pub fn is_parse_artifact(node: &Node) -> boolExpand description
Is node an artifact of HOW a chunk was parsed, rather than part of the
chunk? True for exactly the two things that can sit ABOVE a chunk’s top-level
nodes: the throwaway parse_fragment wrapper, and the parsed document node.
Callers that let someone walk a parsed tree upwards need this. Handing back
the wrapper would leak an internal name into a script-facing API — and worse,
let a script insert <_lxfragment> into the page — while the document node
would serialize the whole chunk from a parent() call. It also removes an
arbitrary inconsistency: a single-root chunk’s top node sits directly under
the document node, a multi-root chunk’s under the wrapper, and neither is
something the caller wrote.