Trim a document to a selector¶

Keep only the part of a parsed document you care about with prune(), the parse-then-keep pattern BeautifulSoup spelled with a SoupStrainer.

prune() keeps only the descendants matching a CSS selector, plus their ancestors up to the node it is called on and the whole subtree under each match, and removes everything else in place. This is the parse-then-keep pattern BeautifulSoup spelled with a SoupStrainer: parse the whole document the WHATWG way, then shrink it to the part you care about. It returns the node, so it chains off parse():

markup = "<body><nav>skip</nav><main><article>keep<b>me</b></article><aside>drop</aside></main></body>"
doc = turbohtml.parse(markup).prune("article")
print(doc.select_one("body").serialize())

<body><main><article>keep<b>me</b></article></main></body>

The ancestors of a match stay so the match keeps its place in the tree, and the match’s own subtree stays whole, so the <b> and the text survive while the unrelated <nav> and <aside> go. A selector that matches nothing empties the subtree.