The AST does not expose the entire ancestry of each node.
The walker methods do provide access to the current parent, but also not the entire ancestry.
To gather the entire ancestry for a a given sub tree of the AST you can use gatherNodeAncestry.
The result is a Map with the child nodes as keys and the parents as values.
This allows you to lookup any ancestor of any node.
import { parseComponentValue } from '@csstools/css-parser-algorithms';const result = parseComponentValue(tokens, options);const ancestry = gatherNodeAncestry(result);
Options
{ onParseError?: (error: ParseError) => void}
onParseError
The parser algorithms are forgiving and won't stop when a parse error is encountered.
Parse errors also aren't tokens.
To receive parsing error information you can set a callback.
Parser errors will try to inform you about the point in the parsing logic the error happened.
This tells you the kind of error.
Goals and non-goals
Things this package aims to be:
specification compliant CSS parser
a reliable low level package to be used in CSS sub-grammars
What it is not:
opinionated
fast
small
a replacement for PostCSS (PostCSS is fast and also an ecosystem)