Compare commits
2 Commits
6e94192fa9
...
a0b8cda00d
Author | SHA1 | Date |
---|---|---|
bog | a0b8cda00d | |
bog | efdbafb645 |
|
@ -1,4 +1,8 @@
|
|||
MODULE ::= INSTR*
|
||||
INSTR ::= EXPR semicolon
|
||||
EXPR ::= BUILTIN
|
||||
BUILTIN ::= bool
|
||||
BUILTIN ::=
|
||||
| bool
|
||||
| float
|
||||
| int
|
||||
| string
|
||||
|
|
207
src/ast/lexer.rs
207
src/ast/lexer.rs
|
@ -1,8 +1,8 @@
|
|||
use crate::ast::node;
|
||||
use crate::ast::node::Node;
|
||||
|
||||
struct Token {
|
||||
position: usize,
|
||||
value: String,
|
||||
value: String
|
||||
}
|
||||
|
||||
pub struct Lexer {
|
||||
|
@ -53,6 +53,128 @@ impl Lexer {
|
|||
false
|
||||
}
|
||||
|
||||
fn scan_string(&mut self) -> Option<Token> {
|
||||
let mut value : String;
|
||||
let mut length = 1;
|
||||
|
||||
if let Some('"') = self.text.chars().nth(self.cursor) {
|
||||
value = String::from("");
|
||||
} else {
|
||||
return None;
|
||||
}
|
||||
|
||||
let mut escape = false;
|
||||
let mut closed = false;
|
||||
|
||||
for c in self.text.chars().skip(length + self.cursor) {
|
||||
if !escape && c == '\\' {
|
||||
escape = true;
|
||||
length += 1;
|
||||
continue;
|
||||
}
|
||||
|
||||
if escape {
|
||||
match c {
|
||||
'"' => value.push('"'),
|
||||
'\\' => value.push('\\'),
|
||||
'n' => value.push('\n'),
|
||||
't' => value.push('\t'),
|
||||
'r' => value.push('\r'),
|
||||
_ => {}
|
||||
}
|
||||
} else if c == '"' {
|
||||
closed = true;
|
||||
break;
|
||||
} else {
|
||||
value.push(c);
|
||||
}
|
||||
|
||||
escape = false;
|
||||
length += 1;
|
||||
}
|
||||
|
||||
if length >= 2 && closed {
|
||||
return Some(Token {
|
||||
position: 1 + self.cursor + length,
|
||||
value
|
||||
});
|
||||
}
|
||||
|
||||
None
|
||||
}
|
||||
|
||||
fn scan_int(&self) -> Option<Token> {
|
||||
let mut value = String::from("");
|
||||
|
||||
if let Some('-') = self.text.chars().nth(self.cursor) {
|
||||
value.push('-');
|
||||
}
|
||||
|
||||
for c in self.text.chars().skip(self.cursor + value.len()) {
|
||||
if c.is_digit(10) {
|
||||
value.push(c);
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
if value.len() > 0 {
|
||||
Some(Token {
|
||||
position: self.cursor + value.len(),
|
||||
value
|
||||
})
|
||||
} else {
|
||||
None
|
||||
}
|
||||
}
|
||||
|
||||
fn scan_float(&self) -> Option<Token> {
|
||||
let mut value = String::from("");
|
||||
|
||||
if let Some('-') = self.text.chars().nth(self.cursor) {
|
||||
value.push('-');
|
||||
}
|
||||
|
||||
for c in self.text.chars().skip(self.cursor + value.len()) {
|
||||
if c.is_digit(10) {
|
||||
value.push(c);
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
if let Some('.') = self.text.chars().nth(self.cursor + value.len()) {
|
||||
value.push('.');
|
||||
} else {
|
||||
return None;
|
||||
}
|
||||
|
||||
for c in self.text.chars().skip(self.cursor + value.len()) {
|
||||
if c.is_digit(10) {
|
||||
value.push(c);
|
||||
} else {
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
if value.len() > 0 && value != "." {
|
||||
if value.chars().nth(0).unwrap() == '.' {
|
||||
value = format!("0{}", value);
|
||||
}
|
||||
|
||||
if value.chars().last().unwrap() == '.' {
|
||||
value = format!("{}0", value);
|
||||
}
|
||||
|
||||
Some(Token {
|
||||
position: self.cursor + value.len(),
|
||||
value
|
||||
})
|
||||
} else {
|
||||
None
|
||||
}
|
||||
}
|
||||
|
||||
fn scan_text(&self, text: &str) -> Option<Token> {
|
||||
let c = self.cursor;
|
||||
|
||||
|
@ -112,24 +234,43 @@ impl Lexer {
|
|||
}
|
||||
|
||||
impl Iterator for Lexer {
|
||||
type Item = node::Node;
|
||||
type Item = Node;
|
||||
|
||||
fn next(&mut self) -> Option<Self::Item> {
|
||||
self.skip_spaces();
|
||||
|
||||
if let Some(token) = self.scan_keyword("true") {
|
||||
self.cursor = token.position;
|
||||
return Some(node::Node::Bool(token.value == "true"));
|
||||
return Some(Node::Bool(token.value == "true"));
|
||||
}
|
||||
|
||||
if let Some(token) = self.scan_keyword("false") {
|
||||
self.cursor = token.position;
|
||||
return Some(node::Node::Bool(token.value == "true"));
|
||||
return Some(Node::Bool(token.value == "true"));
|
||||
}
|
||||
|
||||
if let Some(token) = self.scan_text(";") {
|
||||
self.cursor = token.position;
|
||||
return Some(node::Node::Semicolon);
|
||||
return Some(Node::Semicolon);
|
||||
}
|
||||
|
||||
if let Some(token) = self.scan_string() {
|
||||
self.cursor = token.position;
|
||||
return Some(Node::String(token.value.clone()));
|
||||
}
|
||||
|
||||
if let Some(token) = self.scan_float() {
|
||||
if let Ok(val) = token.value.parse::<f64>() {
|
||||
self.cursor = token.position;
|
||||
return Some(Node::Float(val));
|
||||
}
|
||||
}
|
||||
|
||||
if let Some(token) = self.scan_int() {
|
||||
if let Ok(val) = token.value.parse::<i32>() {
|
||||
self.cursor = token.position;
|
||||
return Some(Node::Int(val));
|
||||
}
|
||||
}
|
||||
|
||||
None
|
||||
|
@ -140,8 +281,9 @@ impl Iterator for Lexer {
|
|||
mod test {
|
||||
use super::*;
|
||||
use crate::ast::error;
|
||||
use crate::ast::node::Node;
|
||||
|
||||
fn lexer_run(lexer: &mut Lexer, text: &str) -> Result<Vec<node::Node>, error::AstError> {
|
||||
fn lexer_run(lexer: &mut Lexer, text: &str) -> Result<Vec<Node>, error::AstError> {
|
||||
lexer.prepare(text);
|
||||
let result = lexer.collect();
|
||||
|
||||
|
@ -168,8 +310,8 @@ mod test {
|
|||
let mut lex = Lexer::new();
|
||||
let res = lexer_run(&mut lex, "true false").unwrap();
|
||||
assert_eq!(2, res.len());
|
||||
assert_eq!("Bool(true)", format!("{:?}", res.get(0).unwrap()));
|
||||
assert_eq!("Bool(false)", format!("{:?}", res.get(1).unwrap()));
|
||||
assert_eq!(Node::Bool(true), *res.get(0).unwrap());
|
||||
assert_eq!(Node::Bool(false), *res.get(1).unwrap());
|
||||
}
|
||||
|
||||
#[test]
|
||||
|
@ -177,8 +319,8 @@ mod test {
|
|||
let mut lex = Lexer::new();
|
||||
let res = lexer_run(&mut lex, "true;").unwrap();
|
||||
assert_eq!(2, res.len());
|
||||
assert_eq!("Bool(true)", format!("{:?}", res.get(0).unwrap()));
|
||||
assert_eq!("Semicolon", format!("{:?}", res.get(1).unwrap()));
|
||||
assert_eq!(Node::Bool(true), *res.get(0).unwrap());
|
||||
assert_eq!(Node::Semicolon, *res.get(1).unwrap());
|
||||
}
|
||||
|
||||
#[test]
|
||||
|
@ -203,8 +345,47 @@ mod test {
|
|||
let mut lex = Lexer::new();
|
||||
let res = lexer_run(&mut lex, " ;; ")?;
|
||||
assert_eq!(2, res.len());
|
||||
assert_eq!("Semicolon", format!("{:?}", res.get(0).unwrap()));
|
||||
assert_eq!("Semicolon", format!("{:?}", res.get(1).unwrap()));
|
||||
assert_eq!(Node::Semicolon, *res.get(0).unwrap());
|
||||
assert_eq!(Node::Semicolon, *res.get(1).unwrap());
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_integers() -> Result<(), error::AstError> {
|
||||
let mut lex = Lexer::new();
|
||||
let res = lexer_run(&mut lex, " 4 -2 328")?;
|
||||
assert_eq!(3, res.len());
|
||||
assert_eq!(Node::Int(4), *res.get(0).unwrap());
|
||||
assert_eq!(Node::Int(-2), *res.get(1).unwrap());
|
||||
assert_eq!(Node::Int(328), *res.get(2).unwrap());
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_floats() -> Result<(), error::AstError> {
|
||||
let mut lex = Lexer::new();
|
||||
let res = lexer_run(&mut lex, " 2.0 .7 5. -3.14 ")?;
|
||||
assert_eq!(4, res.len());
|
||||
assert_eq!(Node::Float(2.0), *res.get(0).unwrap());
|
||||
assert_eq!(Node::Float(0.7), *res.get(1).unwrap());
|
||||
assert_eq!(Node::Float(5.0), *res.get(2).unwrap());
|
||||
assert_eq!(Node::Float(-3.14), *res.get(3).unwrap());
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn test_string() -> Result<(), error::AstError> {
|
||||
let mut lex = Lexer::new();
|
||||
|
||||
let res = lexer_run(&mut lex,
|
||||
" \"s\\ta\\\"lut\\n\" ")?;
|
||||
|
||||
assert_eq!(1, res.len());
|
||||
assert_eq!(Node::String(String::from("s\ta\"lut\n")),
|
||||
*res.get(0).unwrap());
|
||||
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
|
|
@ -3,6 +3,9 @@
|
|||
pub enum Node {
|
||||
Module(String, Vec<Node>),
|
||||
Bool(bool),
|
||||
Int(i32),
|
||||
Float(f64),
|
||||
String(String),
|
||||
Semicolon
|
||||
}
|
||||
|
||||
|
|
|
@ -1,5 +1,5 @@
|
|||
use crate::ast::{
|
||||
node,
|
||||
node::Node,
|
||||
lexer,
|
||||
error
|
||||
};
|
||||
|
@ -9,7 +9,7 @@ pub struct Parser {
|
|||
lexer: lexer::Lexer
|
||||
}
|
||||
|
||||
type ParseResult = Result<Option<node::Node>, error::AstError>;
|
||||
type ParseResult = Result<Option<Node>, error::AstError>;
|
||||
|
||||
impl Parser {
|
||||
pub fn new(module_name: &str, lex: lexer::Lexer) -> Self {
|
||||
|
@ -24,7 +24,7 @@ impl Parser {
|
|||
}
|
||||
|
||||
fn parse_module(&mut self) -> ParseResult {
|
||||
let mut children: Vec<node::Node> = vec![];
|
||||
let mut children: Vec<Node> = vec![];
|
||||
|
||||
loop {
|
||||
match self.parse_instr() {
|
||||
|
@ -41,7 +41,7 @@ impl Parser {
|
|||
});
|
||||
}
|
||||
|
||||
let n = node::Node::Module(self.module_name.clone(), children);
|
||||
let n = Node::Module(self.module_name.clone(), children);
|
||||
Ok(Some(n))
|
||||
}
|
||||
|
||||
|
@ -49,7 +49,7 @@ impl Parser {
|
|||
match self.parse_expr() {
|
||||
Ok(Some(expr)) => {
|
||||
match self.lexer.next() {
|
||||
Some(node::Node::Semicolon) => Ok(Some(expr)),
|
||||
Some(Node::Semicolon) => Ok(Some(expr)),
|
||||
None => Ok(None),
|
||||
_ => Err(error::AstError {
|
||||
msg: String::from("missing semicolon"),
|
||||
|
@ -68,8 +68,17 @@ impl Parser {
|
|||
|
||||
fn parse_builtin(&mut self) -> ParseResult {
|
||||
match self.lexer.next() {
|
||||
Some(node::Node::Bool(value))
|
||||
=> Ok(Some(node::Node::Bool(value))),
|
||||
Some(Node::Int(value))
|
||||
=> Ok(Some(Node::Int(value))),
|
||||
|
||||
Some(Node::Float(value))
|
||||
=> Ok(Some(Node::Float(value))),
|
||||
|
||||
Some(Node::Bool(value))
|
||||
=> Ok(Some(Node::Bool(value))),
|
||||
|
||||
Some(Node::String(value))
|
||||
=> Ok(Some(Node::String(value))),
|
||||
|
||||
Some(other) => {
|
||||
Err(error::AstError {
|
||||
|
@ -86,11 +95,10 @@ impl Parser {
|
|||
#[cfg(test)]
|
||||
mod test {
|
||||
use super::*;
|
||||
use crate::ast::node::Node;
|
||||
|
||||
type TestResult = Result<(), String>;
|
||||
|
||||
fn test_parser(oracle: node::Node, source: &str) -> TestResult {
|
||||
fn test_parser(oracle: Node, source: &str) -> TestResult {
|
||||
let mut lex = lexer::Lexer::new();
|
||||
lex.prepare(source);
|
||||
let mut parser = Parser::new("mod", lex);
|
||||
|
@ -140,4 +148,35 @@ mod test {
|
|||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn numbers() -> TestResult {
|
||||
test_parser(
|
||||
Node::Module("mod".to_owned(), vec![
|
||||
Node::Float(27.3)
|
||||
]),
|
||||
" 27.3; "
|
||||
)?;
|
||||
|
||||
test_parser(
|
||||
Node::Module("mod".to_owned(), vec![
|
||||
Node::Int(-27)
|
||||
]),
|
||||
" -27; "
|
||||
)?;
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn strings() -> TestResult {
|
||||
test_parser(
|
||||
Node::Module("mod".to_owned(), vec![
|
||||
Node::String(String::from("pizza! "))
|
||||
]),
|
||||
" \"pizza! \"; "
|
||||
)?;
|
||||
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
|
|
@ -29,7 +29,12 @@ impl Evaluator {
|
|||
|
||||
Ok(res)
|
||||
},
|
||||
|
||||
Node::Bool(value) => Ok(Some(Value::Bool(*value))),
|
||||
Node::Int(value) => Ok(Some(Value::Int(*value))),
|
||||
Node::Float(value) => Ok(Some(Value::Float(*value))),
|
||||
Node::String(value) => Ok(Some(Value::String(value.clone()))),
|
||||
|
||||
_ => Ok(None)
|
||||
}
|
||||
}
|
||||
|
@ -65,7 +70,10 @@ mod test {
|
|||
Ok(())
|
||||
}
|
||||
} else {
|
||||
Ok(())
|
||||
Err(Box::new(EvalError {
|
||||
msg: format!("unexpected: {:?}", value),
|
||||
line: 0
|
||||
}))
|
||||
}
|
||||
},
|
||||
None => Ok(())
|
||||
|
@ -79,4 +87,20 @@ mod test {
|
|||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn numbers() -> TestResult {
|
||||
test_eval_value(Value::Float(-271.4), " -271.4; ")?;
|
||||
test_eval_value(Value::Int(333), " 333; ")?;
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn strings() -> TestResult {
|
||||
test_eval_value(Value::String(String::from(" pizzza!!!")),
|
||||
" \" pizzza!!!\"; ")?;
|
||||
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
|
|
@ -1,6 +1,9 @@
|
|||
#[derive(PartialEq)]
|
||||
#[derive(Debug)]
|
||||
pub enum Value {
|
||||
Bool(bool)
|
||||
Bool(bool),
|
||||
Int(i32),
|
||||
Float(f64),
|
||||
String(String)
|
||||
}
|
||||
|
||||
|
|
Reference in New Issue