[abo3losh1]

WrapAllHtml parses the wrapper HTML with the first selected node itself as the context, while the wrapper actually takes that node's place in its parent. WrapHtml already uses the parent.

For most elements that difference is invisible, but not for elements whose content the parser reads as raw text:

doc, _ := goquery.NewDocumentFromReader(strings.NewReader(`<body><textarea>hi</textarea></body>`))
doc.Find("textarea").WrapAllHtml(`<div class="wrap"></div>`)
// body: &lt;div class=&#34;wrap&#34;&gt;&lt;/div&gt;

The wrapper is parsed as text inside a <textarea> context, so it becomes a text node, WrapAllNode moves the textarea into that text node, and the textarea is gone from the document. The same happens with iframe, script, style, title and xmp. WrapHtml on the same input gives <div class="wrap"><textarea>hi</textarea></div>.

If the first node is a text or comment node it panics instead:

doc.Find("p").Contents().WrapAllHtml(`<span></span>`)
// panic: goquery: failed to parse HTML: html: ParseFragment of non-element Node

The fix parses in the parent's context, like WrapHtml. When there is no parent, or the parent is not an element (the document node above <html>), it falls back to parseHtml as before.

Added TestWrapAllHtmlRawTextElement and TestWrapAllHtmlTextNode. The first finds 0 nodes and the second panics before the fix. go test ./... and go vet ./... pass.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.