ANTLR: Generating Parse Trees using Python

Question

I'm new to ANTLR and currently I'm trying to use ANTLR 3.1.3 with Python. I've already installed successfully ANTLR runtime for Python. But I don't know how to print out the parse tree of a particular input.

Grammar

grammar Expr;

options {
    language=Python;
    output=AST;
}

expr    :   atom (op atom)* NEWLINE ;
op  :   '+' | '-';
atom    :   ID | INT;
ID  :   ('a'..'z'|'A'..'Z')+ ;  
INT     :   ('0'..'9')+ ;
NEWLINE :   '\r'? '\n' ;
WS  :   (' '|'\t'|'\n'|'\r')+ {self.skip()} ;

Python code

import sys
import antlr3
from antlr3 import *
from ExprLexer import ExprLexer
from ExprParser import ExprParser

# test the parser with an input
char_stream = antlr3.ANTLRStringStream('3+5\n')
lexer = ExprLexer(char_stream)
tokens = antlr3.CommonTokenStream(lexer)
parser = ExprParser(tokens)

# print the parse tree
t = parser.expr().tree
print t.toStringTree()

Although what I want when the code's executed is the parse tree. The code will print out '3+5' only.

Would you please tell me how to modify the code to get the parse tree printed?

Alessandro Mariotti · Accepted Answer · 2014-07-03 08:32:03Z

1

Some weeks ago, I wanted to print out the ANTLR AST (not the parse tree, but it depends on how you implemented your grammar..) for debugging purposes.

You can do something very easy like:

def print_tree(tree, lev):
    print (" " * lev) + "` " + str(tree)
    for c in tree.getChildren():
        print_tree(c, lev + 1)

print_tree(parser.expr().tree, 0)

NB: I've written the code directly in SO, so it is untested

answered Jul 3, 2014 at 8:32

Alessandro Mariotti

631 silver badge7 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

RoyM · Accepted Answer · 2023-05-18 12:49:00Z

Here is another version...

from antlr4 import *
from antlr4.tree.Tree import TerminalNodeImpl
from antlr4.tree.Trees import Trees

def dump(node, depth=0, ruleNames=None):
    depthStr = '. ' * depth
    if isinstance(node, TerminalNodeImpl):
        print(f'{depthStr}{node.symbol}')
    else:
        print(f'{depthStr}{Trees.getNodeText(node, ruleNames)}')
        for child in node.children:
            dump(child, depth + 1, ruleNames)


def test_parse_tree():
    istream = FileStream(INFILE, encoding='utf-8')
    lexer = MyLexer(istream)
    stream = CommonTokenStream(lexer)
    parser = MyParser(stream)
    tree = parser.MAINRULE()
    print(tree.toStringTree(recog=parser))  # lisp repr of tree
    dump(tree, ruleNames=parser.ruleNames)  # pretty printed version

It yields output like so...

prog
. line
. . text
. . . [@0,0:0='@',<5>,1:0]
. . cmnt
. . . [@1,1:2='\n@',<1>,1:1]
. . . [@2,3:3=' ',<4>,2:1]
. . . text
. . . . [@3,4:4=' ',<4>,2:2]

Collectives™ on Stack Overflow

ANTLR: Generating Parse Trees using Python

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related