Next release

.github/workflows/ci.yml

@@ -12,5 +12,3 @@ jobs:
     uses: innmind/github-workflows/.github/workflows/psalm-matrix.yml@main
   cs:
     uses: innmind/github-workflows/.github/workflows/cs.yml@main
-    with:
-      php-version: '8.2'

.github/workflows/extensive.yml

@@ -0,0 +1,12 @@
+name: Extensive CI
+
+on:
+  push:
+    tags:
+      - '*'
+    paths:
+      - '.github/workflows/extensive.yml'
+
+jobs:
+  blackbox:
+    uses: innmind/github-workflows/.github/workflows/extensive.yml@main

CHANGELOG.md

@@ -1,5 +1,12 @@
 # Changelog
 
+## [Unreleased]
+
+### Changed
+
+- Requires PHP `8.4`
+- Requires `innmind/xml:~9.0`
+
 ## 7.0.0 - 2025-06-07
 
 ### Added

blackbox.php

@@ -10,6 +10,10 @@
 };
 
 Application::new($argv)
+    ->when(
+        \getenv('BLACKBOX_SET_SIZE') !== false,
+        static fn(Application $app) => $app->scenariiPerProof((int) \getenv('BLACKBOX_SET_SIZE')),
+    )
     ->when(
         \getenv('ENABLE_COVERAGE') !== false,
         static fn(Application $app) => $app

composer.json

@@ -15,13 +15,12 @@
         "issues": "http://github.com/Innmind/Html/issues"
     },
     "require": {
-        "php": "~8.2",
-        "innmind/xml": "~8.0",
-        "innmind/filesystem": "~8.1",
-        "innmind/immutable": "~5.16",
-        "innmind/validation": "~2.0",
-        "symfony/dom-crawler": ">=6.3 <7.0.7",
-        "innmind/url": "~4.0"
+        "php": "~8.4",
+        "innmind/xml": "~9.0",
+        "innmind/filesystem": "~9.0",
+        "innmind/immutable": "~6.0",
+        "innmind/validation": "~3.0",
+        "innmind/url": "~5.0"
     },
     "autoload": {
         "psr-4": {
@@ -35,7 +34,7 @@
     },
     "require-dev": {
         "innmind/black-box": "^6.4.1",
-        "innmind/static-analysis": "^1.2.1",
+        "innmind/static-analysis": "~1.3",
         "innmind/coding-standard": "~2.0"
     }
 }

src/Reader.php

@@ -10,7 +10,6 @@
 };
 use Innmind\Filesystem\File\Content;
 use Innmind\Immutable\Attempt;
-use Symfony\Component\DomCrawler\Crawler;
 
 /**
  * @psalm-immutable
@@ -26,20 +25,20 @@ private function __construct(private Translator $translate)
      */
     public function __invoke(Content $html): Attempt
     {
-        /** @psalm-suppress ImpureMethodCall */
-        $firstNode = (new Crawler($html->toString(), useHtml5Parser: false))->getNode(0);
+        $content = $html->toString();
 
-        if (!$firstNode instanceof \DOMNode) {
-            /** @var Attempt<Document|Node|Element|Custom> */
-            return Attempt::error(new \RuntimeException('Failed to parse html content'));
+        if ($content === '') {
+            return Attempt::error(new \RuntimeException('Empty content'));
         }
 
-        /** @psalm-suppress RedundantCondition */
-        while ($firstNode->parentNode instanceof \DOMNode) {
-            $firstNode = $firstNode->parentNode;
+        try {
+            return ($this->translate)(\Dom\HTMLDocument::createFromString(
+                $content,
+                \LIBXML_HTML_NOIMPLIED | \LIBXML_NOERROR,
+            ));
+        } catch (\Throwable $e) {
+            return Attempt::error($e);
         }
-
-        return ($this->translate)($firstNode);
     }
 
     public static function new(): self

src/Translator.php

@@ -37,11 +37,9 @@ private function __construct(
     }
 
     /**
-     * @psalm-suppress UndefinedClass Since the package still supports PHP 8.2
-     *
      * @return Attempt<Document|Element|Custom|Node>
      */
-    public function __invoke(\DOMNode|\Dom\Node $node): Attempt
+    public function __invoke(\Dom\Node $node): Attempt
     {
         return $this
             ->buildDocument($node)
@@ -59,11 +57,9 @@ public static function new(): self
     }
 
     /**
-     * @psalm-suppress UndefinedClass Since the package still supports PHP 8.2
-     *
      * @return Attempt<Element|Custom|Node>
      */
-    private function child(\DOMNode|\Dom\Node $node): Attempt
+    private function child(\Dom\Node $node): Attempt
     {
         /** @var Attempt<Element|Custom|Node> Psalm doesn't understand the filter */
         return ($this->translate)($node)
@@ -74,39 +70,27 @@ private function child(\DOMNode|\Dom\Node $node): Attempt
                     ->or(Instance::of(Node::class)),
                 static fn() => new \RuntimeException('Invalid document node'),
             )
-            ->match(
-                Attempt::result(...),
-                Attempt::error(...),
-            );
+            ->attempt(static fn($e) => $e);
     }
 
     /**
-     * @psalm-suppress UndefinedClass Since the package still supports PHP 8.2
      * @psalm-suppress MixedArgument
      * @psalm-suppress MixedMethodCall
      * @psalm-suppress UndefinedPropertyFetch
      *
      * @return Attempt<Document>
      */
-    private function buildDocument(\DOMNode|\Dom\Node $node): Attempt
+    private function buildDocument(\Dom\Node $node): Attempt
     {
         /** @var Sequence<Node|Element|Custom> */
         $children = Sequence::of();
 
         return Maybe::just($node)
-            ->keep(
-                Instance::of(\DOMDocument::class)->or(
-                    Instance::of(\Dom\Document::class),
-                ),
-            )
+            ->keep(Instance::of(\Dom\Document::class))
             ->attempt(static fn() => new \RuntimeException('Not a document'))
             ->flatMap(
                 fn($document) => Sequence::of(...\array_values(\iterator_to_array($document->childNodes)))
-                    ->keep(
-                        Instance::of(\DOMNode::class)->or(
-                            Instance::of(\Dom\Node::class),
-                        ),
-                    )
+                    ->keep(Instance::of(\Dom\Node::class))
                     ->exclude(static fn($child) => $child->nodeType === \XML_DOCUMENT_TYPE_NODE)
                     ->sink($children)
                     ->attempt(
@@ -129,11 +113,10 @@ private function buildDocument(\DOMNode|\Dom\Node $node): Attempt
     /**
      * @psalm-pure
      * @psalm-suppress ImpurePropertyFetch
-     * @psalm-suppress UndefinedClass Since the package still supports PHP 8.2
      *
      * @return Maybe<Type>
      */
-    private static function buildDoctype(\DOMDocumentType|\Dom\DocumentType $type): Maybe
+    private static function buildDoctype(\Dom\DocumentType $type): Maybe
     {
         /** @psalm-suppress MixedArgument */
         return Type::maybe(

tests/DocumentTest.php

@@ -85,8 +85,8 @@ public function testPrependChild()
         $this->assertInstanceOf(Document::class, $document2);
         $this->assertSame($document->type(), $document2->type());
         $this->assertNotSame($document->children(), $document2->children());
-        $this->assertCount(3, $document->children());
-        $this->assertCount(4, $document2->children());
+        $this->assertSame(3, $document->children()->size());
+        $this->assertSame(4, $document2->children()->size());
         $this->assertSame(
             $node,
             $document2->children()->get(0)->match(
@@ -127,8 +127,8 @@ public function testAppendChild()
         $this->assertInstanceOf(Document::class, $document2);
         $this->assertSame($document->type(), $document2->type());
         $this->assertNotSame($document->children(), $document2->children());
-        $this->assertCount(3, $document->children());
-        $this->assertCount(4, $document2->children());
+        $this->assertSame(3, $document->children()->size());
+        $this->assertSame(4, $document2->children()->size());
         $this->assertEquals(
             $document->children()->get(0),
             $document2->children()->get(0),

tests/Element/ScriptTest.php

@@ -26,7 +26,7 @@ public function testInterface()
             '<script>foo</script>'."\n",
             $script->asContent()->toString(),
         );
-        $this->assertCount(1, $script->children());
+        $this->assertSame(1, $script->children()->size());
     }
 
     public function testWithAttributes()

tests/ReaderTest.php

@@ -37,14 +37,17 @@ public function testReadSimple()
             static fn($node) => $node,
             static fn() => null,
         );
+
+        // For some reason the \Dom\HTMLDocument API doesn't respect the
+        // indentation of the input document.
+        $space = '    ';
         $expected = <<<HTML
         <!DOCTYPE html>
-        <html>
-            <head/>
+        <html><head/>
             <body>
                 foo
-            </body>
-        </html>
+        $space
+        </body></html>
         HTML;
 
         $this->assertInstanceOf(Document::class, $node);

tests/Translator/NodeTranslator/ATranslatorTest.php

@@ -14,11 +14,13 @@ class ATranslatorTest extends TestCase
 {
     public function testTranslate()
     {
-        $dom = new \DOMDocument;
-        $dom->loadHTML('<a href="/" class="whatever">foo</a>');
+        $dom = \Dom\HTMLDocument::createFromString(
+            '<a href="/" class="whatever">foo</a>',
+            \LIBXML_HTML_NOIMPLIED | \LIBXML_NOERROR,
+        );
 
         $a = Translator::new()(
-            $dom->childNodes->item(1)->childNodes->item(0)->childNodes->item(0),
+            $dom->childNodes->item(0),
         )->match(
             static fn($a) => $a,
             static fn() => null,
@@ -27,21 +29,23 @@ public function testTranslate()
         $this->assertInstanceOf(A::class, $a);
         $this->assertSame('/', $a->href()->toString());
         $a = $a->normalize();
-        $this->assertCount(2, $a->attributes());
+        $this->assertSame(2, $a->attributes()->size());
         $this->assertSame('whatever', $a->attribute('class')->match(
             static fn($attribute) => $attribute->value(),
             static fn() => null,
         ));
-        $this->assertCount(1, $a->children());
+        $this->assertSame(1, $a->children()->size());
     }
 
     public function testReturnNothingWhenMissingHrefAttribute()
     {
-        $dom = new \DOMDocument;
-        $dom->loadHTML('<a class="whatever">foo</a>');
+        $dom = \Dom\HTMLDocument::createFromString(
+            '<a class="whatever">foo</a>',
+            \LIBXML_HTML_NOIMPLIED | \LIBXML_NOERROR,
+        );
 
         $result = Translator::new()(
-            $dom->childNodes->item(1)->childNodes->item(0)->childNodes->item(0),
+            $dom->childNodes->item(0),
         )
             ->maybe()
             ->keep(Instance::of(A::class));

tests/Translator/NodeTranslator/BaseTranslatorTest.php

@@ -14,11 +14,13 @@ class BaseTranslatorTest extends TestCase
 {
     public function testTranslate()
     {
-        $dom = new \DOMDocument;
-        $dom->loadHTML('<base href="/" target="_blank"/>');
+        $dom = \Dom\HTMLDocument::createFromString(
+            '<base href="/" target="_blank"/>',
+            \LIBXML_HTML_NOIMPLIED | \LIBXML_NOERROR,
+        );
 
         $base = Translator::new()(
-            $dom->childNodes->item(1)->childNodes->item(0)->childNodes->item(0),
+            $dom->childNodes->item(0),
         )->match(
             static fn($base) => $base,
             static fn() => null,
@@ -27,7 +29,7 @@ public function testTranslate()
         $this->assertInstanceOf(Base::class, $base);
         $this->assertSame('/', $base->href()->toString());
         $base = $base->normalize();
-        $this->assertCount(2, $base->attributes());
+        $this->assertSame(2, $base->attributes()->size());
         $this->assertSame('_blank', $base->attribute('target')->match(
             static fn($attribute) => $attribute->value(),
             static fn() => null,
@@ -36,11 +38,13 @@ public function testTranslate()
 
     public function testReturnNothingWhenMissingHrefAttribute()
     {
-        $dom = new \DOMDocument;
-        $dom->loadHTML('<base/>');
+        $dom = \Dom\HTMLDocument::createFromString(
+            '<base/>',
+            \LIBXML_HTML_NOIMPLIED | \LIBXML_NOERROR,
+        );
 
         $result = Translator::new()(
-            $dom->childNodes->item(1)->childNodes->item(0)->childNodes->item(0),
+            $dom->childNodes->item(0),
         )
             ->maybe()
             ->keep(Instance::of(Base::class));

tests/Translator/NodeTranslator/DocumentTranslatorTest.php

@@ -13,8 +13,7 @@ class DocumentTranslatorTest extends TestCase
 {
     public function testTranslate()
     {
-        $document = new \DOMDocument;
-        $document->loadHtml('<!DOCTYPE html><body></body>');
+        $document = \Dom\HTMLDocument::createFromString('<!DOCTYPE html><body></body>');
 
         $node = Translator::new()(
             $document,
@@ -25,11 +24,12 @@ public function testTranslate()
 
         $this->assertInstanceOf(Document::class, $node);
         $this->assertSame('html', $node->type()->name());
-        $this->assertCount(1, $node->children());
+        $this->assertSame(1, $node->children()->size());
         $this->assertSame(
             <<<HTML
             <!DOCTYPE html>
             <html>
+                <head/>
                 <body/>
             </html>
 
@@ -40,8 +40,10 @@ public function testTranslate()
 
     public function testTranslateWithoutDoctype()
     {
-        $document = new \DOMDocument;
-        $document->loadHtml('<!--foo-->');
+        $document = \Dom\HTMLDocument::createFromString(
+            '<!--foo-->',
+            \LIBXML_NOERROR,
+        );
 
         $node = Translator::new()(
             $document,
@@ -51,15 +53,17 @@ public function testTranslateWithoutDoctype()
         );
 
         $this->assertSame(
-            '<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" "http://www.w3.org/TR/REC-html40/loose.dtd">',
+            '<!DOCTYPE html>',
             $node->type()->toString(),
         );
     }
 
     public function testTranslateWithoutChildren()
     {
-        $document = new \DOMDocument;
-        $document->loadHtml('<!DOCTYPE html>');
+        $document = \Dom\HTMLDocument::createFromString(
+            '<!DOCTYPE html>',
+            \LIBXML_HTML_NOIMPLIED | \LIBXML_NOERROR,
+        );
 
         $node = Translator::new()(
             $document,