regenerate/regenerate.js

Generate JavaScript-compatible regular expressions based on a given set of Unicode symbols or code points.

/*! https://mths.be/regenerate v1.4.2 by @mathias | MIT license */
;(function(root) {

	// Detect free variables `exports`.
	var freeExports = typeof exports == 'object' && exports;

	// Detect free variable `module`.
	var freeModule = typeof module == 'object' && module &&
		module.exports == freeExports && module;

	// Detect free variable `global`, from Node.js/io.js or Browserified code,
	// and use it as `root`.
	var freeGlobal = typeof global == 'object' && global;
	if (freeGlobal.global === freeGlobal || freeGlobal.window === freeGlobal) {
		root = freeGlobal;
	}

	/*--------------------------------------------------------------------------*/

	var ERRORS = {
		'rangeOrder': 'A range\u2019s `stop` value must be greater than or equal ' +
			'to the `start` value.',
		'codePointRange': 'Invalid code point value. Code points range from ' +
			'U+000000 to U+10FFFF.'
	};

	// https://mathiasbynens.be/notes/javascript-encoding#surrogate-pairs
	var HIGH_SURROGATE_MIN = 0xD800;
	var HIGH_SURROGATE_MAX = 0xDBFF;
	var LOW_SURROGATE_MIN = 0xDC00;
	var LOW_SURROGATE_MAX = 0xDFFF;

	// In Regenerate output, `\0` is never preceded by `\` because we sort by
	// code point value, so let’s keep this regular expression simple.
	var regexNull = /\\x00([^0123456789]|$)/g;

	var object = {};
	var hasOwnProperty = object.hasOwnProperty;
	var extend = function(destination, source) {
		var key;
		for (key in source) {
			if (hasOwnProperty.call(source, key)) {
				destination[key] = source[key];
			}
		}
		return destination;
	};

	var forEach = function(array, callback) {
		var index = -1;
		var length = array.length;
		while (++index < length) {
			callback(array[index], index);
		}
	};

	var toString = object.toString;
	var isArray = function(value) {
		return toString.call(value) == '[object Array]';
	};
	var isNumber = function(value) {
		return typeof value == 'number' ||
			toString.call(value) == '[object Number]';
	};

	// `hexBytes[n]` is the uppercase, zero-padded, two-digit hexadecimal
	// representation of `n`, for `0 <= n <= 0xFF`.
	var hexBytes = [];
	(function() {
		var digits = '0123456789ABCDEF';
		var index = -1;
		while (++index <= 0xFF) {
			hexBytes[index] = digits.charAt(index >> 4) + digits.charAt(index & 0xF);
		}
	}());

	// These assume that `number` is an integer in the range the output can
	// represent, which is the case for the BMP code points they're used with.
	var hex2 = function(number) {
		return hexBytes[number];
	};

	var hex4 = function(number) {
		return hexBytes[number >> 8] + hexBytes[number & 0xFF];
	};

	var slice = [].slice;

	/*--------------------------------------------------------------------------*/

	var compareNumbers = function(a, b) {
		return a - b;
	};

	// Turns an array of code points into data. The code points may be in any
	// order and may contain duplicates. Note: unsorted `codePoints` are sorted
	// in place.
	var dataFromCodePoints = function(codePoints) {
		var length = codePoints.length;
		var index = 0;
		var start;
		var end;
		var codePoint;
		var result = [];
		// Turn runs of consecutive code points into `(start, end)` pairs, skipping
		// duplicates. Input is often sorted already, so rather than sorting up
		// front, sort and start over only once a code point turns out to be
		// smaller than the one before it.
		while (index < length) {
			// The inner loop only stops at a code point above `end`, so `start` is
			// always greater than result[result.length - 1].
			start = codePoints[index];
			end = start + 1;
			while (++index < length && (codePoint = codePoints[index]) <= end) {
				if (codePoint == end) {
					++end;
				} else if (codePoint < end - 1) {
					codePoints.sort(compareNumbers);
					return dataFromCodePoints(codePoints);
				}
			}
			result.push(start, end);
		}
		return result;
	};

	var dataRemove = function(data, codePoint) {
		// Iterate over the data per `(start, end)` pair.
		var index = 0;
		var start;
		var end;
		var length = data.length;
		while (index < length) {
			start = data[index];
			end = data[index + 1];
			if (codePoint >= start && codePoint < end) {
				// Modify this pair.
				if (codePoint == start) {
					if (end == start + 1) {
						// Just remove `start` and `end`.
						data.splice(index, 2);
						return data;
					} else {
						// Just replace `start` with a new value.
						data[index] = codePoint + 1;
						return data;
					}
				} else if (codePoint == end - 1) {
					// Just replace `end` with a new value.
					data[index + 1] = codePoint;
					return data;
				} else {
					// Replace `[start, end]` with `[startA, endA, startB, endB]`.
					data.splice(index, 2, start, codePoint, codePoint + 1, end);
					return data;
				}
			}
			index += 2;
		}
		return data;
	};

	var dataRemoveRange = function(data, rangeStart, rangeEnd) {
		if (rangeEnd < rangeStart) {
			throw Error(ERRORS.rangeOrder);
		}
		var length = data.length;
		var rangeEndExclusive = rangeEnd + 1;
		var index = 0;
		var first;
		var start;
		var end;
		// Skip the pairs that end before the range starts.
		while (index < length && data[index + 1] <= rangeStart) {
			index += 2;
		}
		// Find the pairs that overlap the range, i.e. `data[first]` up to (but not
		// including) `data[index]`.
		first = index;
		while (index < length && data[index] < rangeEndExclusive) {
			index += 2;
		}
		if (first == index) {
			return data;
		}
		// Replace the overlapping pairs with a single `splice`, keeping the parts of
		// the first and last pair that lie outside the range, if any.
		// E.g. we have `[0, 11, 14, 31]` and want to remove 4-20 -> `[0, 4, 21, 31]`.
		start = data[first];
		end = data[index - 1];
		if (start < rangeStart) {
			if (end > rangeEndExclusive) {
				data.splice(
					first, index - first, start, rangeStart, rangeEndExclusive, end
				);
			} else {
				data.splice(first, index - first, start, rangeStart);
			}
		} else if (end > rangeEndExclusive) {
			data.splice(first, index - first, rangeEndExclusive, end);
		} else {
			data.splice(first, index - first);
		}
		return data;
	};

	 var dataAdd = function(data, codePoint) {
		// Iterate over the data per `(start, end)` pair.
		var index = 0;
		var start;
		var end;
		var lastIndex = null;
		var length = data.length;
		if (codePoint < 0x0 || codePoint > 0x10FFFF) {
			throw RangeError(ERRORS.codePointRange);
		}
		// Fast path: the code point lies at or after the start of the last pair,
		// which is the common case when adding code points in ascending order.
		if (length && codePoint >= data[length - 2]) {
			end = data[length - 1];
			if (codePoint > end) {
				data.push(codePoint, codePoint + 1);
			} else if (codePoint == end) {
				data[length - 1] = codePoint + 1;
			}
			return data;
		}
		while (index < length) {
			start = data[index];
			end = data[index + 1];

			// Check if the code point is already in the set.
			if (codePoint >= start && codePoint < end) {
				return data;
			}

			if (codePoint == start - 1) {
				// Just replace `start` with a new value.
				data[index] = codePoint;
				return data;
			}

			// At this point, if `start` is `greater` than `codePoint`, insert a new
			// `[start, end]` pair before the current pair, or after the current pair
			// if there is a known `lastIndex`.
			if (start > codePoint) {
				data.splice(
					lastIndex != null ? lastIndex + 2 : 0,
					0,
					codePoint,
					codePoint + 1
				);
				return data;
			}

			if (codePoint == end) {
				// Check if adding this code point causes two separate ranges to become
				// a single range, e.g. `dataAdd([0, 4, 5, 10], 4)` → `[0, 10]`.
				if (codePoint + 1 == data[index + 2]) {
					data.splice(index, 4, start, data[index + 3]);
					return data;
				}
				// Else, just replace `end` with a new value.
				data[index + 1] = codePoint + 1;
				return data;
			}
			lastIndex = index;
			index += 2;
		}
		// The loop has finished; add the new pair to the end of the data set.
		data.push(codePoint, codePoint + 1);
		return data;
	};

	var dataAddRange = function(data, rangeStart, rangeEnd) {
		if (rangeEnd < rangeStart) {
			throw Error(ERRORS.rangeOrder);
		}
		if (
			rangeStart < 0x0 || rangeStart > 0x10FFFF ||
			rangeEnd < 0x0 || rangeEnd > 0x10FFFF
		) {
			throw RangeError(ERRORS.codePointRange);
		}
		var length = data.length;
		var rangeEndExclusive = rangeEnd + 1;
		var index = 0;
		var first;
		var start;
		var end;
		// Fast path: the range starts at or after the start of the last pair, which
		// is the common case when adding ranges in ascending order.
		if (length && rangeStart >= data[length - 2]) {
			end = data[length - 1];
			if (rangeStart > end) {
				data.push(rangeStart, rangeEndExclusive);
			} else if (rangeEndExclusive > end) {
				// The range overlaps or touches the last pair; extend it.
				data[length - 1] = rangeEndExclusive;
			}
			return data;
		}
		// Skip the pairs that end before the range starts, without touching it.
		while (index < length && data[index + 1] < rangeStart) {
			index += 2;
		}
		// Find the pairs that overlap or touch the range, i.e. `data[first]` up to
		// (but not including) `data[index]`.
		first = index;
		while (index < length && data[index] <= rangeEndExclusive) {
			index += 2;
		}
		if (first == index) {
			// Nothing to merge with; insert a new pair.
			data.splice(first, 0, rangeStart, rangeEndExclusive);
			return data;
		}
		// Merge the range and the pairs it overlaps or touches into a single pair.
		// E.g. `[0, 11, 12, 14, 20, 31]` and you add 5-15 -> `[0, 16, 20, 31]`.
		start = data[first] < rangeStart ? data[first] : rangeStart;
		end = data[index - 1] > rangeEndExclusive ?
			data[index - 1] :
			rangeEndExclusive;
		if (index - first == 2) {
			data[first] = start;
			data[first + 1] = end;
		} else {
			data.splice(first, index - first, start, end);
		}
		return data;
	};

	var dataContains = function(data, codePoint) {
		// Binary search for the number of `(start, end)` pairs that start at or
		// before `codePoint`. `codePoint` is in the set if it lies before the end of
		// the last of those pairs.
		var length = data.length;
		var low = 0;
		var high = length >> 1;
		var middle;
		// Exit early if `codePoint` is not within `data`'s overall range.
		if (!length || codePoint < data[0] || codePoint >= data[length - 1]) {
			return false;
		}
		while (low < high) {
			middle = (low + high) >> 1;
			if (data[middle << 1] <= codePoint) {
				low = middle + 1;
			} else {
				high = middle;
			}
		}
		return low > 0 && codePoint < data[(low << 1) - 1];
	};

	var dataIntersectionData = function(dataA, dataB) {
		var indexA = 0;
		var indexB = 0;
		var lengthA = dataA.length;
		var lengthB = dataB.length;
		var startA, endA, startB, endB, start, end;
		var result = [];
		while (indexA < lengthA && indexB < lengthB) {
			startA = dataA[indexA];
			endA = dataA[indexA + 1];
			startB = dataB[indexB];
			endB = dataB[indexB + 1];
			start = startA > startB ? startA : startB;
			end = endA < endB ? endA : endB;
			if (start < end) {
				result.push(start, end);
			}
			if (endA < endB) {
				indexA += 2;
			} else {
				indexB += 2;
			}
		}
		return result;
	};

	var dataUnion = function(dataA, dataB) {
		// Walk both sorted lists of `(start, end)` pairs in lockstep, always taking
		// the pair that starts first, and merge it into the previous output pair if
		// the two overlap or touch.
		var indexA = 0;
		var indexB = 0;
		var lengthA = dataA.length;
		var lengthB = dataB.length;
		var start;
		var end;
		var last;
		var result = [];
		while (indexA < lengthA || indexB < lengthB) {
			if (
				indexB >= lengthB ||
				(indexA < lengthA && dataA[indexA] <= dataB[indexB])
			) {
				start = dataA[indexA];
				end = dataA[indexA + 1];
				indexA += 2;
			} else {
				start = dataB[indexB];
				end = dataB[indexB + 1];
				indexB += 2;
			}
			last = result.length - 1;
			if (last > 0 && start <= result[last]) {
				if (end > result[last]) {
					result[last] = end;
				}
			} else {
				result.push(start, end);
			}
		}
		return result;
	};

	var dataDifference = function(dataA, dataB) {
		// For each pair in `dataA`, cut out the overlapping pairs of `dataB`.
		var indexA = 0;
		var indexB = 0;
		var lengthA = dataA.length;
		var lengthB = dataB.length;
		var index;
		var start;
		var end;
		var result = [];
		while (indexA < lengthA) {
			start = dataA[indexA];
			end = dataA[indexA + 1];
			indexA += 2;
			// Skip the pairs of `dataB` that end before this pair starts.
			while (indexB < lengthB && dataB[indexB + 1] <= start) {
				indexB += 2;
			}
			index = indexB;
			while (start < end && index < lengthB && dataB[index] < end) {
				if (dataB[index] > start) {
					result.push(start, dataB[index]);
				}
				if (dataB[index + 1] > start) {
					start = dataB[index + 1];
				}
				index += 2;
			}
			if (start < end) {
				result.push(start, end);
			}
		}
		return result;
	};

	// Collects the code points in `values` (anything `add` or `remove` accepts in
	// an array: code points, symbols, nested arrays, and Regenerate instances)
	// and returns them as data. With `validate`, out-of-range code points throw
	// like `dataAdd` does; otherwise they’re skipped, as `dataRemove` does.
	var dataFromValues = function(values, validate) {
		var codePoints = [];
		var sets = [];
		var collect = function(values) {
			var index = -1;
			var length = values.length;
			var value;
			var codePoint;
			while (++index < length) {
				value = values[index];
				// Check for primitive numbers first, as they’re the common case.
				if (typeof value == 'number') {
					codePoint = value;
				} else if (value == null) {
					continue;
				} else if (value instanceof regenerate) {
					sets.push(value.data);
					continue;
				} else if (isArray(value)) {
					collect(value);
					continue;
				} else {
					codePoint = isNumber(value) ? +value : symbolToCodePoint(value);
				}
				if (codePoint >= 0x0 && codePoint <= 0x10FFFF) {
					codePoints.push(codePoint);
				} else if (validate) {
					throw RangeError(ERRORS.codePointRange);
				}
			}
		};
		collect(values);

		var data = dataFromCodePoints(codePoints);
		var index = -1;
		var length = sets.length;
		while (++index < length) {
			data = dataUnion(data, sets[index]);
		}
		return data;
	};

	var dataIsEmpty = function(data) {
		return !data.length;
	};

	var dataIsSingleton = function(data) {
		// Check if the set only represents a single code point.
		return data.length == 2 && data[0] + 1 == data[1];
	};

	var dataToArray = function(data) {
		// Iterate over the data per `(start, end)` pair.
		var index = 0;
		var start;
		var end;
		var result = [];
		var length = data.length;
		while (index < length) {
			start = data[index];
			end = data[index + 1];
			while (start < end) {
				result.push(start);
				++start;
			}
			index += 2;
		}
		return result;
	};

	/*--------------------------------------------------------------------------*/

	// https://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
	var floor = Math.floor;
	var highSurrogate = function(codePoint) {
		return parseInt(
			floor((codePoint - 0x10000) / 0x400) + HIGH_SURROGATE_MIN,
			10
		);
	};

	var lowSurrogate = function(codePoint) {
		return parseInt(
			(codePoint - 0x10000) % 0x400 + LOW_SURROGATE_MIN,
			10
		);
	};

	var stringFromCharCode = String.fromCharCode;
	var codePointToString = function(codePoint) {
		var string;
		// https://mathiasbynens.be/notes/javascript-escapes#single
		// Note: the `\b` escape sequence for U+0008 BACKSPACE in strings has a
		// different meaning in regular expressions (word boundary), so it cannot
		// be used here.
		if (codePoint == 0x09) {
			string = '\\t';
		}
		// Note: IE < 9 treats `'\v'` as `'v'`, so avoid using it.
		// else if (codePoint == 0x0B) {
		// 	string = '\\v';
		// }
		else if (codePoint == 0x0A) {
			string = '\\n';
		}
		else if (codePoint == 0x0C) {
			string = '\\f';
		}
		else if (codePoint == 0x0D) {
			string = '\\r';
		}
		else if (codePoint == 0x2D) {
			// https://mathiasbynens.be/notes/javascript-escapes#hexadecimal
			// Note: `-` (U+002D HYPHEN-MINUS) is escaped in this way rather
			// than by backslash-escaping, in case the output is used outside
			// of a character class in a `u` RegExp. /\-/u throws, but
			// /\x2D/u is fine.
			string = '\\x2D';
		}
		else if (codePoint == 0x5C) {
			string = '\\\\';
		}
		else if (
			codePoint == 0x24 ||
			(codePoint >= 0x28 && codePoint <= 0x2B) ||
			codePoint == 0x2E || codePoint == 0x2F ||
			codePoint == 0x3F ||
			(codePoint >= 0x5B && codePoint <= 0x5E) ||
			(codePoint >= 0x7B && codePoint <= 0x7D)
		) {
			// The code point maps to an unsafe printable ASCII character;
			// backslash-escape it. Here’s the list of those symbols:
			//
			//     $()*+./?[\]^{|}
			//
			// This matches SyntaxCharacters as well as `/` (U+002F SOLIDUS).
			// https://tc39.github.io/ecma262/#prod-SyntaxCharacter
			string = '\\' + stringFromCharCode(codePoint);
		}
		else if (codePoint >= 0x20 && codePoint <= 0x7E) {
			// The code point maps to one of these printable ASCII symbols
			// (including the space character):
			//
			//      !"#%&',0123456789:;<=>@ABCDEFGHIJKLMNO
			//     PQRSTUVWXYZ_`abcdefghijklmnopqrstuvwxyz~
			//
			// These can safely be used directly.
			string = stringFromCharCode(codePoint);
		}
		else if (codePoint <= 0xFF) {
			string = '\\x' + hex2(codePoint);
		}
		else { // `codePoint <= 0xFFFF` holds true.
			// https://mathiasbynens.be/notes/javascript-escapes#unicode
			string = '\\u' + hex4(codePoint);
		}

		// There’s no need to account for astral symbols / surrogate pairs here,
		// since `codePointToString` is private and only used for BMP code points.
		// But if that’s what you need, just add an `else` block with this code:
		//
		//     string = '\\u' + hex4(highSurrogate(codePoint))
		//     	+ '\\u' + hex4(lowSurrogate(codePoint));

		return string;
	};

	var codePointToStringUnicode = function(codePoint) {
		if (codePoint <= 0xFFFF) {
			if (codePoint < HIGH_SURROGATE_MIN || codePoint > LOW_SURROGATE_MAX) {
				return codePointToString(codePoint);
			}
			// With the `u` flag, `\uD834\uDF06` is read as the single code point
			// U+1D306, so a high surrogate followed by a low surrogate would no
			// longer match either of them. `\u{…}` escapes are never combined.
			return '\\u{' + hex4(codePoint) + '}';
		}
		// Astral code points have five or six hexadecimal digits: their highest
		// byte is 0x01 to 0x10, written without its leading zero, if any.
		var high = codePoint >> 16;
		return '\\u{' +
			(high < 0x10 ? hexBytes[high].charAt(1) : '10') +
			hexBytes[(codePoint >> 8) & 0xFF] +
			hexBytes[codePoint & 0xFF] +
			'}';
	};

	var symbolToCodePoint = function(symbol) {
		var length = symbol.length;
		var first = symbol.charCodeAt(0);
		var second;
		if (
			first >= HIGH_SURROGATE_MIN && first <= HIGH_SURROGATE_MAX &&
			length > 1 // There is a next code unit.
		) {
			// `first` is a high surrogate, and there is a next character. Assume
			// it’s a low surrogate (else it’s invalid usage of Regenerate anyway).
			second = symbol.charCodeAt(1);
			// https://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
			return (first - HIGH_SURROGATE_MIN) * 0x400 +
				second - LOW_SURROGATE_MIN + 0x10000;
		}
		return first;
	};

	// Turns a code point or a symbol into a primitive code point, so that
	// `Number` objects don’t end up in the data. Primitive numbers are returned
	// as is, which keeps them fast to work with.
	var toCodePoint = function(value) {
		return typeof value == 'number' ? value :
			isNumber(value) ? +value : symbolToCodePoint(value);
	};

	var createBMPCharacterClasses = function(data) {
		// Iterate over the data per `(start, end)` pair.
		var result = '';
		var index = 0;
		var start;
		var end;
		var length = data.length;
		if (dataIsSingleton(data)) {
			return codePointToString(data[0]);
		}
		while (index < length) {
			start = data[index];
			end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.
			if (start == end) {
				result += codePointToString(start);
			} else if (start + 1 == end) {
				result += codePointToString(start) + codePointToString(end);
			} else {
				result += codePointToString(start) + '-' + codePointToString(end);
			}
			index += 2;
		}
		return '[' + result + ']';
	};

	var createUnicodeCharacterClasses = function(data) {
		// Iterate over the data per `(start, end)` pair.
		var result = '';
		var index = 0;
		var start;
		var end;
		var length = data.length;
		if (dataIsSingleton(data)) {
			return codePointToStringUnicode(data[0]);
		}
		while (index < length) {
			start = data[index];
			end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.
			if (start == end) {
				result += codePointToStringUnicode(start);
			} else if (start + 1 == end) {
				result += codePointToStringUnicode(start) + codePointToStringUnicode(end);
			} else {
				result += codePointToStringUnicode(start) + '-' + codePointToStringUnicode(end);
			}
			index += 2;
		}
		return '[' + result + ']';
	};

	var splitAtBMP = function(data) {
		// Iterate over the data per `(start, end)` pair.
		var loneHighSurrogates = [];
		var loneLowSurrogates = [];
		var bmp = [];
		var astral = [];
		var index = 0;
		var start;
		var end;
		var length = data.length;
		while (index < length) {
			start = data[index];
			end = data[index + 1] - 1; // Note: the `- 1` makes `end` inclusive.

			if (start < HIGH_SURROGATE_MIN) {

				// The range starts and ends before the high surrogate range.
				// E.g. (0, 0x10).
				if (end < HIGH_SURROGATE_MIN) {
					bmp.push(start, end + 1);
				}

				// The range starts before the high surrogate range and ends within it.
				// E.g. (0, 0xD855).
				if (end >= HIGH_SURROGATE_MIN && end <= HIGH_SURROGATE_MAX) {
					bmp.push(start, HIGH_SURROGATE_MIN);
					loneHighSurrogates.push(HIGH_SURROGATE_MIN, end + 1);
				}

				// The range starts before the high surrogate range and ends in the low
				// surrogate range. E.g. (0, 0xDCFF).
				if (end >= LOW_SURROGATE_MIN && end <= LOW_SURROGATE_MAX) {
					bmp.push(start, HIGH_SURROGATE_MIN);
					loneHighSurrogates.push(HIGH_SURROGATE_MIN, HIGH_SURROGATE_MAX + 1);
					loneLowSurrogates.push(LOW_SURROGATE_MIN, end + 1);
				}

				// The range starts before the high surrogate range and ends after the
				// low surrogate range. E.g. (0, 0x10FFFF).
				if (end > LOW_SURROGATE_MAX) {
					bmp.push(start, HIGH_SURROGATE_MIN);
					loneHighSurrogates.push(HIGH_SURROGATE_MIN, HIGH_SURROGATE_MAX + 1);
					loneLowSurrogates.push(LOW_SURROGATE_MIN, LOW_SURROGATE_MAX + 1);
					if (end <= 0xFFFF) {
						bmp.push(LOW_SURROGATE_MAX + 1, end + 1);
					} else {
						bmp.push(LOW_SURROGATE_MAX + 1, 0xFFFF + 1);
						astral.push(0xFFFF + 1, end + 1);
					}
				}

			} else if (start >= HIGH_SURROGATE_MIN && start <= HIGH_SURROGATE_MAX) {

				// The range starts and ends in the high surrogate range.
				// E.g. (0xD855, 0xD866).
				if (end >= HIGH_SURROGATE_MIN && end <= HIGH_SURROGATE_MAX) {
					loneHighSurrogates.push(start, end + 1);
				}

				// The range starts in the high surrogate range and ends in the low
				// surrogate range. E.g. (0xD855, 0xDCFF).
				if (end >= LOW_SURROGATE_MIN && end <= LOW_SURROGATE_MAX) {
					loneHighSurrogates.push(start, HIGH_SURROGATE_MAX + 1);
					loneLowSurrogates.push(LOW_SURROGATE_MIN, end + 1);
				}

				// The range starts in the high surrogate range and ends after the low
				// surrogate range. E.g. (0xD855, 0x10FFFF).
				if (end > LOW_SURROGATE_MAX) {
					loneHighSurrogates.push(start, HIGH_SURROGATE_MAX + 1);
					loneLowSurrogates.push(LOW_SURROGATE_MIN, LOW_SURROGATE_MAX + 1);
					if (end <= 0xFFFF) {
						bmp.push(LOW_SURROGATE_MAX + 1, end + 1);
					} else {
						bmp.push(LOW_SURROGATE_MAX + 1, 0xFFFF + 1);
						astral.push(0xFFFF + 1, end + 1);
					}
				}

			} else if (start >= LOW_SURROGATE_MIN && start <= LOW_SURROGATE_MAX) {

				// The range starts and ends in the low surrogate range.
				// E.g. (0xDCFF, 0xDDFF).
				if (end >= LOW_SURROGATE_MIN && end <= LOW_SURROGATE_MAX) {
					loneLowSurrogates.push(start, end + 1);
				}

				// The range starts in the low surrogate range and ends after the low
				// surrogate range. E.g. (0xDCFF, 0x10FFFF).
				if (end > LOW_SURROGATE_MAX) {
					loneLowSurrogates.push(start, LOW_SURROGATE_MAX + 1);
					if (end <= 0xFFFF) {
						bmp.push(LOW_SURROGATE_MAX + 1, end + 1);
					} else {
						bmp.push(LOW_SURROGATE_MAX + 1, 0xFFFF + 1);
						astral.push(0xFFFF + 1, end + 1);
					}
				}

			} else if (start > LOW_SURROGATE_MAX && start <= 0xFFFF) {

				// The range starts and ends after the low surrogate range.
				// E.g. (0xFFAA, 0x10FFFF).
				if (end <= 0xFFFF) {
					bmp.push(start, end + 1);
				} else {
					bmp.push(start, 0xFFFF + 1);
					astral.push(0xFFFF + 1, end + 1);
				}

			} else {

				// The range starts and ends in the astral range.
				astral.push(start, end + 1);

			}

			index += 2;
		}
		return {
			'loneHighSurrogates': loneHighSurrogates,
			'loneLowSurrogates': loneLowSurrogates,
			'bmp': bmp,
			'astral': astral
		};
	};

	var optimizeSurrogateMappings = function(surrogateMappings) {
		var result = [];
		var tmpLow = [];
		var addLow = false;
		var mapping;
		var nextMapping;
		var highSurrogates;
		var lowSurrogates;
		var nextHighSurrogates;
		var nextLowSurrogates;
		var index = -1;
		var length = surrogateMappings.length;
		while (++index < length) {
			mapping = surrogateMappings[index];
			nextMapping = surrogateMappings[index + 1];
			if (!nextMapping) {
				result.push(mapping);
				continue;
			}
			highSurrogates = mapping[0];
			lowSurrogates = mapping[1];
			nextHighSurrogates = nextMapping[0];
			nextLowSurrogates = nextMapping[1];

			// Check for identical high surrogate ranges.
			tmpLow = lowSurrogates;
			while (
				nextHighSurrogates &&
				highSurrogates[0] == nextHighSurrogates[0] &&
				highSurrogates[1] == nextHighSurrogates[1]
			) {
				// Merge with the next item.
				if (dataIsSingleton(nextLowSurrogates)) {
					tmpLow = dataAdd(tmpLow, nextLowSurrogates[0]);
				} else {
					tmpLow = dataAddRange(
						tmpLow,
						nextLowSurrogates[0],
						nextLowSurrogates[1] - 1
					);
				}
				++index;
				mapping = surrogateMappings[index];
				highSurrogates = mapping[0];
				lowSurrogates = mapping[1];
				nextMapping = surrogateMappings[index + 1];
				nextHighSurrogates = nextMapping && nextMapping[0];
				nextLowSurrogates = nextMapping && nextMapping[1];
				addLow = true;
			}
			result.push([
				highSurrogates,
				addLow ? tmpLow : lowSurrogates
			]);
			addLow = false;
		}
		return optimizeByLowSurrogates(result);
	};

	var optimizeByLowSurrogates = function(surrogateMappings) {
		if (surrogateMappings.length == 1) {
			return surrogateMappings;
		}
		var index = -1;
		var innerIndex = -1;
		while (++index < surrogateMappings.length) {
			var mapping = surrogateMappings[index];
			var lowSurrogates = mapping[1];
			if (lowSurrogates.length !== 2) {
				// Merge can only be performed when both lowSurrogates.length and
				// otherLowSurrogates.length are equal, that is, both are single
				// ranges of length 2.
				continue;
			}
			var lowSurrogateStart = lowSurrogates[0];
			var lowSurrogateEnd = lowSurrogates[1];
			innerIndex = index; // Note: the loop starts at the next index.
			while (++innerIndex < surrogateMappings.length) {
				var otherMapping = surrogateMappings[innerIndex];
				var otherLowSurrogates = otherMapping[1];
				var otherLowSurrogateStart = otherLowSurrogates[0];
				var otherLowSurrogateEnd = otherLowSurrogates[1];
				if (
					lowSurrogateStart == otherLowSurrogateStart &&
					lowSurrogateEnd == otherLowSurrogateEnd &&
					otherLowSurrogates.length === 2
				) {
					// Add the code points in the other item to this one.
					if (dataIsSingleton(otherMapping[0])) {
						mapping[0] = dataAdd(mapping[0], otherMapping[0][0]);
					} else {
						mapping[0] = dataAddRange(
							mapping[0],
							otherMapping[0][0],
							otherMapping[0][1] - 1
						);
					}
					// Remove the other, now redundant, item.
					surrogateMappings.splice(innerIndex, 1);
					--innerIndex;
				}
			}
		}
		return surrogateMappings;
	};

	var surrogateSet = function(data) {
		// Exit early if `data` is an empty set.
		if (!data.length) {
			return [];
		}

		// Iterate over the data per `(start, end)` pair.
		var index = 0;
		var start;
		var end;
		var startHigh;
		var startLow;
		var endHigh;
		var endLow;
		var surrogateMappings = [];
		var length = data.length;
		while (index < length) {
			start = data[index];
			end = data[index + 1] - 1;

			startHigh = highSurrogate(start);
			startLow = lowSurrogate(start);
			endHigh = highSurrogate(end);
			endLow = lowSurrogate(end);

			var startsWithLowestLowSurrogate = startLow == LOW_SURROGATE_MIN;
			var endsWithHighestLowSurrogate = endLow == LOW_SURROGATE_MAX;
			var complete = false;

			// Append the previous high-surrogate-to-low-surrogate mappings.
			// Step 1: `(startHigh, startLow)` to `(startHigh, LOW_SURROGATE_MAX)`.
			if (
				startHigh == endHigh ||
				startsWithLowestLowSurrogate && endsWithHighestLowSurrogate
			) {
				surrogateMappings.push([
					[startHigh, endHigh + 1],
					[startLow, endLow + 1]
				]);
				complete = true;
			} else {
				surrogateMappings.push([
					[startHigh, startHigh + 1],
					[startLow, LOW_SURROGATE_MAX + 1]
				]);
			}

			// Step 2: `(startHigh + 1, LOW_SURROGATE_MIN)` to
			// `(endHigh - 1, LOW_SURROGATE_MAX)`.
			if (!complete && startHigh + 1 < endHigh) {
				if (endsWithHighestLowSurrogate) {
					// Combine step 2 and step 3.
					surrogateMappings.push([
						[startHigh + 1, endHigh + 1],
						[LOW_SURROGATE_MIN, endLow + 1]
					]);
					complete = true;
				} else {
					surrogateMappings.push([
						[startHigh + 1, endHigh],
						[LOW_SURROGATE_MIN, LOW_SURROGATE_MAX + 1]
					]);
				}
			}

			// Step 3. `(endHigh, LOW_SURROGATE_MIN)` to `(endHigh, endLow)`.
			if (!complete) {
				surrogateMappings.push([
					[endHigh, endHigh + 1],
					[LOW_SURROGATE_MIN, endLow + 1]
				]);
			}

			index += 2;
		}

		// The format of `surrogateMappings` is as follows:
		//
		//     [ surrogateMapping1, surrogateMapping2 ]
		//
		// i.e.:
		//
		//     [
		//       [ highSurrogates1, lowSurrogates1 ],
		//       [ highSurrogates2, lowSurrogates2 ]
		//     ]
		return optimizeSurrogateMappings(surrogateMappings);
	};

	var createSurrogateCharacterClasses = function(surrogateMappings) {
		var result = [];
		forEach(surrogateMappings, function(surrogateMapping) {
			var highSurrogates = surrogateMapping[0];
			var lowSurrogates = surrogateMapping[1];
			result.push(
				createBMPCharacterClasses(highSurrogates) +
				createBMPCharacterClasses(lowSurrogates)
			);
		});
		return result.join('|');
	};

	var createCharacterClassesFromData = function(data, bmpOnly, hasUnicodeFlag) {
		if (hasUnicodeFlag) {
			return createUnicodeCharacterClasses(data);
		}
		var result = [];

		var parts = splitAtBMP(data);
		var loneHighSurrogates = parts.loneHighSurrogates;
		var loneLowSurrogates = parts.loneLowSurrogates;
		var bmp = parts.bmp;
		var astral = parts.astral;
		var hasLoneHighSurrogates = !dataIsEmpty(loneHighSurrogates);
		var hasLoneLowSurrogates = !dataIsEmpty(loneLowSurrogates);

		var surrogateMappings = surrogateSet(astral);

		if (bmpOnly) {
			bmp = dataUnion(bmp, loneHighSurrogates);
			hasLoneHighSurrogates = false;
			bmp = dataUnion(bmp, loneLowSurrogates);
			hasLoneLowSurrogates = false;
		}

		if (!dataIsEmpty(bmp)) {
			// The data set contains BMP code points that are not high surrogates
			// needed for astral code points in the set.
			result.push(createBMPCharacterClasses(bmp));
		}
		if (surrogateMappings.length) {
			// The data set contains astral code points; append character classes
			// based on their surrogate pairs.
			result.push(createSurrogateCharacterClasses(surrogateMappings));
		}
		// https://gist.github.com/mathiasbynens/bbe7f870208abcfec860
		if (hasLoneHighSurrogates) {
			result.push(
				createBMPCharacterClasses(loneHighSurrogates) +
				// Make sure the high surrogates aren’t part of a surrogate pair.
				'(?![\\uDC00-\\uDFFF])'
			);
		}
		if (hasLoneLowSurrogates) {
			result.push(
				// It is not possible to accurately assert the low surrogates aren’t
				// part of a surrogate pair, since JavaScript regular expressions do
				// not support lookbehind.
				'(?:[^\\uD800-\\uDBFF]|^)' +
				createBMPCharacterClasses(loneLowSurrogates)
			);
		}
		return result.join('|');
	};

	/*--------------------------------------------------------------------------*/

	// `regenerate` can be used as a constructor (and new methods can be added to
	// its prototype) but also as a regular function, the latter of which is the
	// documented and most common usage. For that reason, it’s not capitalized.
	var regenerate = function(value) {
		if (arguments.length > 1) {
			value = slice.call(arguments);
		}
		if (this instanceof regenerate) {
			this.data = [];
			return this.add(value);
		}
		return (new regenerate).add(value);
	};

	regenerate.version = '1.4.2';

	var proto = regenerate.prototype;
	extend(proto, {
		'add': function(value) {
			var $this = this;
			if (arguments.length > 1) {
				value = slice.call(arguments);
			}
			if (value == null) {
				return $this;
			}
			if (value instanceof regenerate) {
				// Allow passing other Regenerate instances.
				$this.data = dataUnion($this.data, value.data);
				return $this;
			}
			if (isArray(value)) {
				$this.data = dataUnion($this.data, dataFromValues(value, true));
				return $this;
			}
			$this.data = dataAdd(
				$this.data,
				toCodePoint(value)
			);
			return $this;
		},
		'remove': function(value) {
			var $this = this;
			if (arguments.length > 1) {
				value = slice.call(arguments);
			}
			if (value == null) {
				return $this;
			}
			if (value instanceof regenerate) {
				// Allow passing other Regenerate instances.
				$this.data = dataDifference($this.data, value.data);
				return $this;
			}
			if (isArray(value)) {
				$this.data = dataDifference($this.data, dataFromValues(value, false));
				return $this;
			}
			$this.data = dataRemove(
				$this.data,
				toCodePoint(value)
			);
			return $this;
		},
		'addRange': function(start, end) {
			var $this = this;
			$this.data = dataAddRange($this.data,
				toCodePoint(start),
				toCodePoint(end)
			);
			return $this;
		},
		'removeRange': function(start, end) {
			var $this = this;
			var startCodePoint = toCodePoint(start);
			var endCodePoint = toCodePoint(end);
			$this.data = dataRemoveRange(
				$this.data,
				startCodePoint,
				endCodePoint
			);
			return $this;
		},
		'intersection': function(argument) {
			var $this = this;
			if (argument instanceof regenerate) {
				// Allow passing other Regenerate instances.
				$this.data = dataIntersectionData($this.data, argument.data);
				return $this;
			}
			$this.data = dataIntersectionData(
				$this.data,
				dataFromValues(argument, false)
			);
			return $this;
		},
		'contains': function(codePoint) {
			return dataContains(
				this.data,
				isNumber(codePoint) ? codePoint : symbolToCodePoint(codePoint)
			);
		},
		'clone': function() {
			var set = new regenerate;
			set.data = this.data.slice(0);
			return set;
		},
		'toString': function(options) {
			var result = createCharacterClassesFromData(
				this.data,
				options ? options.bmpOnly : false,
				options ? options.hasUnicodeFlag : false
			);
			if (!result) {
				// For an empty set, return something that can be inserted `/here/` to
				// form a valid regular expression. Avoid `(?:)` since that matches the
				// empty string.
				return '[]';
			}
			// Use `\0` instead of `\x00` where possible.
			return result.replace(regexNull, '\\0$1');
		},
		'toRegExp': function(flags) {
			var pattern = this.toString(
				/[uv]/.test(flags || '') ?
					{ 'hasUnicodeFlag': true } :
					null
			);
			return RegExp(pattern, flags || '');
		},
		'valueOf': function() { // Note: `valueOf` is aliased as `toArray`.
			return dataToArray(this.data);
		}
	});

	proto.toArray = proto.valueOf;

	// Some AMD build optimizers, like r.js, check for specific condition patterns
	// like the following:
	if (
		typeof define == 'function' &&
		typeof define.amd == 'object' &&
		define.amd
	) {
		define(function() {
			return regenerate;
		});
	}	else if (freeExports && !freeExports.nodeType) {
		if (freeModule) { // in Node.js, io.js, or RingoJS v0.8.0+
			freeModule.exports = regenerate;
		} else { // in Narwhal or RingoJS v0.7.0-
			freeExports.regenerate = regenerate;
		}
	} else { // in Rhino or a web browser
		root.regenerate = regenerate;
	}

}(this));