Add HtmlFragment class for improved HTML manipulation in HtmlDiff

[thijskok]

No description provided.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 96.80851% with 3 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/Renderers/Support/HtmlFragment.php	96.10%	3 Missing ⚠️

📢 Thoughts on this report? Let us know!

[Copilot]

Pull request overview

This PR introduces a DOM-based HtmlFragment helper to make HtmlDiff’s before/after view extraction more robust (especially around removing fully-added/fully-deleted nested elements), and adds regression tests for nested list-item cases.

Changes:

• Added HtmlFragment (DOMDocument + XPath wrapper) for targeted, in-place HTML mutations.
• Updated HtmlDiff to use HtmlFragment instead of regex-based post-processing when building before/after views.
• Added tests to ensure nested list items that become empty after diff-marker stripping are omitted rather than left as empty structural tags.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
tests/HtmlDiffRendererTest.php	Adds regression coverage for nested list-item omission behavior in before/after views.
src/Renderers/Support/HtmlFragment.php	Introduces a DOM/XPath-based mutation utility for HTML snippets used by HtmlDiff.
src/Renderers/HtmlDiff.php	Switches view-extraction logic from regexes to HtmlFragment transformations.

Comments suppressed due to low confidence (2)

src/Renderers/Support/HtmlFragment.php:40

• $this->xpath->query(...)->item(0) can return null if parsing fails, which will trigger a TypeError when assigning to the typed DOMElement property. Consider validating the lookup and throwing a clear exception so failures are diagnosable.

        $this->xpath = new DOMXPath($this->dom);
        $this->root = $this->xpath->query('//*[@id="__root__"]')->item(0);
    }

src/Renderers/Support/HtmlFragment.php:162

• toHtml() appends an empty text node to every childless element to force a closing tag. This will also target void elements like
  , , , etc., which must not have children and can serialize into invalid HTML (e.g.,
  
  ) or unexpected output.

        // Childless elements would otherwise have their closing tag silently dropped by
        // libxml's HTML serializer (e.g. <li></li> would come back out as just <li>); a
        // placeholder text node forces it to always emit both tags.
        foreach (iterator_to_array($this->xpath->query('.//*[not(node())]', $this->root)) as $element) {
            $element->appendChild($this->dom->createTextNode(''));

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.