view release on metacpan or  search on metacpan

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

/*! http://mths.be/he v0.4.1 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 or Browserified code,
	// and use it as `root`.
	var freeGlobal = typeof global == 'object' && global;
	if (freeGlobal.global === freeGlobal || freeGlobal.window === freeGlobal) {
		root = freeGlobal;
	}

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

	// All astral symbols.
	var regexAstralSymbols = /[\uD800-\uDBFF][\uDC00-\uDFFF]/g;
	// All ASCII symbols (not just printable ASCII) except those listed in the
	// first column of the overrides table.
	// http://whatwg.org/html/tokenization.html#table-charref-overrides
	var regexAsciiWhitelist = /[\x01-\x7F]/g;
	// All BMP symbols that are not ASCII newlines, printable ASCII symbols, or
	// code points listed in the first column of the overrides table on
	// http://whatwg.org/html/tokenization.html#table-charref-overrides.
	var regexBmpWhitelist = /[\x01-\t\x0B\f\x0E-\x1F\x7F\x81\x8D\x8F\x90\x9D\xA0-\uFFFF]/g;

	var regexEncodeNonAscii = /<\u20D2|=\u20E5|>\u20D2|\u205F\u200A|\u219D\u0338|\u2202\u0338|\u2220\u20D2|\u2229\uFE00|\u222A\uFE00|\u223C\u20D2|\u223D\u0331|\u223E\u0333|\u2242\u0338|\u224B\u0338|\u224D\u20D2|\u224E\u0338|\u224F\u0338|\u2250\u0338|\u2...
	var encodeMap = {'\xC1':'Aacute','\xE1':'aacute','\u0102':'Abreve','\u0103':'abreve','\u223E':'ac','\u223F':'acd','\u223E\u0333':'acE','\xC2':'Acirc','\xE2':'acirc','\xB4':'acute','\u0410':'Acy','\u0430':'acy','\xC6':'AElig','\xE6':'aelig','\u2061':...

	var regexEscape = /["&'<>`]/g;
	var escapeMap = {
		'"': '&quot;',
		'&': '&amp;',
		'\'': '&#x27;',
		'<': '&lt;',
		// See http://mathiasbynens.be/notes/ambiguous-ampersands: in HTML, the
		// following is not strictly necessary unless it’s part of a tag or an
		// unquoted attribute value. We’re only escaping it to support those
		// situations, and for XML support.
		'>': '&gt;',
		// In Internet Explorer ≤ 8, the backtick character can be used
		// to break out of (un)quoted attribute values or HTML comments.
		// See http://html5sec.org/#102, http://html5sec.org/#108, and
		// http://html5sec.org/#133.
		'`': '&#x60;'
	};

	var regexInvalidEntity = /&#(?:[xX][^a-fA-F0-9]|[^0-9xX])/;
	var regexInvalidRawCodePoint = /[\0-\x08\x0B\x0E-\x1F\x7F-\x9F\uFDD0-\uFDEF\uFFFE\uFFFF]|[\uD83F\uD87F\uD8BF\uD8FF\uD93F\uD97F\uD9BF\uD9FF\uDA3F\uDA7F\uDABF\uDAFF\uDB3F\uDB7F\uDBBF\uDBFF][\uDFFE\uDFFF]|[\uD800-\uDBFF](?![\uDC00-\uDFFF])|(?:[^\uD800-...
	var regexDecode = /&#([0-9]+)(;?)|&#[xX]([a-fA-F0-9]+)(;?)|&([0-9a-zA-Z]+);|&(Aacute|iacute|Uacute|plusmn|otilde|Otilde|Agrave|agrave|yacute|Yacute|oslash|Oslash|Atilde|atilde|brvbar|Ccedil|ccedil|ograve|curren|divide|Eacute|eacute|Ograve|oacute|Egr...
	var decodeMap = {'Aacute':'\xC1','aacute':'\xE1','Abreve':'\u0102','abreve':'\u0103','ac':'\u223E','acd':'\u223F','acE':'\u223E\u0333','Acirc':'\xC2','acirc':'\xE2','acute':'\xB4','Acy':'\u0410','acy':'\u0430','AElig':'\xC6','aelig':'\xE6','af':'\u2...
	var decodeMapLegacy = {'Aacute':'\xC1','aacute':'\xE1','Acirc':'\xC2','acirc':'\xE2','acute':'\xB4','AElig':'\xC6','aelig':'\xE6','Agrave':'\xC0','agrave':'\xE0','amp':'&','AMP':'&','Aring':'\xC5','aring':'\xE5','Atilde':'\xC3','atilde':'\xE3','Auml...
	var decodeMapNumeric = {'0':'\uFFFD','128':'\u20AC','130':'\u201A','131':'\u0192','132':'\u201E','133':'\u2026','134':'\u2020','135':'\u2021','136':'\u02C6','137':'\u2030','138':'\u0160','139':'\u2039','140':'\u0152','142':'\u017D','145':'\u2018','1...
	var invalidReferenceCodePoints = [1,2,3,4,5,6,7,8,11,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,64976,6...

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

	var stringFromCharCode = String.fromCharCode;

	var object = {};
	var hasOwnProperty = object.hasOwnProperty;
	var has = function(object, propertyName) {
		return hasOwnProperty.call(object, propertyName);
	};

	var contains = function(array, value) {
		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
			// Encode astral symbols.
			.replace(regexAstralSymbols, function($0) {
				// http://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
				var high = $0.charCodeAt(0);
				var low = $0.charCodeAt(1);
				var codePoint = (high - 0xD800) * 0x400 + low - 0xDC00 + 0x10000;
				return '&#x' + codePoint.toString(16).toUpperCase() + ';';
			})
			// Encode any remaining BMP symbols that are not printable ASCII symbols
			// using a hexadecimal escape.
			.replace(regexBmpWhitelist, hexEscape);
	};
	// Expose default options (so they can be overridden globally).
	encode.options = {
		'encodeEverything': false,
		'strict': false,
		'useNamedReferences': false
	};

	var decode = function(html, options) {
		options = merge(options, decode.options);
		var strict = options.strict;
		if (strict && regexInvalidEntity.test(html)) {
			parseError('malformed character reference');
		}
		return html.replace(regexDecode, function($0, $1, $2, $3, $4, $5, $6, $7) {
			var codePoint;
			var semicolon;
			var hexDigits;
			var reference;
			var next;
			if ($1) {
				// Decode decimal escapes, e.g. `&#119558;`.
				codePoint = $1;
				semicolon = $2;
				if (strict && !semicolon) {
					parseError('character reference was not terminated by a semicolon');
				}
				return codePointToSymbol(codePoint, strict);
			}
			if ($3) {
				// Decode hexadecimal escapes, e.g. `&#x1D306;`.
				hexDigits = $3;
				semicolon = $4;
				if (strict && !semicolon) {
					parseError('character reference was not terminated by a semicolon');
				}
				codePoint = parseInt(hexDigits, 16);
				return codePointToSymbol(codePoint, strict);
			}
			if ($5) {
				// Decode named character references with trailing `;`, e.g. `&copy;`.
				reference = $5;
				if (has(decodeMap, reference)) {
					return decodeMap[reference];
				} else {
					// Ambiguous ampersand; see http://mths.be/notes/ambiguous-ampersands.
					if (strict) {
						parseError(
							'named character reference was not terminated by a semicolon'
						);
					}
					return $0;
				}
			}
			// If we’re still here, it’s a legacy reference for sure. No need for an
			// extra `if` check.
			// Decode named character references without trailing `;`, e.g. `&amp`
			// This is only a parse error if it gets converted to `&`, or if it is
			// followed by `=` in an attribute context.
			reference = $6;
			next = $7;
			if (next && options.isAttributeValue) {
				if (strict && next == '=') {
					parseError('`&` did not start a character reference');
				}
				return $0;
			} else {
				if (strict) {
					parseError(
						'named character reference was not terminated by a semicolon'
					);
				}
				// Note: there is no need to check `has(decodeMapLegacy, reference)`.
				return decodeMapLegacy[reference] + (next || '');
			}
		});
	};
	// Expose default options (so they can be overridden globally).
	decode.options = {
		'isAttributeValue': false,
		'strict': false
	};

	var escape = function(string) {
		return string.replace(regexEscape, function($0) {
			// Note: there is no need to check `has(escapeMap, $0)` here.
			return escapeMap[$0];
		});
	};

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

	var he = {
		'version': '0.4.1',
		'encode': encode,
		'decode': decode,
		'escape': escape,
		'unescape': decode
	};

	// 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 he;
		});
	}	else if (freeExports && !freeExports.nodeType) {
		if (freeModule) { // in Node.js or RingoJS v0.8.0+
			freeModule.exports = he;
		} else { // in Narwhal or RingoJS v0.7.0-
			for (var key in he) {
				has(he, key) && (freeExports[key] = he[key]);
			}
		}
	} else { // in Rhino or a web browser
		root.he = he;
	}

}(this));