markdown-it/lib/parser_inline.js

/** internal
 * class ParserInline
 *
 * Tokenizes paragraph content.
 **/
'use strict';


var Ruler           = require('./ruler');
var replaceEntities = require('./common/utils').replaceEntities;

////////////////////////////////////////////////////////////////////////////////
// Parser rules

var _rules = [
  [ 'text',            require('./rules_inline/text') ],
  [ 'newline',         require('./rules_inline/newline') ],
  [ 'escape',          require('./rules_inline/escape') ],
  [ 'backticks',       require('./rules_inline/backticks') ],
  [ 'strikethrough',   require('./rules_inline/strikethrough') ],
  [ 'emphasis',        require('./rules_inline/emphasis') ],
  [ 'link',            require('./rules_inline/link') ],
  [ 'image',           require('./rules_inline/image') ],
  [ 'autolink',        require('./rules_inline/autolink') ],
  [ 'html_inline',     require('./rules_inline/html_inline') ],
  [ 'entity',          require('./rules_inline/entity') ]
];


var BAD_PROTOCOLS = [ 'vbscript', 'javascript', 'file' ];

function validateLink(url) {
  // Care about digital entities "javascript:alert(1)"
  var str = replaceEntities(url);

  str = str.trim().toLowerCase();

  if (str.indexOf(':') >= 0 && BAD_PROTOCOLS.indexOf(str.split(':')[0]) >= 0) {
    return false;
  }
  return true;
}


/**
 * new ParserInline()
 **/
function ParserInline() {
  /**
   * ParserInline#validateLink(url) -> Boolean
   *
   * Link validation function. CommonMark allows too much in links. By default
   * we disable `javascript:` and `vbscript:` schemas. You can change this
   * behaviour.
   *
   * ```javascript
   * var md = require('markdown-it')();
   * // enable everything
   * md.inline.validateLink = function () { return true; }
   * ```
   **/
  this.validateLink = validateLink;

  /**
   * ParserInline#ruler -> Ruler
   *
   * [[Ruler]] instance. Keep configuration of inline rules.
   **/
  this.ruler = new Ruler();

  for (var i = 0; i < _rules.length; i++) {
    this.ruler.push(_rules[i][0], _rules[i][1]);
  }
}


// Skip single token by running all rules in validation mode;
// returns `true` if any rule reported success
//
ParserInline.prototype.skipToken = function (state) {
  var i, cached_pos, pos = state.pos,
      rules = this.ruler.getRules(''),
      len = rules.length,
      maxNesting = state.md.options.maxNesting;


  if ((cached_pos = state.cacheGet(pos)) > 0) {
    state.pos = cached_pos;
    return;
  }

  if (state.level < maxNesting) {
    for (i = 0; i < len; i++) {
      if (rules[i](state, true)) {
        state.cacheSet(pos, state.pos);
        return;
      }
    }
  }

  state.pos++;
  state.cacheSet(pos, state.pos);
};


// Generate tokens for input range
//
ParserInline.prototype.tokenize = function (state) {
  var ok, i,
      rules = this.ruler.getRules(''),
      len = rules.length,
      end = state.posMax,
      maxNesting = state.md.options.maxNesting;

  while (state.pos < end) {
    // Try all possible rules.
    // On success, rule should:
    //
    // - update `state.pos`
    // - update `state.tokens`
    // - return true

    if (state.level < maxNesting) {
      for (i = 0; i < len; i++) {
        ok = rules[i](state, false);
        if (ok) { break; }
      }
    }

    if (ok) {
      if (state.pos >= end) { break; }
      continue;
    }

    state.pending += state.src[state.pos++];
  }

  if (state.pending) {
    state.pushPending();
  }
};


/**
 * ParserInline.parse(str, md, env, outTokens)
 *
 * Process input string and push inline tokens into `outTokens`
 **/
ParserInline.prototype.parse = function (str, md, env, outTokens) {
  var state = new this.State(str, md, env, outTokens);

  this.tokenize(state);
};


ParserInline.prototype.State = require('./rules_inline/state_inline');


module.exports = ParserInline;