Remove deprecated usage of function (#310)

* remove deprecated usage of function * remove iconv extension, add comments and tests * Remove php 8.1 (#308) * Fix tests * Drop composer dev flag * Support css-selector 6 * Drop php 8.1 * Update changelog * refactor tests * remove deprecated usage of function * remove iconv extension, add comments and tests * refactor tests * fix code style --------- Co-authored-by: Ivan Shcherbak <[email protected]>
xparse · Dec 19, 2024 · bfdb9e6 · bfdb9e6
1 parent d7342ef
commit bfdb9e6
Show file tree

Hide file tree

Showing 2 changed files with 40 additions and 1 deletion.
diff --git a/src/ElementFinder.php b/src/ElementFinder.php
@@ -214,7 +214,19 @@ private function setData(string $data): self
 
         if (static::DOCUMENT_HTML === $this->type) {
             $data = StringHelper::safeEncodeStr($data);
-            $data = mb_convert_encoding($data, 'HTML-ENTITIES', 'UTF-8');
+
+            //Analogue of mb_convert_encoding($data, 'HTML-ENTITIES', 'UTF-8')
+            //Usage of mb_convert_encoding with encoding to HTML_ENTITIES is deprecated since php version 8.2
+            //When passing data to ElementFinder in an encoding other than UTF-8, any unrecognized characters will be ignored
+            $data = mb_encode_numericentity(
+                htmlspecialchars_decode(
+                    htmlentities($data, ENT_NOQUOTES | ENT_IGNORE, 'UTF-8', false),
+                    ENT_NOQUOTES
+                ),
+                [0x80, 0x10FFFF, 0, ~0],
+                'UTF-8'
+            );
+
             $this->dom->loadHTML($data, LIBXML_NOCDATA & LIBXML_NOERROR);
         } elseif (static::DOCUMENT_XML === $this->type) {
             $this->dom->loadXML($data, LIBXML_NOCDATA & LIBXML_NOERROR);

diff --git a/tests/ElementFinderTest.php b/tests/ElementFinderTest.php
@@ -5,6 +5,7 @@
 namespace Test\Xparse\ElementFinder;
 
 use InvalidArgumentException;
+use PHPUnit\Framework\Attributes\DataProvider;
 use PHPUnit\Framework\TestCase;
 use RuntimeException;
 use Test\Xparse\ElementFinder\Dummy\ItemsByClassExpressionTranslator;
@@ -519,4 +520,30 @@ private function getValidXml(): string
       </breakfast_menu>
       ';
     }
+
+    /**
+     * @return string[][]
+     */
+    public static function getDifferentEncodingsSupportDataProvider(): array
+    {
+        return [
+            [
+                '<body>Текст текст text</body>',
+                'Текст текст text',
+            ],
+            [
+                '<body>&#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295; &#4294967295;&#4294967295;&#4294967295;&#4294967295;&#4294967295; text</body>',
+                '  text',
+            ],
+        ];
+    }
+
+    #[DataProvider('getDifferentEncodingsSupportDataProvider')]
+    public function testDifferentEncodingsSupport(string $html, string $bodyText): void
+    {
+        $page = new ElementFinder($html);
+        $pageBodyText = $page->content('//body')->first();
+        self::assertInstanceOf(ElementFinder::class, $page);
+        self::assertEquals($bodyText, $pageBodyText);
+    }
 }