init

2026-09-15 10:35:56 +08:00
commit 782f8ea7aa
4086 changed files with 631288 additions and 0 deletions
@@ -0,0 +1,76 @@
+<?php
+declare(strict_types=1);
+
+namespace Esun3d\LanguageUrlStoreResolver\Model;
+
+class HtmlLinkProcessor
+{
+    /**
+     * @var UrlPathConverter
+     */
+    private $urlPathConverter;
+
+    /**
+     * @param UrlPathConverter $urlPathConverter
+     */
+    public function __construct(UrlPathConverter $urlPathConverter)
+    {
+        $this->urlPathConverter = $urlPathConverter;
+    }
+
+    /**
+     * Convert internal anchor URLs in complete page HTML.
+     *
+     * @param string $html
+     * @param string $language
+     * @return string
+     */
+    public function process(string $html, string $language): string
+    {
+        if (stripos($html, '<a') === false) {
+            return $html;
+        }
+
+        $result = preg_replace_callback(
+            '~(<a\b[^>]*\bhref\s*=\s*)(["\'])(.*?)\2~is',
+            function (array $matches) use ($language): string {
+                $href = $matches[3];
+                $normalizedHref = trim($href);
+
+                if (!$this->isConvertibleHref($normalizedHref)) {
+                    return $matches[0];
+                }
+
+                $convertedHref = $this->urlPathConverter->toLanguageUrl($normalizedHref, $language);
+                if ($convertedHref === $normalizedHref) {
+                    return $matches[0];
+                }
+
+                return $matches[1] . $matches[2] . $convertedHref . $matches[2];
+            },
+            $html
+        );
+
+        return is_string($result) ? $result : $html;
+    }
+
+    /**
+     * Check whether an href represents a navigational HTTP URL.
+     *
+     * @param string $href
+     * @return bool
+     */
+    private function isConvertibleHref(string $href): bool
+    {
+        if ($href === '' || $href[0] === '#' || $href[0] === '?') {
+            return false;
+        }
+
+        if (strpos($href, '{{') !== false || strpos($href, '${') !== false) {
+            return false;
+        }
+
+        return !preg_match('~^[a-z][a-z0-9+.-]*:~i', $href)
+            || preg_match('~^https?:~i', $href) === 1;
+    }
+}