view release on metacpan or  search on metacpan

share/public/javascripts/he.js  view on Meta::CPAN

		var index = -1;
		var length = array.length;
		while (++index < length) {
			if (array[index] == value) {
				return true;
			}
		}
		return false;
	};

	var merge = function(options, defaults) {
		if (!options) {
			return defaults;
		}
		var result = {};
		var key;
		for (key in defaults) {
			// A `hasOwnProperty` check is not needed here, since only recognized
			// option names are used anyway. Any others are ignored.
			result[key] = has(options, key) ? options[key] : defaults[key];
		}
		return result;
	};

	// Modified version of `ucs2encode`; see http://mths.be/punycode.
	var codePointToSymbol = function(codePoint, strict) {
		var output = '';
		if ((codePoint >= 0xD800 && codePoint <= 0xDFFF) || codePoint > 0x10FFFF) {
			// See issue #4:
			// “Otherwise, if the number is in the range 0xD800 to 0xDFFF or is
			// greater than 0x10FFFF, then this is a parse error. Return a U+FFFD
			// REPLACEMENT CHARACTER.”
			if (strict) {
				parseError('character reference outside the permissible Unicode range');
			}
			return '\uFFFD';
		}
		if (has(decodeMapNumeric, codePoint)) {
			if (strict) {
				parseError('disallowed character reference');
			}
			return decodeMapNumeric[codePoint];
		}
		if (strict && contains(invalidReferenceCodePoints, codePoint)) {
			parseError('disallowed character reference');
		}
		if (codePoint > 0xFFFF) {
			codePoint -= 0x10000;
			output += stringFromCharCode(codePoint >>> 10 & 0x3FF | 0xD800);
			codePoint = 0xDC00 | codePoint & 0x3FF;
		}
		output += stringFromCharCode(codePoint);
		return output;
	};

	var hexEscape = function(symbol) {
		return '&#x' + symbol.charCodeAt(0).toString(16).toUpperCase() + ';';
	};

	var parseError = function(message) {
		throw Error('Parse error: ' + message);
	};

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

	var encode = function(string, options) {
		options = merge(options, encode.options);
		var strict = options.strict;
		if (strict && regexInvalidRawCodePoint.test(string)) {
			parseError('forbidden code point');
		}
		var encodeEverything = options.encodeEverything;
		var useNamedReferences = options.useNamedReferences;
		if (encodeEverything) {
			// Encode ASCII symbols.
			string = string.replace(regexAsciiWhitelist, function(symbol) {
				// Use named references if requested & possible.
				if (useNamedReferences && has(encodeMap, symbol)) {
					return '&' + encodeMap[symbol] + ';';
				}
				return hexEscape(symbol);
			});
			// Shorten a few escapes that represent two symbols, of which at least one
			// is within the ASCII range.
			if (useNamedReferences) {
				string = string
					.replace(/&gt;\u20D2/g, '&nvgt;')
					.replace(/&lt;\u20D2/g, '&nvlt;')
					.replace(/&#x66;&#x6A;/g, '&fjlig;');
			}
			// Encode non-ASCII symbols.
			if (useNamedReferences) {
				// Encode non-ASCII symbols that can be replaced with a named reference.
				string = string.replace(regexEncodeNonAscii, function(string) {
					// Note: there is no need to check `has(encodeMap, string)` here.
					return '&' + encodeMap[string] + ';';
				});
			}
			// Note: any remaining non-ASCII symbols are handled outside of the `if`.
		} else if (useNamedReferences) {
			// Apply named character references.
			// Encode `<>"'&` using named character references.
			string = string.replace(regexEscape, function(string) {
				return '&' + encodeMap[string] + ';'; // no need to check `has()` here
			});
			// Shorten escapes that represent two symbols, of which at least one is
			// `<>"'&`.
			string = string
				.replace(/&gt;\u20D2/g, '&nvgt;')
				.replace(/&lt;\u20D2/g, '&nvlt;');
			// Encode non-ASCII symbols that can be replaced with a named reference.
			string = string.replace(regexEncodeNonAscii, function(string) {
				// Note: there is no need to check `has(encodeMap, string)` here.
				return '&' + encodeMap[string] + ';';
			});
		} else {
			// Encode `<>"'&` using hexadecimal escapes, now that they’re not handled
			// using named character references.
			string = string.replace(regexEscape, hexEscape);
		}
		return string