Lexer rules in v4

Actions and semantic predicates

A lexer action is a block of arbitrary code in the target language surrounded by {...}, which is executed during matching:

IDENTIFIER: [A-Z]+ { log("matched rule"); };

A semantic predicate is a block of arbitrary code in the target language surrounded by {...}?, which evaluates to a boolean value. If the returned value is false, the lexer rule is skipped.

IDENTIFIER: [A-Z]+ { identifierIsValid() }?;

Semantic predicates should be defined at the end of the rule whenever possible for performance reasons.

Fragments

Fragments are reusable parts of lexer rules which cannot match on their own - they need to be referenced from a lexer rule.

INTEGER: DIGIT+
       | '0' [Xx] HEX_DIGIT+
       ;

fragment DIGIT: [0-9];
fragment HEX_DIGIT: [0-9A-Fa-f];

Implicit lexer rules

When tokens like '{' are used in a parser rule, an implicit lexer rule will be created for them unless an explicit rule exists.

In other words, if you have a lexer rule:

OPEN_BRACE: '{';

Then both of these parser rules are equivalent:

parserRule: '{';
parserRule: OPEN_BRACE;

But if the OPEN_BRACE lexer rule is not defined, an implicit anonymous rule will be created. In that case, the implicit rule will be defined as if it were defined before the other rules: it will have a higher priority than other rules.

Lexer commands

A lexer rule can have associated commands:

WHITESPACE: [ \r\n] -> skip;

Commands are defined after a -> at the end of the rule.

  • skip: Skips the matched text, no token will be emited
  • channel(n): Emits the token on a different channel
  • type(n): Changes the emitted token type
  • mode(n), pushMode(n), popMode, more: Controls lexer modes

Priority rules

Several lexer rules can match the same input text. In that case, the token type will be chosen as follows:

  • First, select the lexer rule which matches the longest input
  • If the text matches an implicitly defined token (like '{'), use the implicit rule
  • If several lexer rules match the same input length, choose the first one, based on definition order

The following combined grammar:

grammar LexerPriorityRulesExample;

// Parser rules

randomParserRule: 'foo'; // Implicitly declared token type
    
// Lexer rules
    
BAR: 'bar';
IDENTIFIER: [A-Za-z]+;
BAZ: 'baz';

WS: [ \t\r\n]+ -> skip;

Given the following input:

aaa foo bar baz barz

Will produce the following token sequence from the lexer:

IDENTIFIER 'foo' BAR IDENTIFIER IDENTIFIER
  • aaa is of type IDENTIFIER

    Only the IDENTIFIER rule can match this token, there is no ambiguity.

  • foo is of type 'foo'

    The parser rule randomParserRule introduces the implicit 'foo' token type, which is prioritary over the IDENTIFIER rule.

  • bar is of type BAR

    This text matches the BAR rule, which is defined before the IDENTIFIER rule, and therefore has precedence.

  • baz is of type IDENTIFIER

    This text matches the BAZ rule, but it also matches the IDENTIFIER rule. The latter is chosen as it is defined before BAR.

    Given the grammar, BAZ will never be able to match, as the IDENTIFIER rule already covers everything BAZ can match.

  • barz is of type IDENTIFIER

    The BAR rule can match the first 3 characters of this string (bar), but the IDENTIFIER rule will match 4 characters. As IDENTIFIER matches a longer substring, it is chosen over BAR.

As a rule of thumb, specific rules should de defined before more generic rules. If a rule can only match an input which is already covered by a previously defined rule, it will never be used.

Implicitly defined rules such as 'foo' act as if they were defined before all other lexer rules.

Simple rules

Lexer rules define token types. Their name has to start with an uppercase letter to distinguish them from parser rules.

INTEGER: [0-9]+;
IDENTIFIER: [a-zA-Z_] [a-zA-Z_0-9]*;

OPEN_PAREN: '(';
CLOSE_PAREN: ')';

Basic syntax:

SyntaxMeaning
AMatch lexer rule or fragment named A
A BMatch A followed by B
(A|B)Match either A or B
'text'Match literal "text"
A?Match A zero or one time
A*Match A zero or more times
A+Match A one or more times
[A-Z0-9]Match one character in the defined ranges (in this example between A-Z or 0-9)
'a'..'z'Alternative syntax for a character range
~[A-Z]Negation of a range - match any single character not in the range
.Match any single character


2016-07-24
2016-09-15
ANTLR Pedia
Icon