ANTLR extraneous input x expecting x

Question

I am trying to match a very basic ANTLR grammar. But ANTLR is keep telling me that he got the input '.' and expects '.' .

The full error is:

line 1:0 extraneous input '.' expecting '.'
line 1:2 missing '*' at '<EOF>'

With the grammar:

grammar regex;

@parser::header
{
    package antlr;
}

@lexer::header
{
    package antlr;
}


WHITESPACE : (' ' | '\t' | '\n' | '\r') -> channel(HIDDEN);
COMP       : '.';
KLEENE     : '*';

start : COMP KLEENE;

And input:

.*

Both files have the same charset:

regex.g: text/plain; charset=us-ascii
test.grammar: text/plain; charset=us-ascii

There should be no Lexer rule mix up. Why does this not work as expected?

Have you checked that the dot is the same character both in grammar and input, i.e. this is not yet another different-unicode-character-that-looks-the-same mixup? — Jiri Tousek
I copied the dot from the grammar into the input file. So, they should be the same. — lauw

Bart Kiers Bart Kiers · Accepted Answer · 2018-03-26T08:12:34

Given your example grammar and this test class:

import org.antlr.v4.runtime.*;

public class Main {
  public static void main(String[] args) {
    String source = ".*";
    regexLexer lexer = new regexLexer(CharStreams.fromString(source));
    regexParser parser = new regexParser(new CommonTokenStream(lexer));
    System.out.println(parser.start().toStringTree(parser));
  }
}

the following is printed to my console:

(start . *)

My guess is you have either dumbed down the grammar too much causing the error in your original grammar to disappear, or you haven't generated new lexer/parser classes.

ANTLR extraneous input x expecting x

1 Answers