...
You can still do something like this manually with very little work. Described in the reference book Chapter 10, "Error Reporting and Recovery" : "Altering Recognizer Error Messages". The new v3 mechanism lends itself to a more flexible "top down" paraphrase mechanism. Rules as well as tokens may be paraphrased. As a simple start the paraphrase option may be replaced trivially...
| Code Block |
|---|
LPAREN options { paraphrase = "("; } : '(';
|
...by...
| Code Block |
|---|
...
@members {
...
Stack<String> paraphrase = new Stack<String>();
...
}
...
LPAREN
@init { paraphrase.push("("); }
@after { paraphrase.pop(); }
: '(';
|
Per-Token AST Type Specs
ANTLR 2 allowed the grammar to specify an AST implementation class per token type.
...