seconds'] ?? 3600));
        $cacheDirectory = runtime_path('seo');
        $cacheFile = $cacheDirectory
            . 'sitemap-'
            . substr(hash('sha256', $baseUrl), 0, 16)
            . '.xml';
        $cachedXml = $this->readCache($cacheFile);
        $cacheTime = is_file($cacheFile) ? filemtime($cacheFile) : false;
        if (
            $cachedXml !== ''
            && $cacheTime !== false
            && $cacheTime + $cacheSeconds > time()
        ) {
            return $this->xmlResponse($cachedXml, $cacheSeconds);
        }

        $service = new SitemapService($baseUrl);
        $entries = $this->staticEntries($seoConfig);
        [$newsEntries, $newsComplete] = $this->newsEntries($sitemapConfig);

        if (!$newsComplete && $cachedXml !== '') {
            return $this->xmlResponse($cachedXml, 300);
        }

        $xml = $service->render(array_merge($entries, $newsEntries));
        if ($newsComplete) {
            $this->writeCache($cacheDirectory, $cacheFile, $xml);
        }

        return $this->xmlResponse($xml, $newsComplete ? $cacheSeconds : 300);
    }

    public function robots()
    {
        $seoConfig = (array) config('seo');
        $baseUrl = (new SeoMetaService($seoConfig))->baseUrl();
        $service = new SitemapService($baseUrl);
        $robots = $service->renderRobots($this->disallowPaths($seoConfig));

        return response($robots, 200, [
            'Content-Type' => 'text/plain; charset=UTF-8',
            'Cache-Control' => 'public, max-age=3600',
        ]);
    }

    private function staticEntries(array $seoConfig): array
    {
        $entries = [];
        foreach ((array) ($seoConfig['pages'] ?? []) as $route => $page) {
            $page = (array) $page;
            if (!str_starts_with((string) $route, 'home/')) {
                continue;
            }
            if (!str_starts_with((string) ($page['robots'] ?? ''), 'index')) {
                continue;
            }
            if ($route === 'home/news/detail') {
                continue;
            }

            $path = (string) ($page['canonical_path'] ?? '');
            if ($path !== '') {
                $entries[] = ['path' => $path];
            }
        }

        return $entries;
    }

    private function newsEntries(array $sitemapConfig): array
    {
        $url = trim((string) ($sitemapConfig['news_list_url'] ?? ''));
        if (filter_var($url, FILTER_VALIDATE_URL) === false) {
            Log::warning('SEO sitemap news URL is invalid.');
            return [[], false];
        }

        $pageSize = max(1, min(500, (int) ($sitemapConfig['news_page_size'] ?? 100)));
        $maxPages = max(1, min(500, (int) ($sitemapConfig['max_news_pages'] ?? 100)));
        $entries = [];
        $fetchedItems = 0;

        for ($page = 1; $page <= $maxPages; $page++) {
            try {
                $response = Httpcurl::request($url, 'post', [
                    'model' => 'article',
                    'source_code' => 'cn',
                    'page' => $page,
                    'size' => $pageSize,
                ], [], 5);
            } catch (Throwable $exception) {
                Log::warning('SEO sitemap news request failed: ' . $exception->getMessage());
                return [$entries, false];
            }

            if (!is_array($response) || !isset($response[0]) || !empty($response[3])) {
                Log::warning('SEO sitemap news request returned an invalid response.');
                return [$entries, false];
            }

            $statusCode = (int) ($response[2]['http_code'] ?? 0);
            if ($statusCode < 200 || $statusCode >= 300) {
                Log::warning('SEO sitemap news request returned an invalid HTTP status.');
                return [$entries, false];
            }

            $result = json_decode((string) $response[0], true);
            if (
                !is_array($result)
                || (int) ($result['code'] ?? 0) !== 1
                || !is_array($result['data'] ?? null)
            ) {
                Log::warning('SEO sitemap news response could not be decoded.');
                return [$entries, false];
            }

            $items = $result['data'];
            $fetchedItems += count($items);
            foreach ($items as $item) {
                if (!is_array($item)) {
                    continue;
                }

                $id = filter_var($item['id'] ?? null, FILTER_VALIDATE_INT, [
                    'options' => ['min_range' => 1],
                ]);
                if ($id === false) {
                    continue;
                }

                $entries[] = [
                    'path' => '/home/news/detail',
                    'query' => ['id' => $id],
                    'lastmod' => $this->articleDate((string) ($item['create_time'] ?? '')),
                ];
            }

            $total = max(0, (int) ($result['count'] ?? 0));
            if (
                $items === []
                || ($total > 0 && $fetchedItems >= $total)
                || ($total === 0 && count($items) < $pageSize)
            ) {
                return [$entries, true];
            }
        }

        Log::warning('SEO sitemap news pagination reached the configured page limit.');
        return [$entries, false];
    }

    private function disallowPaths(array $seoConfig): array
    {
        $paths = [];
        foreach ((array) ($seoConfig['noindex_controllers'] ?? []) as $controller) {
            $controller = trim((string) $controller, '/');
            if ($controller !== '') {
                $paths[] = '/home/' . $controller . '/';
            }
        }

        return $paths;
    }

    private function articleDate(string $date): string
    {
        $timestamp = strtotime($date);
        return $timestamp === false ? '' : date('Y-m-d', $timestamp);
    }

    private function readCache(string $cacheFile): string
    {
        if (!is_file($cacheFile) || !is_readable($cacheFile)) {
            return '';
        }

        $contents = file_get_contents($cacheFile);
        return is_string($contents) ? $contents : '';
    }

    private function writeCache(string $directory, string $cacheFile, string $xml): void
    {
        if (!is_dir($directory) && !mkdir($directory, 0775, true) && !is_dir($directory)) {
            Log::warning('SEO sitemap cache directory could not be created.');
            return;
        }

        if (file_put_contents($cacheFile, $xml, LOCK_EX) === false) {
            Log::warning('SEO sitemap cache could not be written.');
        }
    }

    private function xmlResponse(string $xml, int $maxAge)
    {
        return response($xml, 200, [
            'Content-Type' => 'application/xml; charset=UTF-8',
            'Cache-Control' => 'public, max-age=' . max(0, $maxAge),
            'X-Robots-Tag' => 'noindex',
        ]);
    }
}