markdown-it/lib/rules_core/linkify.js

// Replace link-like texts with link nodes.
//
// Currently restricted by `inline.validateLink()` to http/https/ftp
//
'use strict';


var Autolinker = require('autolinker');


var LINK_SCAN_RE = /www|@|\:\/\//;


function isLinkOpen(str) {
  return /^<a[>\s]/i.test(str);
}
function isLinkClose(str) {
  return /^<\/a\s*>/i.test(str);
}

// Stupid fabric to avoid singletons, for thread safety.
// Required for engines like Nashorn.
//
function createLinkifier() {
  var links = [];
  var autolinker = new Autolinker({
    stripPrefix: false,
    url: true,
    email: true,
    twitter: false,
    replaceFn: function (autolinker, match) {
      // Only collect matched strings but don't change anything.
      switch (match.getType()) {
        /*eslint default-case:0*/
        case 'url':
          links.push({
            text: match.matchedText,
            url: match.getUrl()
          });
          break;
        case 'email':
          links.push({
            text: match.matchedText,
            // normalize email protocol
            url: 'mailto:' + match.getEmail().replace(/^mailto:/i, '')
          });
          break;
      }
      return false;
    }
  });

  return {
    links: links,
    autolinker: autolinker
  };
}


module.exports = function linkify(state) {
  var i, j, l, tokens, token, text, nodes, ln, pos, level, htmlLinkLevel,
      blockTokens = state.tokens,
      linkifier = null, links, autolinker;

  if (!state.options.linkify) { return; }

  for (j = 0, l = blockTokens.length; j < l; j++) {
    if (blockTokens[j].type !== 'inline') { continue; }
    tokens = blockTokens[j].children;

    htmlLinkLevel = 0;

    // We scan from the end, to keep position when new tags added.
    // Use reversed logic in links start/end match
    for (i = tokens.length - 1; i >= 0; i--) {
      token = tokens[i];

      // Skip content of markdown links
      if (token.type === 'link_close') {
        i--;
        while (tokens[i].level !== token.level && tokens[i].type !== 'link_open') {
          i--;
        }
        continue;
      }

      // Skip content of html tag links
      if (token.type === 'htmltag') {
        if (isLinkOpen(token.content) && htmlLinkLevel > 0) {
          htmlLinkLevel--;
        }
        if (isLinkClose(token.content)) {
          htmlLinkLevel++;
        }
      }
      if (htmlLinkLevel > 0) { continue; }

      if (token.type === 'text' && LINK_SCAN_RE.test(token.content)) {

        // Init linkifier in lazy manner, only if required.
        if (!linkifier) {
          linkifier = createLinkifier();
          links = linkifier.links;
          autolinker = linkifier.autolinker;
        }

        text = token.content;
        links.length = 0;
        autolinker.link(text);

        if (!links.length) { continue; }

        // Now split string to nodes
        nodes = [];
        level = token.level;

        for (ln = 0; ln < links.length; ln++) {

          if (!state.inline.validateLink(links[ln].url)) { continue; }

          pos = text.indexOf(links[ln].text);

          if (pos) {
            level = level;
            nodes.push({
              type: 'text',
              content: text.slice(0, pos),
              level: level
            });
          }
          nodes.push({
            type: 'link_open',
            href: links[ln].url,
            title: '',
            level: level++
          });
          nodes.push({
            type: 'text',
            content: links[ln].text,
            level: level
          });
          nodes.push({
            type: 'link_close',
            level: --level
          });
          text = text.slice(pos + links[ln].text.length);
        }
        if (text.length) {
          nodes.push({
            type: 'text',
            content: text,
            level: level
          });
        }

        // replace current node
        blockTokens[j].children = tokens = [].concat(tokens.slice(0, i), nodes, tokens.slice(i + 1));
      }
    }
  }
};