Utility	Description
`TokenType`	Re-exported `TokenType` enum from `@csstools/css-tokenizer`.
`tokens`	Tokenize a CSS string into a `CSSToken[]` array with trailing `EOF` tokens stripped.
`tokenStart`	Return the inclusive start character offset of a token (`token[2]`).
`tokenEnd`	Return the exclusive end character offset of a token (`token[3] + 1`) for string slicing.
`decoded`	Return the decoded string value of a token, falling back to its raw representation.
`numeric`	Parse a single token into `{ number, unit }`, or `false` if not a numeric token.
`endsWithEscapingBackslash`	Whether a string ends in an odd run of backslashes, so the last one escapes whatever follows.
`isHexDigitCode`	Whether a UTF-16 code unit is an ASCII hex digit (`0-9`, `a-f`, `A-F`).
`mathFunctions`	Read-only `Map` from each CSS Values 4 math function name to its `[min, max]` inclusive argument count (`max` may be `Infinity`). Iterating yields `[name, range]` pairs; use `.keys()` for names.
`balancedTokens`	Parse CSS into a `BalancedTokens` index of matching delimiters (`()`, `[]`, `{}`), or `undefined` if unbalanced.
`applyEdits`	Apply validated, non-overlapping string edits. Invalid bounds or overlaps fail closed by returning the complete original source.

`BalancedTokens` Methods

Instances returned by balancedTokens(source) provide:

endForOpening(index): Returns the closing token index for an opening delimiter token at index.
topLevelSegments(startIndex?, endIndex?, delimiter?): Splits a balanced token range by a top-level delimiter (defaults to comma).

Usage

import cssnanoUtils from 'cssnano-utils';

const { tokens, balancedTokens, applyEdits } = cssnanoUtils;

const structure = balancedTokens('cubic-bezier(0.1, 0.7, 1.0, 0.1)');
if (structure) {
  const segments = structure.topLevelSegments();
  // Process balanced segments
}

Contributors

See CONTRIBUTORS.md.

License

MIT