symfony · nicolas-grekas · Aug 21, 2025 · Aug 21, 2025
diff --git a/UPGRADE-8.0.md b/UPGRADE-8.0.md
@@ -14,6 +14,11 @@ AssetMapper
 
  * Remove `ImportMapConfigReader::splitPackageNameAndFilePath()`, use `ImportMapEntry::splitPackageNameAndFilePath()` instead
 
+BrowserKit
+----------
+
+ * Remove `AbstractBrowser::useHtml5Parser()`; the native HTML5 parser is used unconditionally
+
 Cache
 -----
 
@@ -117,6 +122,11 @@ DoctrineBridge
  * Remove support for auto-mapping Doctrine entities to controller arguments; use explicit mapping instead
  * Make `ProxyCacheWarmer` class `final`
 
+DomCrawler
+----------
+
+ * Remove argument `$useHtml5Parser` of `Crawler`'s constructor; the native HTML5 parser is used unconditionally
+
 ExpressionLanguage
 ------------------
 

diff --git a/src/Symfony/Component/BrowserKit/AbstractBrowser.php b/src/Symfony/Component/BrowserKit/AbstractBrowser.php
@@ -46,8 +46,6 @@ abstract class AbstractBrowser
     /** @psalm-var TResponse */
     protected object $response;
     protected Crawler $crawler;
-    /** @deprecated since Symfony 7.4, to be removed in Symfony 8 */
-    protected bool $useHtml5Parser = true;
     protected bool $insulated = false;
     protected ?string $redirect;
     protected bool $followRedirects = true;
@@ -202,24 +200,6 @@ public function getCrawler(): Crawler
         return $this->crawler ?? throw new BadMethodCallException(\sprintf('The "request()" method must be called before "%s()".', __METHOD__));
     }
 
-    /**
-     * Sets whether parsing should be done using "masterminds/html5".
-     *
-     * @deprecated since Symfony 7.4, Symfony 8 will unconditionally use the native HTML5 parser
-     *
-     * @return $this
-     */
-    public function useHtml5Parser(bool $useHtml5Parser): static
-    {
-        if (\PHP_VERSION_ID >= 80400) {
-            trigger_deprecation('symfony/browser-kit', '7.4', 'Method "%s()" is deprecated. Symfony 8 will unconditionally use the native HTML5 parser.', __METHOD__);
-        }
-
-        $this->useHtml5Parser = $useHtml5Parser;
-
-        return $this;
-    }
-
     /**
      * Returns the current BrowserKit Response instance.
      */
@@ -507,7 +487,7 @@ protected function createCrawlerFromContent(string $uri, string $content, string
             return null;
         }
 
-        $crawler = new Crawler(null, $uri, null, $this->useHtml5Parser);
+        $crawler = new Crawler(null, $uri, null);
         $crawler->addContent($content, $type);
 
         return $crawler;

diff --git a/src/Symfony/Component/BrowserKit/CHANGELOG.md b/src/Symfony/Component/BrowserKit/CHANGELOG.md
@@ -1,6 +1,11 @@
 CHANGELOG
 =========
 
+8.0
+---
+
+ * Remove `AbstractBrowser::useHtml5Parser()`; the native HTML5 parser is used unconditionally
+
 7.4
 ---
 

diff --git a/src/Symfony/Component/DomCrawler/CHANGELOG.md b/src/Symfony/Component/DomCrawler/CHANGELOG.md
@@ -1,6 +1,11 @@
 CHANGELOG
 =========
 
+8.0
+---
+
+ * Remove argument `$useHtml5Parser` of `Crawler`'s constructor; the native HTML5 parser is used unconditionally
+
 7.4
 ---
 

diff --git a/src/Symfony/Component/DomCrawler/Crawler.php b/src/Symfony/Component/DomCrawler/Crawler.php
@@ -11,7 +11,6 @@
 
 namespace Symfony\Component\DomCrawler;
 
-use Masterminds\HTML5;
 use Symfony\Component\CssSelector\CssSelectorConverter;
 
 /**
@@ -53,23 +52,15 @@ class Crawler implements \Countable, \IteratorAggregate
      */
     private bool $isHtml = true;
 
-    private ?HTML5 $html5Parser = null;
-
     /**
      * @param \DOMNodeList|\DOMNode|\DOMNode[]|string|null $node A Node to use as the base for the crawling
      */
     public function __construct(
         \DOMNodeList|\DOMNode|array|string|null $node = null,
         protected ?string $uri = null,
         ?string $baseHref = null,
-        private bool $useHtml5Parser = true,
     ) {
-        if (\PHP_VERSION_ID >= 80400 && !$useHtml5Parser) {
-            trigger_deprecation('symfony/dom-crawler', '7.4', 'Disabling HTML5 parsing is deprecated. Symfony 8 will unconditionally use the native HTML5 parser.');
-        }
-
         $this->baseHref = $baseHref ?: $uri;
-        $this->html5Parser = \PHP_VERSION_ID < 80400 && $useHtml5Parser ? new HTML5(['disable_html_ns' => true]) : null;
         $this->cachedNamespaces = new \ArrayObject();
 
         $this->add($node);
@@ -175,7 +166,7 @@ public function addContent(string $content, ?string $type = null): void
      */
     public function addHtmlContent(string $content, string $charset = 'UTF-8'): void
     {
-        $dom = $this->parseHtmlString($content, $charset);
+        $dom = $this->parseXhtml($content, $charset);
         $this->addDocument($dom);
 
         $base = $this->filterRelativeXPath('descendant-or-self::base')->extract(['href']);
@@ -609,10 +600,6 @@ public function html(?string $default = null): string
         $node = $this->getNode(0);
         $owner = $node->ownerDocument;
 
-        if ($this->html5Parser && '<!DOCTYPE html>' === $owner->saveXML($owner->childNodes[0])) {
-            $owner = $this->html5Parser;
-        }
-
         $html = '';
         foreach ($node->childNodes as $child) {
             $html .= $owner->saveHTML($child);
@@ -630,10 +617,6 @@ public function outerHtml(): string
         $node = $this->getNode(0);
         $owner = $node->ownerDocument;
 
-        if ($this->html5Parser && '<!DOCTYPE html>' === $owner->saveXML($owner->childNodes[0])) {
-            $owner = $this->html5Parser;
-        }
-
         return $owner->saveHTML($node);
     }
 
@@ -1064,48 +1047,8 @@ protected function sibling(\DOMNode $node, string $siblingDir = 'nextSibling'):
         return $nodes;
     }
 
-    private function parseHtml5(string $htmlContent, string $charset = 'UTF-8'): \DOMDocument
-    {
-        if (!$this->supportsEncoding($charset)) {
-            $htmlContent = $this->convertToHtmlEntities($htmlContent, $charset);
-            $charset = 'UTF-8';
-        }
-
-        return $this->html5Parser->parse($htmlContent, ['encoding' => $charset]);
-    }
-
-    private function supportsEncoding(string $encoding): bool
-    {
-        try {
-            return '' === @mb_convert_encoding('', $encoding, 'UTF-8');
-        } catch (\Throwable $e) {
-            return false;
-        }
-    }
-
     private function parseXhtml(string $htmlContent, string $charset = 'UTF-8'): \DOMDocument
     {
-        if (\PHP_VERSION_ID < 80400 || !$this->useHtml5Parser) {
-            if ('UTF-8' === $charset && preg_match('//u', $htmlContent)) {
-                $htmlContent = '<?xml encoding="UTF-8">'.$htmlContent;
-            } else {
-                $htmlContent = $this->convertToHtmlEntities($htmlContent, $charset);
-            }
-
-            $internalErrors = libxml_use_internal_errors(true);
-
-            $dom = new \DOMDocument('1.0', $charset);
-            $dom->validateOnParse = true;
-
-            if ('' !== trim($htmlContent)) {
-                @$dom->loadHTML($htmlContent);
-            }
-
-            libxml_use_internal_errors($internalErrors);
-
-            return $dom;
-        }
-
         $document = @\Dom\HTMLDocument::createFromString($htmlContent, \Dom\HTML_NO_DEFAULT_NS, $charset);
         $htmlContent = $document->saveXml();
         $charset = $document->inputEncoding;
@@ -1202,7 +1145,6 @@ private function createSubCrawler(\DOMNodeList|\DOMNode|array|string|null $nodes
         $crawler->document = $this->document;
         $crawler->namespaces = $this->namespaces;
         $crawler->cachedNamespaces = $this->cachedNamespaces;
-        $crawler->html5Parser = $this->html5Parser;
 
         return $crawler;
     }
@@ -1219,39 +1161,6 @@ private function createCssSelectorConverter(): CssSelectorConverter
         return new CssSelectorConverter($this->isHtml);
     }
 
-    /**
-     * Parse string into DOMDocument object using HTML5 parser if the content is HTML5 and the library is available.
-     * Use libxml parser otherwise.
-     */
-    private function parseHtmlString(string $content, string $charset): \DOMDocument
-    {
-        if ($this->canParseHtml5String($content)) {
-            return $this->parseHtml5($content, $charset);
-        }
-
-        return $this->parseXhtml($content, $charset);
-    }
-
-    private function canParseHtml5String(string $content): bool
-    {
-        if (!$this->html5Parser) {
-            return false;
-        }
-
-        if (false === $pos = stripos($content, '<!doctype html>')) {
-            return false;
-        }
-
-        $header = substr($content, 0, $pos);
-
-        return '' === $header || $this->isValidHtml5Heading($header);
-    }
-
-    private function isValidHtml5Heading(string $heading): bool
-    {
-        return 1 === preg_match('/^\x{FEFF}?\s*(<!--[^>]*?-->\s*)*$/u', $heading);
-    }
-
     private function normalizeWhitespace(string $string): string
     {
         return trim(preg_replace("/(?:[ \n\r\t\x0C]{2,}+|[\n\r\t\x0C])/", ' ', $string), " \n\r\t\x0C");

diff --git a/src/Symfony/Component/DomCrawler/Tests/CrawlerTestCase.php b/src/Symfony/Component/DomCrawler/Tests/CrawlerTestCase.php
@@ -12,8 +12,6 @@
 namespace Symfony\Component\DomCrawler\Tests;
 
 use PHPUnit\Framework\Attributes\DataProvider;
-use PHPUnit\Framework\Attributes\Group;
-use PHPUnit\Framework\Attributes\IgnoreDeprecations;
 use PHPUnit\Framework\Attributes\RequiresPhpExtension;
 use PHPUnit\Framework\Error\Notice;
 use PHPUnit\Framework\TestCase;
@@ -1329,22 +1327,6 @@ public function testAddXmlContentWithErrors()
         libxml_use_internal_errors($internalErrors);
     }
 
-    #[IgnoreDeprecations]
-    #[Group('legacy')]
-    public function testHtml5ParserNotSameAsNativeParserForSpecificHtml()
-    {
-        // Html who create a bug specific to the DOM extension (see https://github.com/symfony/symfony/issues/28596)
-        $html = '<!DOCTYPE html><html><body><h1><p>Foo</p></h1></body></html>';
-
-        $html5Crawler = new Crawler(null, null, null, true);
-        $html5Crawler->add($html);
-
-        $nativeCrawler = new Crawler(null, null, null, false);
-        $nativeCrawler->add($html);
-
-        $this->assertNotEquals($nativeCrawler->filterXPath('//h1')->text(), $html5Crawler->filterXPath('//h1')->text(), 'Native parser and Html5 parser must be different');
-    }
-
     public function testAddHtml5()
     {
         // Ensure a bug specific to the DOM extension is fixed (see https://github.com/symfony/symfony/issues/28596)

diff --git a/src/Symfony/Component/DomCrawler/Tests/LegacyHtml5ParserCrawlerTest.php b/src/Symfony/Component/DomCrawler/Tests/LegacyHtml5ParserCrawlerTest.php