Files
dealplustech/node_modules/micromark-util-sanitize-uri/index.js
Kunthawat 77ac4d2d05 feat: Upgrade to Astro with full PDPA compliance
PDPA Features:
 Cookie consent banner
 Consent logging API
 Admin dashboard
 Privacy Policy
 Terms & Conditions

Technical:
 Astro 5.x + Tailwind v4
 Docker on port 80
 SQLite database
 15 pages built

Ready for Easypanel deployment.
2026-03-12 10:01:04 +07:00

107 lines
3.1 KiB
JavaScript
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
import { asciiAlphanumeric } from 'micromark-util-character';
import { encode } from 'micromark-util-encode';
/**
* Make a value safe for injection as a URL.
*
* This encodes unsafe characters with percent-encoding and skips already
* encoded sequences (see `normalizeUri`).
* Further unsafe characters are encoded as character references (see
* `micromark-util-encode`).
*
* A regex of allowed protocols can be given, in which case the URL is
* sanitized.
* For example, `/^(https?|ircs?|mailto|xmpp)$/i` can be used for `a[href]`, or
* `/^https?$/i` for `img[src]` (this is what `github.com` allows).
* If the URL includes an unknown protocol (one not matched by `protocol`, such
* as a dangerous example, `javascript:`), the value is ignored.
*
* @param {string | null | undefined} url
* URI to sanitize.
* @param {RegExp | null | undefined} [protocol]
* Allowed protocols.
* @returns {string}
* Sanitized URI.
*/
export function sanitizeUri(url, protocol) {
const value = encode(normalizeUri(url || ''));
if (!protocol) {
return value;
}
const colon = value.indexOf(':');
const questionMark = value.indexOf('?');
const numberSign = value.indexOf('#');
const slash = value.indexOf('/');
if (
// If there is no protocol, its relative.
colon < 0 ||
// If the first colon is after a `?`, `#`, or `/`, its not a protocol.
slash > -1 && colon > slash || questionMark > -1 && colon > questionMark || numberSign > -1 && colon > numberSign ||
// It is a protocol, it should be allowed.
protocol.test(value.slice(0, colon))) {
return value;
}
return '';
}
/**
* Normalize a URL.
*
* Encode unsafe characters with percent-encoding, skipping already encoded
* sequences.
*
* @param {string} value
* URI to normalize.
* @returns {string}
* Normalized URI.
*/
export function normalizeUri(value) {
/** @type {Array<string>} */
const result = [];
let index = -1;
let start = 0;
let skip = 0;
while (++index < value.length) {
const code = value.charCodeAt(index);
/** @type {string} */
let replace = '';
// A correct percent encoded value.
if (code === 37 && asciiAlphanumeric(value.charCodeAt(index + 1)) && asciiAlphanumeric(value.charCodeAt(index + 2))) {
skip = 2;
}
// ASCII.
else if (code < 128) {
if (!/[!#$&-;=?-Z_a-z~]/.test(String.fromCharCode(code))) {
replace = String.fromCharCode(code);
}
}
// Astral.
else if (code > 55_295 && code < 57_344) {
const next = value.charCodeAt(index + 1);
// A correct surrogate pair.
if (code < 56_320 && next > 56_319 && next < 57_344) {
replace = String.fromCharCode(code, next);
skip = 1;
}
// Lone surrogate.
else {
replace = "\uFFFD";
}
}
// Unicode.
else {
replace = String.fromCharCode(code);
}
if (replace) {
result.push(value.slice(start, index), encodeURIComponent(replace));
start = index + skip + 1;
replace = '';
}
if (skip) {
index += skip;
skip = 0;
}
}
return result.join('') + value.slice(start);
}