Crate markab_parser
source ·Expand description
This crate provides simple, copy-less and rich-error-message parser combinator for parsing string.
Create simple parser
You can create simple parser with character, character_class and string.
Each methods returns a class implementing Parser trait, CharacterParser, CharacterClassParser, StringParser, and returns a slice of the source string by calling the parse method of Parser trait.
use markab_parser::{
string,
Parser,
};
let src = "requirement";
let mut pos = 0;
let parser = string("requirement");
let result = parser.parse(src, &mut pos);
assert!(result.is_ok());
assert_eq!(result.ok().unwrap(), "requirement");
assert_eq!(pos, 11);
Combinate parser
Parser trait has methods for parser combination.
use markab_parser::{
character,
character_class,
Parser,
};
// require "1" and return "1"
let parser = character('1');
// require "12" and return ("1","2")
let seq = parser.and_then(character('2'));
// require "1" or "2" and return Either("1","2")
let ord = character_class(false, &[], &['1'..'2']);
// require [0-9]+ and return as usize
let map = character_class(false, &[], &['0'..'9'])
.one_or_more()
.stringify()
.map(&|num| num.parse::<usize>().unwrap());
Create new parser
If the parser has parameters, you should implement Parser trait.
If the parser does not have parameters, you should implement Parseable trait.
Parseable trait provides static method Parseable::parse and
Parseable::get_parser to get a parser instance for parser combination.
Here is an example of Parseable class WS
, which parses whitespace characters.
use markab_parser::{
character_class,
character_class_parser::CharacterClassParser,
repetition_parser::RepetitionParserError,
Parseable,
Parser,
};
pub struct WS;
impl<'a> Parseable<'a> for WS
{
type Error = RepetitionParserError<'a, CharacterClassParser<'a>>;
type Output = ();
fn parse(src: &'a str, pos: &mut usize) -> Result<Self::Output, Self::Error>
{
character_class(false, &[' ', '\t', '\n', '\r'], &[])
.one_or_more()
.skip(src, pos)
}
fn name() -> &'a str
{
"WS"
}
}