From e6dda6dc9813471359cb2441a9aeda1fa1abea33 Mon Sep 17 00:00:00 2001
From: Petr Chromec <mortal_flesh@seznam.cz>
Date: Wed, 25 Oct 2023 17:22:27 +0200
Subject: [PATCH] Encode html entities in HtmlHelper to prevent parse errors

---
 CHANGELOG.md                     |  1 +
 src/Service/HtmlHelper.php       | 15 +++++++++++++++
 tests/Service/HtmlHelperTest.php | 18 ++++++++++++++++++
 3 files changed, 34 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 76c8be8..0126e28 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,7 @@
 
 <!-- There is always Unreleased section on the top. Subsections (Added, Changed, Fixed, Removed) should be added as needed. -->
 ## Unreleased
+- Encode html entities in `HtmlHelper` to prevent parse errors
 
 ## 7.3.0 - 2023-05-23
 - Support `figure` html tag in `HtmlHelper::xpathHtmlDocument` method
diff --git a/src/Service/HtmlHelper.php b/src/Service/HtmlHelper.php
index 50ff40c..83a7c61 100644
--- a/src/Service/HtmlHelper.php
+++ b/src/Service/HtmlHelper.php
@@ -26,6 +26,7 @@ public function xpathHtmlDocument(string $content, string $xpathQuery): ISeq
     {
         return Seq::init(function () use ($xpathQuery, $content) {
             $htmlContent = $this->transformUnsupportedHtml($content);
+            $htmlContent = $this->encodeHtml($htmlContent);
 
             // @see https://www.php.net/manual/en/domdocument.loadhtml.php#95251
             $dom = new \DOMDocument();
@@ -40,6 +41,20 @@ public function xpathHtmlDocument(string $content, string $xpathQuery): ISeq
         });
     }
 
+    /**
+     * @see https://stackoverflow.com/questions/1685277/warning-domdocumentloadhtml-htmlparseentityref-expecting-in-entity
+     *
+     * It is meant to encode html entities, which would otherwise break the DOMDocument::loadHTML() method.
+     */
+    private function encodeHtml(string $content): string
+    {
+        return str_replace(
+            ['&gt;', '&lt;'],
+            ['>', '<'],
+            htmlentities($content, ENT_NOQUOTES, 'UTF-8', false)
+        );
+    }
+
     private function transformUnsupportedHtml(string $originalContent): string
     {
         $unsupportedTags = Map::from([
diff --git a/tests/Service/HtmlHelperTest.php b/tests/Service/HtmlHelperTest.php
index 3a4c711..58bd7a0 100644
--- a/tests/Service/HtmlHelperTest.php
+++ b/tests/Service/HtmlHelperTest.php
@@ -263,6 +263,24 @@ public function linksProvider(): array
                     ]),
                 ],
             ],
+            'with one link containing not encoded entities' => [
+                '<div>content<a href="https://www.vysokeskoly.cz?param1=1&param2=2" class="active"></div>',
+                [
+                    ' href="https://www.vysokeskoly.cz?param1=1&param2=2" class="active"' => new Link([
+                        'href' => 'https://www.vysokeskoly.cz?param1=1&param2=2',
+                        'class' => 'active',
+                    ]),
+                ],
+            ],
+            'with one link containing both encoded and not encoded entities' => [
+                '<div>content<a href="https://www.vysokeskoly.cz?param1=1&param2=2&amp;param3=3" class="active"></div>',
+                [
+                    ' href="https://www.vysokeskoly.cz?param1=1&param2=2&amp;param3=3" class="active"' => new Link([
+                        'href' => 'https://www.vysokeskoly.cz?param1=1&param2=2&param3=3',
+                        'class' => 'active',
+                    ]),
+                ],
+            ],
             'with multi line link' => [
                 '<figure class="wp-block-image size-large is-resized">
                     content