Adjust code style in HTML5 lexer

xemlock · Dec 31, 2019 · 956ad0f · 956ad0f
1 parent 92cce65
commit 956ad0f
Show file tree

Hide file tree

Showing 3 changed files with 29 additions and 16 deletions.
diff --git a/composer.json b/composer.json
@@ -1,7 +1,7 @@
 {
     "name": "xemlock/htmlpurifier-html5",
     "type": "library",
-    "description": "HTML5 element definitions for HTML Purifier",
+    "description": "HTML5 support for HTML Purifier",
     "keywords": [
         "HtmlPurifier",
         "html",
@@ -36,6 +36,6 @@
         "test": "phpunit --coverage-html build/coverage"
     },
     "suggest": {
-        "masterminds/html5": "Required to use HTMLPurifier_Lexer_HTML5."
+        "masterminds/html5": "Required to use HTMLPurifier_Lexer_HTML5"
     }
 }
diff --git a/library/HTMLPurifier/Lexer/HTML5.php b/library/HTMLPurifier/Lexer/HTML5.php
@@ -1,14 +1,24 @@
 <?php
 
 /**
- * Class HTMLPurifier_Lexer_HTML5
- *
- * Experimental HTML5-based parser using masterminds/html5 library.
+ * Experimental HTML5-compliant parser using masterminds/html5 library.
  */
 class HTMLPurifier_Lexer_HTML5 extends HTMLPurifier_Lexer_DOMLex
 {
     /**
-     * Lexes an HTML string into tokens.
+     * @throws HTMLPurifier_Exception
+     * @codeCoverageIgnore
+     */
+    public function __construct()
+    {
+        if (!class_exists('\Masterminds\HTML5')) {
+            throw new HTMLPurifier_Exception('Cannot instantiate HTML5 lexer. \Masterminds\HTML5 class is not available');
+        }
+        parent::__construct();
+    }
+
+    /**
+     * Transforms an HTML string into tokens.
      *
      * @param  string                $html
      * @param  HTMLPurifier_Config   $config
@@ -20,16 +30,15 @@ public function tokenizeHTML($html, $config, $context)
         $html = $this->normalize($html, $config, $context);
         $html = $this->armor($html, $config);
 
-        // preprocess html. masterminds/html5 requires <html>, <head> and <body> tags.
-        // <meta charset> is also essential for utf-8
+        // masterminds/html5 requires <html>, <head> and <body> tags
         $html = $this->wrapHTML($html, $config, $context, false);
 
         // Parse the document. $dom is a DOMDocument.
         $html5 = new \Masterminds\HTML5(array('disable_html_ns' => true));
         $doc = $html5->loadHTML($html);
 
-        $body = $doc->getElementsByTagName('html')->item(0) // <html>
-                    ->getElementsByTagName('body')->item(0);            // <body>
+        $body = $doc->getElementsByTagName('html')->item(0)  // <html>
+                    ->getElementsByTagName('body')->item(0); // <body>
 
         $tokens = array();
         $this->tokenizeDOM($body, $tokens, $config);
@@ -45,7 +54,7 @@ public function tokenizeHTML($html, $config, $context)
      * @param  HTMLPurifier_Config  $config
      * @return string
      */
-    protected function armor($html, $config)
+    protected function armor($html, HTMLPurifier_Config $config)
     {
         if ($config->get('Core.AggressivelyFixLt')) {
             $char = '[^a-z!\/]';

diff --git a/tests/HTMLPurifier/Lexer/HTML5Test.php b/tests/HTMLPurifier/Lexer/HTML5Test.php
@@ -1,5 +1,8 @@
 <?php
 
+/** @noinspection PhpDocMissingThrowsInspection */
+/** @noinspection PhpUnhandledExceptionInspection */
+
 class HTMLPurifier_Lexer_HTML5Test extends BaseTestCase
 {
     /**
@@ -12,11 +15,11 @@ class HTMLPurifier_Lexer_HTML5Test extends BaseTestCase
      *
      * @return void
      */
-    public function setUp()
+    protected function setUp()
     {
         parent::setUp();
 
-        $this->context = new HTMLPurifier_Context;
+        $this->context = new HTMLPurifier_Context();
     }
 
     /**
@@ -27,7 +30,7 @@ public function setUp()
      */
     public function test_create_objectLexerImpl()
     {
-        $this->config->set('Core.LexerImpl', new HTMLPurifier_Lexer_HTML5);
+        $this->config->set('Core.LexerImpl', new HTMLPurifier_Lexer_HTML5());
 
         $lexer = HTMLPurifier_Lexer::create($this->config);
 
@@ -560,7 +563,8 @@ public function test_tokenizeHTML_()
     public function test_tokenizeHTML_ignoreIECondComment()
     {
         $this->assertTokenization(
-            '<!--[if IE]>foo<a>bar<!-- baz --><![endif]-->',
+            // typecast disables 'Bad character' inspection in PHPStorm
+            '<!--[if IE]>foo<a>bar' . (string) '<!-- baz --><![endif]-->',
             array()
         );
     }
@@ -678,7 +682,7 @@ public function test_tokenizeHTML_prematureDivClose()
      */
     protected function assertTokenization($input, $expect)
     {
-        $lexer = new HTMLPurifier_Lexer_HTML5;
+        $lexer = new HTMLPurifier_Lexer_HTML5();
 
         $result = $lexer->tokenizeHTML($input, $this->config, $this->context);