116 lines
3.9 KiB
JavaScript
116 lines
3.9 KiB
JavaScript
export const xmlReplacer = /["&'<>$\x80-\uFFFF]/g;
|
|
const xmlCodeMap = new Map([
|
|
[34, """],
|
|
[38, "&"],
|
|
[39, "'"],
|
|
[60, "<"],
|
|
[62, ">"],
|
|
]);
|
|
// For compatibility with node < 4, we wrap `codePointAt`
|
|
export const getCodePoint =
|
|
// eslint-disable-next-line @typescript-eslint/no-unnecessary-condition
|
|
String.prototype.codePointAt != null
|
|
? (str, index) => str.codePointAt(index)
|
|
: // http://mathiasbynens.be/notes/javascript-encoding#surrogate-formulae
|
|
(c, index) => (c.charCodeAt(index) & 0xfc00) === 0xd800
|
|
? (c.charCodeAt(index) - 0xd800) * 0x400 +
|
|
c.charCodeAt(index + 1) -
|
|
0xdc00 +
|
|
0x10000
|
|
: c.charCodeAt(index);
|
|
/**
|
|
* Encodes all non-ASCII characters, as well as characters not valid in XML
|
|
* documents using XML entities.
|
|
*
|
|
* If a character has no equivalent entity, a
|
|
* numeric hexadecimal reference (eg. `ü`) will be used.
|
|
*/
|
|
export function encodeXML(str) {
|
|
let ret = "";
|
|
let lastIdx = 0;
|
|
let match;
|
|
while ((match = xmlReplacer.exec(str)) !== null) {
|
|
const i = match.index;
|
|
const char = str.charCodeAt(i);
|
|
const next = xmlCodeMap.get(char);
|
|
if (next !== undefined) {
|
|
ret += str.substring(lastIdx, i) + next;
|
|
lastIdx = i + 1;
|
|
}
|
|
else {
|
|
ret += `${str.substring(lastIdx, i)}&#x${getCodePoint(str, i).toString(16)};`;
|
|
// Increase by 1 if we have a surrogate pair
|
|
lastIdx = xmlReplacer.lastIndex += Number((char & 0xfc00) === 0xd800);
|
|
}
|
|
}
|
|
return ret + str.substr(lastIdx);
|
|
}
|
|
/**
|
|
* Encodes all non-ASCII characters, as well as characters not valid in XML
|
|
* documents using numeric hexadecimal reference (eg. `ü`).
|
|
*
|
|
* Have a look at `escapeUTF8` if you want a more concise output at the expense
|
|
* of reduced transportability.
|
|
*
|
|
* @param data String to escape.
|
|
*/
|
|
export const escape = encodeXML;
|
|
/**
|
|
* Creates a function that escapes all characters matched by the given regular
|
|
* expression using the given map of characters to escape to their entities.
|
|
*
|
|
* @param regex Regular expression to match characters to escape.
|
|
* @param map Map of characters to escape to their entities.
|
|
*
|
|
* @returns Function that escapes all characters matched by the given regular
|
|
* expression using the given map of characters to escape to their entities.
|
|
*/
|
|
function getEscaper(regex, map) {
|
|
return function escape(data) {
|
|
let match;
|
|
let lastIdx = 0;
|
|
let result = "";
|
|
while ((match = regex.exec(data))) {
|
|
if (lastIdx !== match.index) {
|
|
result += data.substring(lastIdx, match.index);
|
|
}
|
|
// We know that this character will be in the map.
|
|
result += map.get(match[0].charCodeAt(0));
|
|
// Every match will be of length 1
|
|
lastIdx = match.index + 1;
|
|
}
|
|
return result + data.substring(lastIdx);
|
|
};
|
|
}
|
|
/**
|
|
* Encodes all characters not valid in XML documents using XML entities.
|
|
*
|
|
* Note that the output will be character-set dependent.
|
|
*
|
|
* @param data String to escape.
|
|
*/
|
|
export const escapeUTF8 = getEscaper(/[&<>'"]/g, xmlCodeMap);
|
|
/**
|
|
* Encodes all characters that have to be escaped in HTML attributes,
|
|
* following {@link https://html.spec.whatwg.org/multipage/parsing.html#escapingString}.
|
|
*
|
|
* @param data String to escape.
|
|
*/
|
|
export const escapeAttribute = getEscaper(/["&\u00A0]/g, new Map([
|
|
[34, """],
|
|
[38, "&"],
|
|
[160, " "],
|
|
]));
|
|
/**
|
|
* Encodes all characters that have to be escaped in HTML text,
|
|
* following {@link https://html.spec.whatwg.org/multipage/parsing.html#escapingString}.
|
|
*
|
|
* @param data String to escape.
|
|
*/
|
|
export const escapeText = getEscaper(/[&<>\u00A0]/g, new Map([
|
|
[38, "&"],
|
|
[60, "<"],
|
|
[62, ">"],
|
|
[160, " "],
|
|
]));
|
|
//# sourceMappingURL=escape.js.map
|