LilHTML

It's lil html parser with ample opportunities for optimization.

It uses libxml2 to parse an HTML string into a pure Swift tree, rather than just wrapping a libxml2 tree.

When you call Parser.parse with some HTML, you get back a mutable tree consisting of MutableElementNodes and MutableTextNodes. These are reference types that let you manipulate the tree. If you want value semantics you can call immutableCopy() on the tree to get it as structs.

Usage

let parser = Parser(html: """
<div>
  <div class="container">
    <h1>Hi it's lil tidy</h1>
    <article>
      <p class="one two">hello</p>
      <p>world</p>
    </article>
    <footer>
      <p>it's the footer</p>
    </footer>
  </div>
</div>
""")

// Get an ElementNode
let parsed = try parser.parse().get()

// Search for tags
let paragraphs = parsed.find(.p)
XCTAssertEqual(3, paragraphs.count) // Passes!

// Search for tags inside tags
let footerParagraphs = parsed.find(.p, in: [.footer])
XCTAssertEqual(1, footerParagraphs.count) // Passes!

// Search by attribute
let firstParagraph = parsed.find(attributes: ["class": .contains("one")])
XCTAssertEqual(1, firstParagraph.count) // Passes!