marked/src/helpers.ts

import { other } from './rules.ts';

/**
 * Helpers
 */
const escapeReplacements: { [index: string]: string } = {
  '&': '&',
  '<': '&lt;',
  '>': '&gt;',
  '"': '&quot;',
  "'": '&#39;',
};
const getEscapeReplacement = (ch: string) => escapeReplacements[ch];

export function escape(html: string, encode?: boolean) {
  if (encode) {
    if (other.escapeTest.test(html)) {
      return html.replace(other.escapeReplace, getEscapeReplacement);
    }
  } else {
    if (other.escapeTestNoEncode.test(html)) {
      return html.replace(other.escapeReplaceNoEncode, getEscapeReplacement);
    }
  }

  return html;
}

export function unescape(html: string) {
  // explicitly match decimal, hex, and named HTML entities
  return html.replace(other.unescapeTest, (_, n) => {
    n = n.toLowerCase();
    if (n === 'colon') return ':';
    if (n.charAt(0) === '#') {
      return n.charAt(1) === 'x'
        ? String.fromCharCode(parseInt(n.substring(2), 16))
        : String.fromCharCode(+n.substring(1));
    }
    return '';
  });
}

export function cleanUrl(href: string) {
  try {
    href = encodeURI(href).replace(other.percentDecode, '%');
  } catch {
    return null;
  }
  return href;
}

export function splitCells(tableRow: string, count?: number) {
  // ensure that every cell-delimiting pipe has a space
  // before it to distinguish it from an escaped pipe
  const row = tableRow.replace(other.findPipe, (match, offset, str) => {
      let escaped = false;
      let curr = offset;
      while (--curr >= 0 && str[curr] === '\\') escaped = !escaped;
      if (escaped) {
        // odd number of slashes means | is escaped
        // so we leave it alone
        return '|';
      } else {
        // add space before unescaped |
        return ' |';
      }
    }),
    cells = row.split(other.splitPipe);
  let i = 0;

  // First/last cell in a row cannot be empty if it has no leading/trailing pipe
  if (!cells[0].trim()) {
    cells.shift();
  }
  if (cells.length > 0 && !cells.at(-1)?.trim()) {
    cells.pop();
  }

  if (count) {
    if (cells.length > count) {
      cells.splice(count);
    } else {
      while (cells.length < count) cells.push('');
    }
  }

  for (; i < cells.length; i++) {
    // leading or trailing whitespace is ignored per the gfm spec
    cells[i] = cells[i].trim().replace(other.slashPipe, '|');
  }
  return cells;
}

/**
 * Remove trailing 'c's. Equivalent to str.replace(/c*$/, '').
 * /c*$/ is vulnerable to REDOS.
 *
 * @param str
 * @param c
 * @param invert Remove suffix of non-c chars instead. Default falsey.
 */
export function rtrim(str: string, c: string, invert?: boolean) {
  const l = str.length;
  if (l === 0) {
    return '';
  }

  // Length of suffix matching the invert condition.
  let suffLen = 0;

  // Step left until we fail to match the invert condition.
  while (suffLen < l) {
    const currChar = str.charAt(l - suffLen - 1);
    if (currChar === c && !invert) {
      suffLen++;
    } else if (currChar !== c && invert) {
      suffLen++;
    } else {
      break;
    }
  }

  return str.slice(0, l - suffLen);
}

export function findClosingBracket(str: string, b: string) {
  if (str.indexOf(b[1]) === -1) {
    return -1;
  }

  let level = 0;
  for (let i = 0; i < str.length; i++) {
    if (str[i] === '\\') {
      i++;
    } else if (str[i] === b[0]) {
      level++;
    } else if (str[i] === b[1]) {
      level--;
      if (level < 0) {
        return i;
      }
    }
  }
  return -1;
}