Utilize latest ext-dom

Author: adamjgriffith

.github/workflows/ci.yml

@@ -43,15 +43,9 @@ 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": "*",

src/DOMBuilder.php

@@ -4,32 +4,27 @@
 
 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
+     * @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,
         );
 

src/Reader/MicrodataReader.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;
 
@@ -36,9 +36,9 @@
 final class MicrodataReader implements Reader
 {
     #[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);
 
         /**
          * An item is a top-level Microdata item if its element does not have an itemprop attribute.
@@ -49,19 +49,19 @@ public function read(DOMDocument $document, string $url): array
         $nodes = iterator_to_array($nodes);
 
         return array_map(
-            fn (DOMNode $node) => $this->nodeToItem($node, $xpath, $url),
+            fn (Node $node) => $this->nodeToItem($node, $xpath, $url),
             $nodes,
         );
     }
 
     /**
-     * Extracts information from a DOMNode into an Item.
+     * Extracts information from a Node into an Item.
      *
-     * @param DOMNode  $node  A DOMNode representing an element with the itemscope attribute.
-     * @param DOMXPath $xpath A DOMXPath object created from the node's document element.
+     * @param Node  $node  A Node representing an element with the itemscope attribute.
+     * @param XPath $xpath A XPath object created from the node's document element.
      * @param string   $url   The URL the document was retrieved from, for relative URL resolution.
      */
-    private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url): Item
+    private function nodeToItem(Node $node, XPath $xpath, string $url): Item
     {
         $itemid = $node->attributes->getNamedItem('itemid');
 
@@ -106,7 +106,7 @@ private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url): Item
 
         // Exclude properties that are inside a nested item; XPath does not seem to provide a way to do this.
         // See: https://stackoverflow.com/q/26365495/759866
-        $itemprops = array_filter($itemprops, function (DOMNode $itemprop) use ($node, $xpath) {
+        $itemprops = array_filter($itemprops, function (Node $itemprop) use ($node, $xpath) {
             for (; ;) {
                 $itemprop = $itemprop->parentNode;
 
@@ -122,7 +122,7 @@ private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url): Item
 
         $vocabularyIdentifier = $this->getVocabularyIdentifier($types);
 
-        /** @var DOMNode[] $itemprops */
+        /** @var Node[] $itemprops */
         foreach ($itemprops as $itemprop) {
             /**
              * An element introducing a property can introduce multiple properties at once, to avoid duplication when
@@ -159,11 +159,11 @@ private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url): Item
     /**
      * @see https://www.w3.org/TR/microdata/#values
      *
-     * @param DOMNode  $node  A DOMNode representing an element with the itemprop attribute.
-     * @param DOMXPath $xpath A DOMXPath object created from the node's document element.
+     * @param Node  $node  A Node representing an element with the itemprop attribute.
+     * @param XPath $xpath A XPath object created from the node's document element.
      * @param string   $url   The URL the document was retrieved from, for relative URL resolution.
      */
-    private function getPropertyValue(DOMNode $node, DOMXPath $xpath, string $url): Item|string
+    private function getPropertyValue(Node $node, XPath $xpath, string $url): Item|string
     {
         /**
          * If the element also has an itemscope attribute: the value is the item created by the element.

src/Reader/RdfaLiteReader.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;
 
@@ -93,9 +93,9 @@ final class RdfaLiteReader implements Reader
     ];
 
     #[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);
 
         /**
          * Top-level item has a typeof attribute and no property attribute.
@@ -104,22 +104,22 @@ public function read(DOMDocument $document, string $url): array
         $nodes = iterator_to_array($nodes);
 
         return array_map(
-            fn (DOMNode $node) => $this->nodeToItem($node, $xpath, $url, self::PREDEFINED_PREFIXES, null),
+            fn (Node $node) => $this->nodeToItem($node, $xpath, $url, self::PREDEFINED_PREFIXES, null),
             $nodes,
         );
     }
 
     /**
-     * Extracts information from a DOMNode into an Item.
+     * Extracts information from a Node into an Item.
      *
-     * @param DOMNode     $node       A DOMNode representing an element with the typeof attribute.
-     * @param DOMXPath    $xpath      A DOMXPath object created from the node's document element.
+     * @param Node        $node       A Node representing an element with the typeof attribute.
+     * @param XPath       $xpath      A XPath object created from the node's document element.
      * @param string      $url        The URL the document was retrieved from, for relative URL resolution.
      * @param string[]    $prefixes   The prefixes in use, as a map of prefix to vocabulary URL.
      * @param string|null $vocabulary The URL of the vocabulary in use, if any.
      *                                This is the content of the vocab attribute of the closest item ancestor.
      */
-    private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url, array $prefixes, ?string $vocabulary): Item
+    private function nodeToItem(Node $node, XPath $xpath, string $url, array $prefixes, ?string $vocabulary): Item
     {
         $vocabulary = $this->updateVocabulary($node, $vocabulary);
 
@@ -165,7 +165,7 @@ private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url, array $
 
         // Exclude properties that are inside a nested item; XPath does not seem to provide a way to do this.
         // See: https://stackoverflow.com/q/26365495/759866
-        $properties = array_filter($properties, function (DOMNode $itemprop) use ($node, $xpath) {
+        $properties = array_filter($properties, function (Node $itemprop) use ($node, $xpath) {
             for (; ;) {
                 $itemprop = $itemprop->parentNode;
 
@@ -182,7 +182,7 @@ private function nodeToItem(DOMNode $node, DOMXPath $xpath, string $url, array $
             return false;
         });
 
-        /** @var DOMNode[] $properties */
+        /** @var Node[] $properties */
         foreach ($properties as $property) {
             $names = $property->attributes->getNamedItem('property')->textContent;
 
@@ -261,12 +261,12 @@ private function isValidAbsoluteURL(string $url): bool
     /**
      * Replaces the current vocabulary with the one from the vocab attribute of the current node, if set.
      *
-     * @param DOMNode     $node       The DOMNode that may contain a vocab attribute.
+     * @param Node        $node       The Node that may contain a vocab attribute.
      * @param string|null $vocabulary The URL of the vocabulary in use, if any.
      *
      * @return string|null The updated vocabulary URL, if any.
      */
-    private function updateVocabulary(DOMNode $node, ?string $vocabulary): ?string
+    private function updateVocabulary(Node $node, ?string $vocabulary): ?string
     {
         $vocab = $node->attributes->getNamedItem('vocab');
 
@@ -310,13 +310,13 @@ private function checkVocabularyUrl(string $url): ?string
     /**
      * @see https://www.w3.org/TR/microdata/#values
      *
-     * @param DOMNode     $node       A DOMNode representing an element with the property attribute.
-     * @param DOMXPath    $xpath      A DOMXPath object created from the node's document element.
+     * @param Node        $node       A Node representing an element with the property attribute.
+     * @param XPath       $xpath      A XPath object created from the node's document element.
      * @param string      $url        The URL the document was retrieved from, for relative URL resolution.
      * @param string[]    $prefixes   The prefixes in use, as a map of prefix to vocabulary URL.
      * @param string|null $vocabulary The URL of the vocabulary in use, if any.
      */
-    private function getPropertyValue(DOMNode $node, DOMXPath $xpath, string $url, array $prefixes, ?string $vocabulary): Item|string
+    private function getPropertyValue(Node $node, XPath $xpath, string $url, array $prefixes, ?string $vocabulary): Item|string
     {
         // If the element also has an typeof attribute, create an item from the element
         $attr = $node->attributes->getNamedItem('typeof');

src/Reader/ReaderChain.php

@@ -5,7 +5,7 @@
 namespace Brick\StructuredData\Reader;
 
 use Brick\StructuredData\Reader;
-use DOMDocument;
+use DOM\HTMLDocument;
 use Override;
 
 use function array_merge;
@@ -29,7 +29,7 @@ public function __construct(Reader ...$readers)
     }
 
     #[Override]
-    public function read(DOMDocument $document, string $url): array
+    public function read(HTMLDocument $document, string $url): array
     {
         if (! $this->readers) {
             return [];