showdown/src/helpers.js
Estevao Soares dos Santos 6fbc072c2c fix(HTML parsing): fix HTML parsing issues with nested tags
Deeply nested HTML tags and recursive tags broke the HTML parser.

Closes #357, closes #387
2017-05-28 17:20:07 +01:00

381 lines
10 KiB
JavaScript

/**
* showdownjs helper functions
*/
if (!showdown.hasOwnProperty('helper')) {
showdown.helper = {};
}
/**
* Check if var is string
* @static
* @param {string} a
* @returns {boolean}
*/
showdown.helper.isString = function (a) {
'use strict';
return (typeof a === 'string' || a instanceof String);
};
/**
* Check if var is a function
* @static
* @param {*} a
* @returns {boolean}
*/
showdown.helper.isFunction = function (a) {
'use strict';
var getType = {};
return a && getType.toString.call(a) === '[object Function]';
};
/**
* isArray helper function
* @static
* @param {*} a
* @returns {boolean}
*/
showdown.helper.isArray = function (a) {
'use strict';
return a.constructor === Array;
};
/**
* Check if value is undefined
* @static
* @param {*} value The value to check.
* @returns {boolean} Returns `true` if `value` is `undefined`, else `false`.
*/
showdown.helper.isUndefined = function (value) {
'use strict';
return typeof value === 'undefined';
};
/**
* ForEach helper function
* Iterates over Arrays and Objects (own properties only)
* @static
* @param {*} obj
* @param {function} callback Accepts 3 params: 1. value, 2. key, 3. the original array/object
*/
showdown.helper.forEach = function (obj, callback) {
'use strict';
// check if obj is defined
if (showdown.helper.isUndefined(obj)) {
throw new Error('obj param is required');
}
if (showdown.helper.isUndefined(callback)) {
throw new Error('callback param is required');
}
if (!showdown.helper.isFunction(callback)) {
throw new Error('callback param must be a function/closure');
}
if (typeof obj.forEach === 'function') {
obj.forEach(callback);
} else if (showdown.helper.isArray(obj)) {
for (var i = 0; i < obj.length; i++) {
callback(obj[i], i, obj);
}
} else if (typeof (obj) === 'object') {
for (var prop in obj) {
if (obj.hasOwnProperty(prop)) {
callback(obj[prop], prop, obj);
}
}
} else {
throw new Error('obj does not seem to be an array or an iterable object');
}
};
/**
* Standardidize extension name
* @static
* @param {string} s extension name
* @returns {string}
*/
showdown.helper.stdExtName = function (s) {
'use strict';
return s.replace(/[_?*+\/\\.^-]/g, '').replace(/\s/g, '').toLowerCase();
};
function escapeCharactersCallback (wholeMatch, m1) {
'use strict';
var charCodeToEscape = m1.charCodeAt(0);
return '¨E' + charCodeToEscape + 'E';
}
/**
* Callback used to escape characters when passing through String.replace
* @static
* @param {string} wholeMatch
* @param {string} m1
* @returns {string}
*/
showdown.helper.escapeCharactersCallback = escapeCharactersCallback;
/**
* Escape characters in a string
* @static
* @param {string} text
* @param {string} charsToEscape
* @param {boolean} afterBackslash
* @returns {XML|string|void|*}
*/
showdown.helper.escapeCharacters = function (text, charsToEscape, afterBackslash) {
'use strict';
// First we have to escape the escape characters so that
// we can build a character class out of them
var regexString = '([' + charsToEscape.replace(/([\[\]\\])/g, '\\$1') + '])';
if (afterBackslash) {
regexString = '\\\\' + regexString;
}
var regex = new RegExp(regexString, 'g');
text = text.replace(regex, escapeCharactersCallback);
return text;
};
var rgxFindMatchPos = function (str, left, right, flags) {
'use strict';
var f = flags || '',
g = f.indexOf('g') > -1,
x = new RegExp(left + '|' + right, 'g' + f.replace(/g/g, '')),
l = new RegExp(left, f.replace(/g/g, '')),
pos = [],
t, s, m, start, end;
do {
t = 0;
while ((m = x.exec(str))) {
if (l.test(m[0])) {
if (!(t++)) {
s = x.lastIndex;
start = s - m[0].length;
}
} else if (t) {
if (!--t) {
end = m.index + m[0].length;
var obj = {
left: {start: start, end: s},
match: {start: s, end: m.index},
right: {start: m.index, end: end},
wholeMatch: {start: start, end: end}
};
pos.push(obj);
if (!g) {
return pos;
}
}
}
}
} while (t && (x.lastIndex = s));
return pos;
};
/**
* matchRecursiveRegExp
*
* (c) 2007 Steven Levithan <stevenlevithan.com>
* MIT License
*
* Accepts a string to search, a left and right format delimiter
* as regex patterns, and optional regex flags. Returns an array
* of matches, allowing nested instances of left/right delimiters.
* Use the "g" flag to return all matches, otherwise only the
* first is returned. Be careful to ensure that the left and
* right format delimiters produce mutually exclusive matches.
* Backreferences are not supported within the right delimiter
* due to how it is internally combined with the left delimiter.
* When matching strings whose format delimiters are unbalanced
* to the left or right, the output is intentionally as a
* conventional regex library with recursion support would
* produce, e.g. "<<x>" and "<x>>" both produce ["x"] when using
* "<" and ">" as the delimiters (both strings contain a single,
* balanced instance of "<x>").
*
* examples:
* matchRecursiveRegExp("test", "\\(", "\\)")
* returns: []
* matchRecursiveRegExp("<t<<e>><s>>t<>", "<", ">", "g")
* returns: ["t<<e>><s>", ""]
* matchRecursiveRegExp("<div id=\"x\">test</div>", "<div\\b[^>]*>", "</div>", "gi")
* returns: ["test"]
*/
showdown.helper.matchRecursiveRegExp = function (str, left, right, flags) {
'use strict';
var matchPos = rgxFindMatchPos (str, left, right, flags),
results = [];
for (var i = 0; i < matchPos.length; ++i) {
results.push([
str.slice(matchPos[i].wholeMatch.start, matchPos[i].wholeMatch.end),
str.slice(matchPos[i].match.start, matchPos[i].match.end),
str.slice(matchPos[i].left.start, matchPos[i].left.end),
str.slice(matchPos[i].right.start, matchPos[i].right.end)
]);
}
return results;
};
/**
*
* @param {string} str
* @param {string|function} replacement
* @param {string} left
* @param {string} right
* @param {string} flags
* @returns {string}
*/
showdown.helper.replaceRecursiveRegExp = function (str, replacement, left, right, flags) {
'use strict';
if (!showdown.helper.isFunction(replacement)) {
var repStr = replacement;
replacement = function () {
return repStr;
};
}
var matchPos = rgxFindMatchPos(str, left, right, flags),
finalStr = str,
lng = matchPos.length;
if (lng > 0) {
var bits = [];
if (matchPos[0].wholeMatch.start !== 0) {
bits.push(str.slice(0, matchPos[0].wholeMatch.start));
}
for (var i = 0; i < lng; ++i) {
bits.push(
replacement(
str.slice(matchPos[i].wholeMatch.start, matchPos[i].wholeMatch.end),
str.slice(matchPos[i].match.start, matchPos[i].match.end),
str.slice(matchPos[i].left.start, matchPos[i].left.end),
str.slice(matchPos[i].right.start, matchPos[i].right.end)
)
);
if (i < lng - 1) {
bits.push(str.slice(matchPos[i].wholeMatch.end, matchPos[i + 1].wholeMatch.start));
}
}
if (matchPos[lng - 1].wholeMatch.end < str.length) {
bits.push(str.slice(matchPos[lng - 1].wholeMatch.end));
}
finalStr = bits.join('');
}
return finalStr;
};
/**
* Returns the index within the passed String object of the first occurrence of the specified regex,
* starting the search at fromIndex. Returns -1 if the value is not found.
*
* @param {string} str string to search
* @param {RegExp} regex Regular expression to search
* @param {int} [fromIndex = 0] Index to start the search
* @returns {Number}
* @throws InvalidArgumentError
*/
showdown.helper.regexIndexOf = function (str, regex, fromIndex) {
'use strict';
if (!showdown.helper.isString(str)) {
throw 'InvalidArgumentError: first parameter of showdown.helper.regexIndexOf function must be a string';
}
if (regex instanceof RegExp === false) {
throw 'InvalidArgumentError: second parameter of showdown.helper.regexIndexOf function must be an instance of RegExp';
}
var indexOf = str.substring(fromIndex || 0).search(regex);
return (indexOf >= 0) ? (indexOf + (fromIndex || 0)) : indexOf;
};
/**
* Splits the passed string object at the defined index, and returns an array composed of the two substrings
* @param {string} str string to split
* @param {int} index index to split string at
* @returns {[string,string]}
* @throws InvalidArgumentError
*/
showdown.helper.splitAtIndex = function (str, index) {
'use strict';
if (!showdown.helper.isString(str)) {
throw 'InvalidArgumentError: first parameter of showdown.helper.regexIndexOf function must be a string';
}
return [str.substring(0, index), str.substring(index)];
};
/**
* Obfuscate an e-mail address through the use of Character Entities,
* transforming ASCII characters into their equivalent decimal or hex entities.
*
* Since it has a random component, subsequent calls to this function produce different results
*
* @param {string} mail
* @returns {string}
*/
showdown.helper.encodeEmailAddress = function (mail) {
'use strict';
var encode = [
function (ch) {
return '&#' + ch.charCodeAt(0) + ';';
},
function (ch) {
return '&#x' + ch.charCodeAt(0).toString(16) + ';';
},
function (ch) {
return ch;
}
];
mail = mail.replace(/./g, function (ch) {
if (ch === '@') {
// this *must* be encoded. I insist.
ch = encode[Math.floor(Math.random() * 2)](ch);
} else {
var r = Math.random();
// roughly 10% raw, 45% hex, 45% dec
ch = (
r > 0.9 ? encode[2](ch) : r > 0.45 ? encode[1](ch) : encode[0](ch)
);
}
return ch;
});
return mail;
};
/**
* POLYFILLS
*/
// use this instead of builtin is undefined for IE8 compatibility
if (typeof(console) === 'undefined') {
console = {
warn: function (msg) {
'use strict';
alert(msg);
},
log: function (msg) {
'use strict';
alert(msg);
},
error: function (msg) {
'use strict';
throw msg;
}
};
}
/**
* Common regexes.
* We declare some common regexes to improve performance
*/
showdown.helper.regexes = {
asteriskAndDash: /([*_])/g
};