A parser combinator like nom but specialized in parsing &[Token].
It has similar combinators as nom, but also provides convenient parser generator macro for each Token variants.
If your parser system has tokenize
or lex
stage, then it will be happy to code with picktok
to make your Syntax Tree.
Here is an example of parsing integer list from Token with variant of LParen
, RParent
, Integer
.
// mod token
mod token {
use token_combinator::TokenParser;
#[derive(Clone, TokenParser)]
pub enum Token {
LParen, // (
RParen, // )
Integer(i64), // 10, 0xFF, 0b01, 0o70...
}
}
mod ast {
pub enum AST {
IntegerLiteral,
List(Vec<AST>),
}
}
use token::parser::*;
pub fn parse_list(tokens: &[token::Token]) -> ast::AST {
map(
delimited(
l_paren,
many0(map(integer, |i| AST::IntegerLiteral(i))),
r_paren,
),
|integers| AST::List(integers),
)(tokens)
}
// mod parser
use token::parser::*
fn parse_list(tokens: &[Token]) -> AST {
}