@futpib/parser 1.0.2 → 1.0.4

package/src/regularExpressionParser.ts

@@ -0,0 +1,921 @@
+import { type Parser } from './parser.js';
+import { createUnionParser } from './unionParser.js';
+import { createExactSequenceParser } from './exactSequenceParser.js';
+import { promiseCompose } from './promiseCompose.js';
+import { createTupleParser } from './tupleParser.js';
+import { createArrayParser } from './arrayParser.js';
+import { createParserAccessorParser } from './parserAccessorParser.js';
+import { createElementParser } from './elementParser.js';
+import { parserCreatorCompose } from './parserCreatorCompose.js';
+import { createOptionalParser } from './optionalParser.js';
+import { createFixedLengthSequenceParser } from './fixedLengthSequenceParser.js';
+import { createTerminatedArrayParser } from './terminatedArrayParser.js';
+import { createDisjunctionParser } from './disjunctionParser.js';
+import { createNegativeLookaheadParser } from './negativeLookaheadParser.js';
+import {
+	type CharacterSet,
+	type CodePointRange,
+	type RegularExpression,
+	type RepeatBounds,
+} from './regularExpression.js';
+
+// CharacterSet helpers
+
+const emptyCharacterSet: CharacterSet = { type: 'empty' };
+
+function codePointRangeIsEmpty(range: CodePointRange): boolean {
+	return range.start > range.end;
+}
+
+function codePointRangeIsStrictlyBefore(rangeA: CodePointRange, rangeB: CodePointRange): boolean {
+	return rangeA.end + 1 < rangeB.start;
+}
+
+function codePointRangeIsStrictlyAfter(rangeA: CodePointRange, rangeB: CodePointRange): boolean {
+	return codePointRangeIsStrictlyBefore(rangeB, rangeA);
+}
+
+function codePointRangeLeastUpperBound(rangeA: CodePointRange, rangeB: CodePointRange): CodePointRange {
+	if (codePointRangeIsEmpty(rangeA)) return rangeB;
+	if (codePointRangeIsEmpty(rangeB)) return rangeA;
+	return {
+		start: Math.min(rangeA.start, rangeB.start),
+		end: Math.max(rangeA.end, rangeB.end),
+	};
+}
+
+function codePointRangeStrictlyDisjoint(rangeA: CodePointRange, rangeB: CodePointRange): boolean {
+	return codePointRangeIsStrictlyBefore(rangeA, rangeB) || codePointRangeIsStrictlyAfter(rangeA, rangeB);
+}
+
+function characterSetNode(range: CodePointRange, left: CharacterSet, right: CharacterSet): CharacterSet {
+	return { type: 'node', range, left, right };
+}
+
+function* characterSetGetRanges(set: CharacterSet): Generator<CodePointRange> {
+	if (set.type === 'node') {
+		yield* characterSetGetRanges(set.left);
+		yield set.range;
+		yield* characterSetGetRanges(set.right);
+	}
+}
+
+function characterSetExtractOverlap(set: CharacterSet, range: CodePointRange): { restCharSet: CharacterSet; extendedRange: CodePointRange } {
+	if (set.type === 'empty') {
+		return { restCharSet: set, extendedRange: range };
+	}
+
+	let extendedRange = range;
+	let newLeft = set.left;
+	let newRight = set.right;
+
+	if (range.start < set.range.start) {
+		const resultLeft = characterSetExtractOverlap(set.left, range);
+		extendedRange = codePointRangeLeastUpperBound(extendedRange, resultLeft.extendedRange);
+		newLeft = resultLeft.restCharSet;
+	}
+
+	if (range.end > set.range.end) {
+		const resultRight = characterSetExtractOverlap(set.right, range);
+		extendedRange = codePointRangeLeastUpperBound(extendedRange, resultRight.extendedRange);
+		newRight = resultRight.restCharSet;
+	}
+
+	if (codePointRangeStrictlyDisjoint(range, set.range)) {
+		return {
+			extendedRange,
+			restCharSet: characterSetNode(set.range, newLeft, newRight),
+		};
+	}
+
+	return {
+		extendedRange: codePointRangeLeastUpperBound(set.range, extendedRange),
+		restCharSet: characterSetUnion(newLeft, newRight),
+	};
+}
+
+function characterSetInsertRange(set: CharacterSet, range: CodePointRange): CharacterSet {
+	if (codePointRangeIsEmpty(range)) {
+		return set;
+	}
+
+	if (set.type === 'empty') {
+		return characterSetNode(range, emptyCharacterSet, emptyCharacterSet);
+	}
+
+	if (codePointRangeIsStrictlyBefore(range, set.range)) {
+		return characterSetNode(set.range, characterSetInsertRange(set.left, range), set.right);
+	}
+
+	if (codePointRangeIsStrictlyAfter(range, set.range)) {
+		return characterSetNode(set.range, set.left, characterSetInsertRange(set.right, range));
+	}
+
+	const resultLeft = characterSetExtractOverlap(set.left, range);
+	const resultRight = characterSetExtractOverlap(set.right, range);
+	const resultRange = [set.range, resultLeft.extendedRange, resultRight.extendedRange].reduce(codePointRangeLeastUpperBound);
+
+	if (codePointRangeIsEmpty(resultRange)) {
+		return emptyCharacterSet;
+	}
+
+	return characterSetNode(resultRange, resultLeft.restCharSet, resultRight.restCharSet);
+}
+
+function characterSetUnion(setA: CharacterSet, setB: CharacterSet): CharacterSet {
+	return [...characterSetGetRanges(setB)].reduce(characterSetInsertRange, setA);
+}
+
+function codePointRangeSplitAt(point: number, range: CodePointRange): [CodePointRange, CodePointRange] {
+	return [
+		{ start: range.start, end: Math.min(range.end, point) },
+		{ start: Math.max(range.start, point + 1), end: range.end },
+	];
+}
+
+function codePointRangeUnion(rangeA: CodePointRange, rangeB: CodePointRange): CodePointRange[] {
+	if (codePointRangeIsEmpty(rangeA) && codePointRangeIsEmpty(rangeB)) return [];
+	if (codePointRangeIsEmpty(rangeA)) return [rangeB];
+	if (codePointRangeIsEmpty(rangeB)) return [rangeA];
+	if (rangeA.end + 1 < rangeB.start) return [rangeA, rangeB];
+	if (rangeB.end + 1 < rangeA.start) return [rangeB, rangeA];
+	return [{
+		start: Math.min(rangeA.start, rangeB.start),
+		end: Math.max(rangeA.end, rangeB.end),
+	}];
+}
+
+function codePointRangeDifference(rangeA: CodePointRange, rangeB: CodePointRange): CodePointRange[] {
+	const [before, restRangeA] = codePointRangeSplitAt(rangeB.start - 1, rangeA);
+	const [, after] = codePointRangeSplitAt(rangeB.end, restRangeA);
+	return codePointRangeUnion(before, after);
+}
+
+function characterSetDeleteRange(set: CharacterSet, range: CodePointRange): CharacterSet {
+	if (codePointRangeIsEmpty(range)) {
+		return set;
+	}
+
+	if (set.type === 'empty') {
+		return emptyCharacterSet;
+	}
+
+	const [rangeBeforeStart] = codePointRangeSplitAt(set.range.start - 1, range);
+	const [rangeRest2, rangeAfterEnd] = codePointRangeSplitAt(set.range.end, range);
+	const newLeft = characterSetDeleteRange(set.left, rangeBeforeStart);
+	const newRight = characterSetDeleteRange(set.right, rangeAfterEnd);
+	const setRangeRest = codePointRangeDifference(set.range, rangeRest2);
+
+	if (setRangeRest.length === 0) {
+		return characterSetUnion(newLeft, newRight);
+	}
+
+	if (setRangeRest.length === 1) {
+		return characterSetNode(setRangeRest[0]!, newLeft, newRight);
+	}
+
+	// setRangeRest.length === 2
+	return characterSetUnion(
+		characterSetInsertRange(newLeft, setRangeRest[0]!),
+		characterSetInsertRange(newRight, setRangeRest[1]!),
+	);
+}
+
+function characterSetDifference(setA: CharacterSet, setB: CharacterSet): CharacterSet {
+	return [...characterSetGetRanges(setB)].reduce(characterSetDeleteRange, setA);
+}
+
+function characterSetFromRange(range: CodePointRange): CharacterSet {
+	if (codePointRangeIsEmpty(range)) {
+		return emptyCharacterSet;
+	}
+	return characterSetNode(range, emptyCharacterSet, emptyCharacterSet);
+}
+
+function characterSetSingleton(char: string): CharacterSet {
+	const codePoint = char.codePointAt(0)!;
+	return characterSetFromRange({ start: codePoint, end: codePoint });
+}
+
+function characterSetCharRange(startChar: string, endChar: string): CharacterSet {
+	const start = startChar.codePointAt(0)!;
+	const end = endChar.codePointAt(0)!;
+	return characterSetFromRange({ start, end });
+}
+
+function characterSetFromArray(chars: string[]): CharacterSet {
+	return chars.map(characterSetSingleton).reduce(characterSetUnion, emptyCharacterSet);
+}
+
+function characterSetComplement(set: CharacterSet): CharacterSet {
+	return characterSetDifference(alphabet, set);
+}
+
+// Pre-defined character sets
+const alphabet: CharacterSet = characterSetDifference(
+	characterSetFromRange({ start: 0, end: 0x10FFFF }),
+	characterSetFromArray(['\r', '\n', '\u2028', '\u2029']),
+);
+
+const wildcardCharacterSet: CharacterSet = characterSetDifference(
+	alphabet,
+	characterSetFromArray(['\r', '\n', '\u2028', '\u2029']),
+);
+
+const digitChars: CharacterSet = characterSetCharRange('0', '9');
+const nonDigitChars: CharacterSet = characterSetComplement(digitChars);
+
+const wordChars: CharacterSet = [
+	characterSetCharRange('a', 'z'),
+	characterSetCharRange('A', 'Z'),
+	characterSetCharRange('0', '9'),
+	characterSetSingleton('_'),
+].reduce(characterSetUnion);
+const nonWordChars: CharacterSet = characterSetComplement(wordChars);
+
+const whiteSpaceChars: CharacterSet = [
+	characterSetSingleton('\f'),
+	characterSetSingleton('\n'),
+	characterSetSingleton('\r'),
+	characterSetSingleton('\t'),
+	characterSetSingleton('\v'),
+	characterSetSingleton('\u0020'),
+	characterSetSingleton('\u00a0'),
+	characterSetSingleton('\u1680'),
+	characterSetCharRange('\u2000', '\u200a'),
+	characterSetSingleton('\u2028'),
+	characterSetSingleton('\u2029'),
+	characterSetSingleton('\u202f'),
+	characterSetSingleton('\u205f'),
+	characterSetSingleton('\u3000'),
+	characterSetSingleton('\ufeff'),
+].reduce(characterSetUnion);
+const nonWhiteSpaceChars: CharacterSet = characterSetComplement(whiteSpaceChars);
+
+// AST constructors
+
+const epsilon: RegularExpression = { type: 'epsilon' };
+
+function literal(charset: CharacterSet): RegularExpression {
+	return { type: 'literal', charset };
+}
+
+function concat(left: RegularExpression, right: RegularExpression): RegularExpression {
+	return { type: 'concat', left, right };
+}
+
+function union(left: RegularExpression, right: RegularExpression): RegularExpression {
+	return { type: 'union', left, right };
+}
+
+function star(inner: RegularExpression): RegularExpression {
+	return { type: 'star', inner };
+}
+
+function plus(inner: RegularExpression): RegularExpression {
+	return { type: 'plus', inner };
+}
+
+function optional(inner: RegularExpression): RegularExpression {
+	return { type: 'optional', inner };
+}
+
+function repeat(inner: RegularExpression, bounds: RepeatBounds): RegularExpression {
+	return { type: 'repeat', inner, bounds };
+}
+
+function captureGroup(inner: RegularExpression, name?: string): RegularExpression {
+	if (name === undefined) {
+		return { type: 'capture-group', inner };
+	}
+	return { type: 'capture-group', inner, name };
+}
+
+function lookahead(isPositive: boolean, inner: RegularExpression, right: RegularExpression): RegularExpression {
+	return { type: 'lookahead', isPositive, inner, right };
+}
+
+function startAnchor(left: RegularExpression, right: RegularExpression): RegularExpression {
+	return { type: 'start-anchor', left, right };
+}
+
+function endAnchor(left: RegularExpression, right: RegularExpression): RegularExpression {
+	return { type: 'end-anchor', left, right };
+}
+
+// Parser implementation
+
+const elementParser: Parser<string, string> = createElementParser();
+
+const metaCharacters = new Set(['\\', '^', '$', '.', '|', '?', '*', '+', '(', ')', '[', ']', '{', '}']);
+
+// Escape sequences for control characters
+const escapeNParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\n'),
+	() => literal(characterSetSingleton('\n')),
+);
+
+const escapeRParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\r'),
+	() => literal(characterSetSingleton('\r')),
+);
+
+const escapeTParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\t'),
+	() => literal(characterSetSingleton('\t')),
+);
+
+const escapeFParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\f'),
+	() => literal(characterSetSingleton('\f')),
+);
+
+const escapeVParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\v'),
+	() => literal(characterSetSingleton('\v')),
+);
+
+const escape0Parser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\0'),
+	() => literal(characterSetSingleton('\0')),
+);
+
+// Character class escapes
+const escapeDigitParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\d'),
+	() => literal(digitChars),
+);
+
+const escapeNonDigitParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\D'),
+	() => literal(nonDigitChars),
+);
+
+const escapeWordParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\w'),
+	() => literal(wordChars),
+);
+
+const escapeNonWordParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\W'),
+	() => literal(nonWordChars),
+);
+
+const escapeSpaceParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\s'),
+	() => literal(whiteSpaceChars),
+);
+
+const escapeNonSpaceParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('\\S'),
+	() => literal(nonWhiteSpaceChars),
+);
+
+// Hex escape \xHH
+const escapeHexParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('\\x'),
+		createFixedLengthSequenceParser<string>(2),
+	]),
+	([, hexCode]) => literal(characterSetSingleton(String.fromCharCode(Number.parseInt(hexCode, 16)))),
+);
+
+// Unicode escape \uHHHH
+const escapeUnicodeParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('\\u'),
+		createFixedLengthSequenceParser<string>(4),
+	]),
+	([, hexCode]) => literal(characterSetSingleton(String.fromCharCode(Number.parseInt(hexCode, 16)))),
+);
+
+// Escaped metacharacter (e.g., \., \*, etc.)
+const escapeMetacharacterParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('\\'),
+		elementParser,
+	]),
+	([, char]) => literal(characterSetSingleton(char)),
+);
+
+// All escape sequences - use createDisjunctionParser to try specific escapes first
+const escapeParser: Parser<RegularExpression, string> = createDisjunctionParser([
+	escapeNParser,
+	escapeRParser,
+	escapeTParser,
+	escapeFParser,
+	escapeVParser,
+	escape0Parser,
+	escapeDigitParser,
+	escapeNonDigitParser,
+	escapeWordParser,
+	escapeNonWordParser,
+	escapeSpaceParser,
+	escapeNonSpaceParser,
+	escapeHexParser,
+	escapeUnicodeParser,
+	escapeMetacharacterParser, // Must be last - matches any escaped char
+]);
+
+// Dot (matches any character except newline)
+const dotParser: Parser<RegularExpression, string> = promiseCompose(
+	createExactSequenceParser('.'),
+	() => literal(wildcardCharacterSet),
+);
+
+// Literal character (non-metacharacter)
+const literalCharacterParser: Parser<RegularExpression, string> = parserCreatorCompose(
+	() => elementParser,
+	char => async parserContext => {
+		parserContext.invariant(!metaCharacters.has(char), 'Unexpected metacharacter "%s"', char);
+		return literal(characterSetSingleton(char));
+	},
+)();
+
+// Character class internals
+
+// Character in a character class (different rules than outside)
+const charClassMetaCharacters = new Set(['\\', ']', '^', '-']);
+
+// Escape sequences inside character class (returns CharacterSet)
+const charClassEscapeNParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\n'),
+	() => characterSetSingleton('\n'),
+);
+
+const charClassEscapeRParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\r'),
+	() => characterSetSingleton('\r'),
+);
+
+const charClassEscapeTParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\t'),
+	() => characterSetSingleton('\t'),
+);
+
+const charClassEscapeFParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\f'),
+	() => characterSetSingleton('\f'),
+);
+
+const charClassEscapeVParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\v'),
+	() => characterSetSingleton('\v'),
+);
+
+const charClassEscape0Parser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\0'),
+	() => characterSetSingleton('\0'),
+);
+
+const charClassEscapeDigitParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\d'),
+	() => digitChars,
+);
+
+const charClassEscapeNonDigitParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\D'),
+	() => nonDigitChars,
+);
+
+const charClassEscapeWordParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\w'),
+	() => wordChars,
+);
+
+const charClassEscapeNonWordParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\W'),
+	() => nonWordChars,
+);
+
+const charClassEscapeSpaceParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\s'),
+	() => whiteSpaceChars,
+);
+
+const charClassEscapeNonSpaceParser: Parser<CharacterSet, string> = promiseCompose(
+	createExactSequenceParser('\\S'),
+	() => nonWhiteSpaceChars,
+);
+
+const charClassEscapeHexParser: Parser<CharacterSet, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('\\x'),
+		createFixedLengthSequenceParser<string>(2),
+	]),
+	([, hexCode]) => characterSetSingleton(String.fromCharCode(Number.parseInt(hexCode, 16))),
+);
+
+const charClassEscapeUnicodeParser: Parser<CharacterSet, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('\\u'),
+		createFixedLengthSequenceParser<string>(4),
+	]),
+	([, hexCode]) => characterSetSingleton(String.fromCharCode(Number.parseInt(hexCode, 16))),
+);
+
+const charClassEscapeMetacharacterParser: Parser<CharacterSet, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('\\'),
+		elementParser,
+	]),
+	([, char]) => characterSetSingleton(char),
+);
+
+// Use createDisjunctionParser to try specific escapes before generic metacharacter escape
+const charClassEscapeParser: Parser<CharacterSet, string> = createDisjunctionParser([
+	charClassEscapeNParser,
+	charClassEscapeRParser,
+	charClassEscapeTParser,
+	charClassEscapeFParser,
+	charClassEscapeVParser,
+	charClassEscape0Parser,
+	charClassEscapeDigitParser,
+	charClassEscapeNonDigitParser,
+	charClassEscapeWordParser,
+	charClassEscapeNonWordParser,
+	charClassEscapeSpaceParser,
+	charClassEscapeNonSpaceParser,
+	charClassEscapeHexParser,
+	charClassEscapeUnicodeParser,
+	charClassEscapeMetacharacterParser, // Must be last - matches any escaped char
+]);
+
+// Single character (not escape, not ], not -)
+const charClassLiteralParser: Parser<CharacterSet, string> = parserCreatorCompose(
+	() => elementParser,
+	char => async parserContext => {
+		parserContext.invariant(!charClassMetaCharacters.has(char), 'Unexpected character class metacharacter "%s"', char);
+		return characterSetSingleton(char);
+	},
+)();
+
+// Single char in character class (escape or literal) - returns the character string for range checking
+const charClassSingleCharParser: Parser<string, string> = createUnionParser([
+	// Escape sequences that produce single chars
+	promiseCompose(createExactSequenceParser('\\n'), () => '\n'),
+	promiseCompose(createExactSequenceParser('\\r'), () => '\r'),
+	promiseCompose(createExactSequenceParser('\\t'), () => '\t'),
+	promiseCompose(createExactSequenceParser('\\f'), () => '\f'),
+	promiseCompose(createExactSequenceParser('\\v'), () => '\v'),
+	promiseCompose(createExactSequenceParser('\\0'), () => '\0'),
+	promiseCompose(
+		createTupleParser([
+			createExactSequenceParser('\\x'),
+			createFixedLengthSequenceParser<string>(2),
+		]),
+		([, hexCode]) => String.fromCharCode(Number.parseInt(hexCode, 16)),
+	),
+	promiseCompose(
+		createTupleParser([
+			createExactSequenceParser('\\u'),
+			createFixedLengthSequenceParser<string>(4),
+		]),
+		([, hexCode]) => String.fromCharCode(Number.parseInt(hexCode, 16)),
+	),
+	promiseCompose(
+		createTupleParser([
+			createExactSequenceParser('\\'),
+			elementParser,
+		]),
+		([, char]) => char,
+	),
+	// Literal char (not metacharacter, not -)
+	parserCreatorCompose(
+		() => elementParser,
+		char => async parserContext => {
+			parserContext.invariant(
+				!charClassMetaCharacters.has(char) && char !== '-',
+				'Unexpected character "%s"',
+				char,
+			);
+			return char;
+		},
+	)(),
+]);
+
+// Character range (a-z)
+const charClassRangeParser: Parser<CharacterSet, string> = promiseCompose(
+	createTupleParser([
+		charClassSingleCharParser,
+		createExactSequenceParser('-'),
+		charClassSingleCharParser,
+	]),
+	([startChar, , endChar]) => characterSetCharRange(startChar, endChar),
+);
+
+// Character class element: range, escape (for \d, \w, etc.), or single char
+const charClassElementParser: Parser<CharacterSet, string> = createDisjunctionParser([
+	charClassRangeParser,
+	charClassEscapeParser,
+	charClassLiteralParser,
+	// Literal hyphen at end or after negation
+	promiseCompose(
+		createTupleParser([
+			createExactSequenceParser('-'),
+			createNegativeLookaheadParser(createExactSequenceParser(']')),
+		]),
+		() => characterSetSingleton('-'),
+	),
+]);
+
+// Character class [...]
+const characterClassParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('['),
+		createOptionalParser(createExactSequenceParser('^')),
+		createTerminatedArrayParser(
+			charClassElementParser,
+			createExactSequenceParser(']'),
+		),
+	]),
+	([, negation, [elements]]) => {
+		let charset = elements.reduce(
+			(acc, el) => characterSetUnion(acc, el),
+			emptyCharacterSet,
+		);
+		if (negation !== undefined) {
+			charset = characterSetComplement(charset);
+		}
+		return literal(charset);
+	},
+);
+
+// Quantifiers
+type Quantifier =
+	| { type: 'star' }
+	| { type: 'plus' }
+	| { type: 'optional' }
+	| { type: 'repeat'; bounds: RepeatBounds };
+
+const starQuantifierParser: Parser<Quantifier, string> = promiseCompose(
+	createExactSequenceParser('*'),
+	() => ({ type: 'star' as const }),
+);
+
+const plusQuantifierParser: Parser<Quantifier, string> = promiseCompose(
+	createExactSequenceParser('+'),
+	() => ({ type: 'plus' as const }),
+);
+
+const optionalQuantifierParser: Parser<Quantifier, string> = promiseCompose(
+	createExactSequenceParser('?'),
+	() => ({ type: 'optional' as const }),
+);
+
+// Parse a number for quantifiers
+const numberParser: Parser<number, string> = parserCreatorCompose(
+	() => createArrayParser(parserCreatorCompose(
+		() => elementParser,
+		char => async parserContext => {
+			parserContext.invariant(char >= '0' && char <= '9', 'Expected digit, got "%s"', char);
+			return char;
+		},
+	)()),
+	digits => async parserContext => {
+		parserContext.invariant(digits.length > 0, 'Expected at least one digit');
+		return Number.parseInt(digits.join(''), 10);
+	},
+)();
+
+// {n}, {n,}, {n,m}
+const braceQuantifierParser: Parser<Quantifier, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('{'),
+		numberParser,
+		createOptionalParser(
+			createTupleParser([
+				createExactSequenceParser(','),
+				createOptionalParser(numberParser),
+			]),
+		),
+		createExactSequenceParser('}'),
+	]),
+	([, min, comma]): Quantifier => {
+		if (comma === undefined) {
+			// {n} - exactly n
+			return { type: 'repeat', bounds: min };
+		}
+		const [, max] = comma;
+		if (max === undefined) {
+			// {n,} - at least n
+			return { type: 'repeat', bounds: { min } };
+		}
+		// {n,m} - between n and m
+		return { type: 'repeat', bounds: { min, max } };
+	},
+);
+
+const quantifierParser: Parser<Quantifier, string> = createUnionParser([
+	starQuantifierParser,
+	plusQuantifierParser,
+	optionalQuantifierParser,
+	braceQuantifierParser,
+]);
+
+// Groups
+// Capture group (...)
+const captureGroupParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('('),
+		createNegativeLookaheadParser(createExactSequenceParser('?')),
+		createParserAccessorParser(() => alternationParser),
+		createExactSequenceParser(')'),
+	]),
+	([, , inner]) => captureGroup(inner),
+);
+
+// Named capture group (?<name>...)
+const namedCaptureGroupParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('(?<'),
+		createTerminatedArrayParser(
+			parserCreatorCompose(
+				() => elementParser,
+				char => async parserContext => {
+					parserContext.invariant(char !== '>', 'Unexpected ">"');
+					return char;
+				},
+			)(),
+			createExactSequenceParser('>'),
+		),
+		createParserAccessorParser(() => alternationParser),
+		createExactSequenceParser(')'),
+	]),
+	([, [nameChars], inner]) => captureGroup(inner, nameChars.join('')),
+);
+
+// Non-capture group (?:...)
+const nonCaptureGroupParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('(?:'),
+		createParserAccessorParser(() => alternationParser),
+		createExactSequenceParser(')'),
+	]),
+	([, inner]) => inner,
+);
+
+// Lookahead markers for internal use during parsing
+type LookaheadMarker = { type: 'lookahead-marker'; isPositive: boolean; inner: RegularExpression };
+
+// Positive lookahead (?=...)
+const positiveLookaheadMarkerParser: Parser<LookaheadMarker, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('(?='),
+		createParserAccessorParser(() => alternationParser),
+		createExactSequenceParser(')'),
+	]),
+	([, inner]) => ({ type: 'lookahead-marker' as const, isPositive: true, inner }),
+);
+
+// Negative lookahead (?!...)
+const negativeLookaheadMarkerParser: Parser<LookaheadMarker, string> = promiseCompose(
+	createTupleParser([
+		createExactSequenceParser('(?!'),
+		createParserAccessorParser(() => alternationParser),
+		createExactSequenceParser(')'),
+	]),
+	([, inner]) => ({ type: 'lookahead-marker' as const, isPositive: false, inner }),
+);
+
+const groupParser: Parser<RegularExpression, string> = createUnionParser([
+	namedCaptureGroupParser,
+	nonCaptureGroupParser,
+	captureGroupParser,
+]);
+
+// Anchors
+// Anchor markers for internal use during parsing
+type AnchorMarker = { type: 'start-anchor-marker' } | { type: 'end-anchor-marker' };
+type ParsedElement = RegularExpression | AnchorMarker | LookaheadMarker;
+
+const startAnchorMarkerParser: Parser<AnchorMarker, string> = promiseCompose(
+	createExactSequenceParser('^'),
+	() => ({ type: 'start-anchor-marker' as const }),
+);
+
+const endAnchorMarkerParser: Parser<AnchorMarker, string> = promiseCompose(
+	createExactSequenceParser('$'),
+	() => ({ type: 'end-anchor-marker' as const }),
+);
+
+// Atom: the basic unit that can be quantified (excluding anchors)
+const atomParser: Parser<RegularExpression, string> = createUnionParser([
+	groupParser,
+	characterClassParser,
+	escapeParser,
+	dotParser,
+	literalCharacterParser,
+]);
+
+// Quantified atom
+const quantifiedParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		atomParser,
+		createOptionalParser(quantifierParser),
+	]),
+	([atom, quantifier]) => {
+		if (quantifier === undefined) {
+			return atom;
+		}
+		switch (quantifier.type) {
+			case 'star':
+				return star(atom);
+			case 'plus':
+				return plus(atom);
+			case 'optional':
+				return optional(atom);
+			case 'repeat':
+				return repeat(atom, quantifier.bounds);
+		}
+	},
+);
+
+// Element in a sequence: either a quantified atom, anchor marker, or lookahead marker
+const sequenceElementParser: Parser<ParsedElement, string> = createUnionParser<ParsedElement, string>([
+	startAnchorMarkerParser,
+	endAnchorMarkerParser,
+	positiveLookaheadMarkerParser,
+	negativeLookaheadMarkerParser,
+	quantifiedParser,
+]);
+
+// Helper to concatenate a list of RegularExpressions (right-associative)
+function concatList(parts: RegularExpression[]): RegularExpression {
+	if (parts.length === 0) {
+		return epsilon;
+	}
+	return parts.reduceRight((acc, part) => concat(part, acc));
+}
+
+// Process elements with anchor markers and lookahead markers into proper AST
+// Handles anchors and lookahead as infix operators like @gruhn/regex-utils
+// Precedence order (lowest to highest): union -> start-anchor -> end-anchor -> lookahead -> concat
+function processElements(elements: ParsedElement[]): RegularExpression {
+	if (elements.length === 0) {
+		return epsilon;
+	}
+
+	// Process start anchors first (lowest precedence among infix operators)
+	const startAnchorIdx = elements.findIndex(e => 'type' in e && e.type === 'start-anchor-marker');
+	if (startAnchorIdx !== -1) {
+		const left = elements.slice(0, startAnchorIdx);
+		const right = elements.slice(startAnchorIdx + 1);
+		return startAnchor(processElements(left), processElements(right));
+	}
+
+	// Then end anchors
+	const endAnchorIdx = elements.findIndex(e => 'type' in e && e.type === 'end-anchor-marker');
+	if (endAnchorIdx !== -1) {
+		const left = elements.slice(0, endAnchorIdx);
+		const right = elements.slice(endAnchorIdx + 1);
+		return endAnchor(processElements(left), processElements(right));
+	}
+
+	// Then lookaheads (higher precedence than anchors)
+	const lookaheadIdx = elements.findIndex(e => 'type' in e && e.type === 'lookahead-marker');
+	if (lookaheadIdx !== -1) {
+		const marker = elements[lookaheadIdx] as LookaheadMarker;
+		const left = elements.slice(0, lookaheadIdx);
+		const right = elements.slice(lookaheadIdx + 1);
+		const lookaheadExpr = lookahead(marker.isPositive, marker.inner, processElements(right));
+		if (left.length === 0) {
+			return lookaheadExpr;
+		}
+		// If there's content before the lookahead, concatenate it
+		return concat(processElements(left), lookaheadExpr);
+	}
+
+	// No markers, just regular expressions - concatenate them
+	const regexParts = elements as RegularExpression[];
+	return concatList(regexParts);
+}
+
+// Concatenation: sequence of quantified atoms and anchors
+const concatParser: Parser<RegularExpression, string> = promiseCompose(
+	createArrayParser(sequenceElementParser),
+	processElements,
+);
+
+// Alternation: concat ('|' concat)*
+const alternationParser: Parser<RegularExpression, string> = promiseCompose(
+	createTupleParser([
+		concatParser,
+		createArrayParser(
+			promiseCompose(
+				createTupleParser([
+					createExactSequenceParser('|'),
+					concatParser,
+				]),
+				([, right]) => right,
+			),
+		),
+	]),
+	([first, rest]) => {
+		// Right-associative union like @gruhn/regex-utils
+		const allParts = [first, ...rest];
+		return allParts.reduceRight((acc, part) => union(part, acc));
+	},
+);
+
+export const regularExpressionParser: Parser<RegularExpression, string> = alternationParser;