Class Lexer

static/katex/katex.mjs:12828–12894

Main Lexer class

/** Main Lexer class */

class Lexer {
  // category codes, only supports comment characters (14) for now
  constructor(input, settings) {
    this.input = void 0;
    this.settings = void 0;
    this.tokenRegex = void 0;
    this.catcodes = void 0;
    // Separate accents from characters
    this.input = input;
    this.settings = settings;
    this.tokenRegex = new RegExp(tokenRegexString, 'g');
    this.catcodes = {
      "%": 14 // comment character

    };
  }

  setCatcode(char, code) {
    this.catcodes[char] = code;
  }
  /**
   * This function lexes a single token.
   */


  lex() {
    const input = this.input;
    const pos = this.tokenRegex.lastIndex;

    if (pos === input.length) {
      return new Token("EOF", new SourceLocation(this, pos, pos));
    }

    const match = this.tokenRegex.exec(input);

    if (match === null || match.index !== pos) {
      throw new ParseError(`Unexpected character: '${input[pos]}'`, new Token(input[pos], new SourceLocation(this, pos, pos + 1)));
    }

    let text = match[2] || " ";

    if (this.catcodes[text] === 14) {
      // comment character
      const nlIndex = input.indexOf('\n', this.tokenRegex.lastIndex);

      if (nlIndex === -1) {
        this.tokenRegex.lastIndex = input.length; // EOF

        this.settings.reportNonstrict("commentAtEnd", "% comment has no terminating newline; LaTeX would " + "fail because of commenting the end of math mode (e.g. $)");
      } else {
        this.tokenRegex.lastIndex = nlIndex + 1;
      }

      return this.lex();
    } // Trim any trailing whitespace from control word match


    const controlMatch = text.match(controlWordWhitespaceRegex);