Generating java code using antlr input file

The source request rules described using *.g antlr files.

I'm trying to generate java code using antlr4 and getting errors like:

error(50): mql2.g4:9:7: syntax error: mismatched input ';' expecting RBRACE
error(50): mql2.g4:10:6: syntax error: mismatched input ';' expecting COLON while matching a lexer rule
error(50): mql2.g4:11:11: syntax error: mismatched input ';' expecting COLON while matching a lexer rule
error(50): mql2.g4:12:10: syntax error: mismatched input ';' expecting COLON while matching a lexer rule
error(50): mql2.g4:16:16: syntax error: '{package com.proquest.mql.queryTranslator;}' came as a complete surprise to me while matching rule preamble
error(50): mql2.g4:17:1: syntax error: 'lexer' came as a complete surprise to me while looking for an identifier
error(50): mql2.g4:19:11: syntax error: '^' came as a complete surprise to me
error(50): mql2.g4:19:16: syntax error: '!' came as a complete surprise to me
...

the input file example is

grammar mql2;

options {
    output=AST;
    k=2;
}

tokens {
    AND_OP;
    OR_OP;
    FIELD_CODE;
    FC_SUFFIX;
    }


@parser::header {package com.company.mql.queryTranslator;}
@lexer::header {package com.company.mql.queryTranslator;}

parse   : mql^ EOF!
    ;

mql : WS!* mqlx WS!* ( and_or^ WS! mqlx WS!*)*;

and_or
    : and_operator
    | or_operator
    ;

mqlx : search_item
     | LPAREN! mql^ RPAREN!
     | field_code field_phrase RPAREN -> ^(FIELD_CODE field_code field_phrase)
     | field_code_prefix field_code_suffix ->^(FIELD_CODE field_code_prefix field_code_suffix)
     ;

field_code
    : w=WORD^ LPAREN!
    ;

field_phrase
    : (WS!* (WORD|PHRASE|AND|OR))+
    ;

field_code_prefix
    : WORD^ '.'!;

field_code_suffix
    : field_code  (WORD|PHRASE) RPAREN!;

and_operator
    : AND->AND_OP | (/*empty*/->AND_OP) ;

or_operator
    : OR->OR_OP;

search_item
        :  NOT^ WS!* mqlx
    |  (WORD|PHRASE);

LPAREN : '(';

RPAREN : ')';

AND : ('a'|'A')('n'|'N')('d'|'D');

NOT : ('n'|'N')('o'|'O')('t'|'T');

OR : ('o'|'O')('r'|'R');

fragment
DIGIT  : ('0'..'9') ;

fragment
LETTER  : ('a'..'z' | 'A'..'Z'| 'á'| '*' | '&' | '-' | '.' | ',' | '?' | '!' | '/' | '\u0080'..'\ufffe');

SPECIAL_CHAR : ('\'' | '&');

WORD    : (LETTER|DIGIT|SPECIAL_CHAR)+;

WS : ( '\t' | ' ' | '\r' | '\n' | '|')+ /*{ $channel = HIDDEN; }*/;

fragment
QUOTE :   '"' ;

PHRASE  :   QUOTE (options {greedy=false;} : . )* QUOTE ;

So the questions are:

Which version of antlr the file should be supported by? (I started reading antlr4 reference, and continued with antlr3 on their confluence but not realized yet the current version)
How to fix for antlr4 errors like a syntax error: '^' came as a complete surprise to me or syntax error: '!' came as a complete surprise to me?

Solution

As mentioned in the comments: the grammar is for ANTLR3. I recommend stop using v3 grammars: it's rather old. Converting it into a v4 grammar is easy:

grammar mql2;

@parser::header {package com.company.mql.queryTranslator;}
@lexer::header {package com.company.mql.queryTranslator;}

parse   : mql EOF
    ;

mql : mqlx (and_or mqlx)*;

and_or
    : and_operator
    | or_operator
    ;

mqlx : search_item
     | LPAREN mql RPAREN
     | field_code field_phrase RPAREN
     | field_code_prefix field_code_suffix
     ;

field_code
    : WORD LPAREN
    ;

field_phrase
    : (WORD|PHRASE|AND|OR)+
    ;

field_code_prefix
    : WORD '.';

field_code_suffix
    : field_code  (WORD|PHRASE) RPAREN;

and_operator
    : AND;

or_operator
    : OR;

search_item
        :  NOT mqlx
    |  (WORD|PHRASE);

LPAREN : '(';

RPAREN : ')';

AND : [aA] [nN] [dD];

NOT : [nN] [oO] [tT];

OR : [oO] [rR];

fragment
DIGIT  : [0-9];

fragment
LETTER  : [a-zA-Zá*&\-.,?!/\u0080-\ufffe];

fragment
SPECIAL_CHAR : ('\'' | '&');

WORD    : (LETTER|DIGIT|SPECIAL_CHAR)+;

WS : [\t \r\n|]+ -> channel(HIDDEN);

fragment
QUOTE :   '"' ;

PHRASE  :   QUOTE .*? QUOTE ;

Version 4 just gives you a parse tree which you cannot transform (into an AST) in the grammar as was possible in version 3.