Utilize latest ext-dom

.github/workflows/ci.yml

@@ -33,6 +33,7 @@ jobs:
     name: Coding Standard
     uses: brick/coding-standard/.github/workflows/coding-standard.yml@v2
     with:
+      php-version: 8.4
       working-directory: "tools/ecs"
 
   phpunit:
@@ -43,15 +44,9 @@ jobs:
       fail-fast: false
       matrix:
         php-version:
-          - "8.1"
-          - "8.2"
-          - "8.3"
           - "8.4"
         deps:
           - "highest"
-        include:
-          - php-version: "8.1"
-            deps: "lowest"
 
     steps:
       - name: Checkout

README.md

@@ -23,7 +23,7 @@ composer require brick/structured-data
 
 ### Requirements
 
-This library requires PHP 8.1 or later. It makes use of the following extensions:
+This library requires PHP 8.4 or later. It makes use of the following extensions:
 
 - [dom](https://www.php.net/manual/en/book.dom.php)
 - [json](https://www.php.net/manual/en/book.json.php)
@@ -55,13 +55,13 @@ interface Brick\StructuredData\Reader
     /**
      * Reads the items contained in the given document.
      *
-     * @param DOMDocument $document The DOM document to read.
-     * @param string      $url      The URL the document was retrieved from. This will be used only to resolve relative
-     *                              URLs in property values. No attempt will be performed to connect to this URL.
+     * @param HTMLDocument $document The DOM document to read.
+     * @param string       $url      The URL the document was retrieved from. This will be used only to resolve relative
+     *                               URLs in property values. No attempt will be performed to connect to this URL.
      *
      * @return Item[] The top-level items.
      */
-    public function read(DOMDocument $document, string $url) : array;
+    public function read(HTMLDocument $document, string $url) : array;
 }
 ```
 
@@ -93,7 +93,7 @@ use Brick\StructuredData\Item;
 $microdataReader = new MicrodataReader();
 
 // Wrap into HTMLReader to be able to read HTML strings or files directly,
-// i.e. without manually converting them to DOMDocument instances first
+// i.e. without manually converting them to HTMLDocument instances first
 $htmlReader = new HTMLReader($microdataReader);
 
 // Replace this URL with that of a website you know is using Microdata

composer.json

@@ -11,7 +11,7 @@
   ],
   "license": "MIT",
   "require": {
-    "php": "^8.1",
+    "php": "^8.4",
     "ext-dom": "*",
     "ext-json": "*",
     "ext-libxml": "*",

psalm-baseline.xml

@@ -14,13 +14,14 @@
     </MissingClosureParamType>
     <MixedArgument>
       <code><![CDATA[$name]]></code>
+      <code><![CDATA[$node->textContent]]></code>
     </MixedArgument>
     <MixedArrayAccess>
       <code><![CDATA[$name[0]]]></code>
     </MixedArrayAccess>
-    <PossiblyInvalidArgument>
-      <code><![CDATA[fn (DOMNode $node) => $this->readJson($node->textContent, $url)]]></code>
-    </PossiblyInvalidArgument>
+    <NamedArgumentNotAllowed>
+      <code><![CDATA[$items]]></code>
+    </NamedArgumentNotAllowed>
     <RawObjectIteration>
       <code><![CDATA[$item]]></code>
     </RawObjectIteration>
@@ -29,106 +30,100 @@
     </RedundantConditionGivenDocblockType>
   </file>
   <file src="src/Reader/MicrodataReader.php">
-    <InvalidArgument>
-      <code><![CDATA[function (DOMNode $itemprop) use ($node, $xpath) {
-            for (; ;) {
-                $itemprop = $itemprop->parentNode;
-
-                if ($itemprop->isSameNode($node)) {
-                    return true;
-                }
-
-                if ($itemprop->attributes->getNamedItem('itemscope')) {
-                    return false;
-                }
-            }
-        }]]></code>
-    </InvalidArgument>
-    <PossiblyInvalidArgument>
-      <code><![CDATA[fn (DOMNode $node) => $this->nodeToItem($node, $xpath, $url)]]></code>
-      <code><![CDATA[function (DOMNode $itemprop) use ($node, $xpath) {
-            for (; ;) {
-                $itemprop = $itemprop->parentNode;
-
-                if ($itemprop->isSameNode($node)) {
-                    return true;
-                }
-
-                if ($itemprop->attributes->getNamedItem('itemscope')) {
-                    return false;
-                }
-            }
-        }]]></code>
-    </PossiblyInvalidArgument>
-    <PossiblyNullArgument>
+    <MixedArgument>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$itemid->textContent]]></code>
+      <code><![CDATA[$itemtype->textContent]]></code>
       <code><![CDATA[$names]]></code>
-      <code><![CDATA[preg_replace('/\s+/', ' ', $node->textContent)]]></code>
-    </PossiblyNullArgument>
-    <PossiblyNullPropertyFetch>
-      <code><![CDATA[$itemprop->attributes->getNamedItem('itemprop')->textContent]]></code>
-    </PossiblyNullPropertyFetch>
-    <PossiblyNullReference>
+      <code><![CDATA[$node->textContent]]></code>
+    </MixedArgument>
+    <MixedMethodCall>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[isSameNode]]></code>
-    </PossiblyNullReference>
-  </file>
-  <file src="src/Reader/RdfaLiteReader.php">
-    <InvalidArgument>
-      <code><![CDATA[function (DOMNode $itemprop) use ($node, $xpath) {
-            for (; ;) {
-                $itemprop = $itemprop->parentNode;
-
-                if ($itemprop->isSameNode($node)) {
-                    return true;
-                }
-
-                if ($itemprop->attributes->getNamedItem('typeof')) {
-                    return false;
-                }
-            }
-
-            // Unreachable, but makes static analysis happy
-            return false;
-        }]]></code>
-    </InvalidArgument>
+    </MixedMethodCall>
+    <MixedPropertyFetch>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$itemid->textContent]]></code>
+      <code><![CDATA[$itemprop->attributes]]></code>
+      <code><![CDATA[$itemprop->attributes->getNamedItem('itemprop')->textContent]]></code>
+      <code><![CDATA[$itemprop->parentNode]]></code>
+      <code><![CDATA[$itemtype->textContent]]></code>
+    </MixedPropertyFetch>
+    <MixedReturnStatement>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+    </MixedReturnStatement>
     <PossiblyInvalidArgument>
-      <code><![CDATA[fn (DOMNode $node) => $this->nodeToItem($node, $xpath, $url, self::PREDEFINED_PREFIXES, null)]]></code>
-      <code><![CDATA[function (DOMNode $itemprop) use ($node, $xpath) {
-            for (; ;) {
-                $itemprop = $itemprop->parentNode;
-
-                if ($itemprop->isSameNode($node)) {
-                    return true;
-                }
-
-                if ($itemprop->attributes->getNamedItem('typeof')) {
-                    return false;
-                }
-            }
-
-            // Unreachable, but makes static analysis happy
-            return false;
-        }]]></code>
+      <code><![CDATA[preg_replace('/\s+/', ' ', $node->textContent)]]></code>
     </PossiblyInvalidArgument>
-    <PossiblyNullArgument>
+    <UndefinedPropertyFetch>
+      <code><![CDATA[$itemprop->attributes]]></code>
+      <code><![CDATA[$node->attributes]]></code>
+      <code><![CDATA[$node->attributes]]></code>
+    </UndefinedPropertyFetch>
+  </file>
+  <file src="src/Reader/RdfaLiteReader.php">
+    <MixedArgument>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
       <code><![CDATA[$names]]></code>
+      <code><![CDATA[$node->textContent]]></code>
+      <code><![CDATA[$resource->textContent]]></code>
       <code><![CDATA[$typeof->textContent]]></code>
-      <code><![CDATA[preg_replace('/\s+/', ' ', $node->textContent)]]></code>
-    </PossiblyNullArgument>
-    <PossiblyNullPropertyFetch>
-      <code><![CDATA[$property->attributes->getNamedItem('property')->textContent]]></code>
-      <code><![CDATA[$typeof->textContent]]></code>
-    </PossiblyNullPropertyFetch>
-    <PossiblyNullReference>
+      <code><![CDATA[$vocab->textContent]]></code>
+    </MixedArgument>
+    <MixedMethodCall>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
+      <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[getNamedItem]]></code>
       <code><![CDATA[isSameNode]]></code>
-    </PossiblyNullReference>
+    </MixedMethodCall>
+    <MixedPropertyFetch>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$attr->textContent]]></code>
+      <code><![CDATA[$itemprop->attributes]]></code>
+      <code><![CDATA[$itemprop->parentNode]]></code>
+      <code><![CDATA[$property->attributes->getNamedItem('property')->textContent]]></code>
+      <code><![CDATA[$resource->textContent]]></code>
+      <code><![CDATA[$typeof->textContent]]></code>
+      <code><![CDATA[$vocab->textContent]]></code>
+    </MixedPropertyFetch>
+    <MixedReturnStatement>
+      <code><![CDATA[$attr->textContent]]></code>
+    </MixedReturnStatement>
+    <PossiblyInvalidArgument>
+      <code><![CDATA[preg_replace('/\s+/', ' ', $node->textContent)]]></code>
+    </PossiblyInvalidArgument>
+    <UndefinedPropertyFetch>
+      <code><![CDATA[$node->attributes]]></code>
+      <code><![CDATA[$node->attributes]]></code>
+      <code><![CDATA[$node->attributes]]></code>
+      <code><![CDATA[$property->attributes]]></code>
+    </UndefinedPropertyFetch>
   </file>
 </files>

src/DOMBuilder.php

@@ -4,32 +4,26 @@
 
 namespace Brick\StructuredData;
 
-use DOMDocument;
+use Dom\HTMLDocument;
 
+use const Dom\HTML_NO_DEFAULT_NS;
 use const LIBXML_NOERROR;
-use const LIBXML_NOWARNING;
 
 final class DOMBuilder
 {
     /**
-     * Builds a DOMDocument from an HTML string.
+     * Builds a HTMLDocument from an HTML string.
      */
-    public static function fromHTML(string $html): DOMDocument
+    public static function fromHTML(string $html): HTMLDocument
     {
-        $document = new DOMDocument();
-        $document->loadHTML($html, LIBXML_NOWARNING | LIBXML_NOERROR);
-
-        return $document;
+        return HTMLDocument::createFromString($html, LIBXML_NOERROR | HTML_NO_DEFAULT_NS);
     }
 
     /**
-     * Builds a DOMDocument from an HTML file.
+     * Builds a HTMLDocument from an HTML file.
      */
-    public static function fromHTMLFile(string $file): DOMDocument
+    public static function fromHTMLFile(string $file): HTMLDocument
     {
-        $document = new DOMDocument();
-        $document->loadHTMLFile($file, LIBXML_NOWARNING | LIBXML_NOERROR);
-
-        return $document;
+        return HTMLDocument::createFromFile($file, LIBXML_NOERROR | HTML_NO_DEFAULT_NS);
     }
 }

src/Reader.php

@@ -4,7 +4,7 @@
 
 namespace Brick\StructuredData;
 
-use DOMDocument;
+use Dom\HTMLDocument;
 
 /**
  * Common interface for readers of each format: Microdata, RDFa Lite, JSON-LD.
@@ -14,11 +14,11 @@ interface Reader
     /**
      * Reads the items contained in the given document.
      *
-     * @param DOMDocument $document The DOM document to read.
-     * @param string      $url      The URL the document was retrieved from. This will be used only to resolve relative
-     *                              URLs in property values. The implementation must not attempt to connect to this URL.
+     * @param HTMLDocument $document The HTMLDocument to read.
+     * @param string       $url      The URL the document was retrieved from. This will be used only to resolve relative
+     *                               URLs in property values. The implementation must not attempt to connect to this URL.
      *
      * @return Item[] The top-level items.
      */
-    public function read(DOMDocument $document, string $url): array;
+    public function read(HTMLDocument $document, string $url): array;
 }

src/Reader/JsonLdReader.php

@@ -6,9 +6,9 @@
 
 use Brick\StructuredData\Item;
 use Brick\StructuredData\Reader;
-use DOMDocument;
-use DOMNode;
-use DOMXPath;
+use Dom\HTMLDocument;
+use Dom\Node;
+use Dom\XPath;
 use Override;
 use Sabre\Uri\InvalidUriException;
 use stdClass;
@@ -66,9 +66,9 @@ public function __construct(array $iriProperties = [])
     }
 
     #[Override]
-    public function read(DOMDocument $document, string $url): array
+    public function read(HTMLDocument $document, string $url): array
     {
-        $xpath = new DOMXPath($document);
+        $xpath = new XPath($document);
 
         $nodes = $xpath->query('//script[@type="application/ld+json"]');
         $nodes = iterator_to_array($nodes);
@@ -78,7 +78,7 @@ public function read(DOMDocument $document, string $url): array
         }
 
         $items = array_map(
-            fn (DOMNode $node) => $this->readJson($node->textContent, $url),
+            fn (Node $node) => $this->readJson($node->textContent, $url),
             $nodes,
         );