Very basic tokenizer implementation

src/lib.rs

@@ -1,3 +1,6 @@
+use std::fmt;
+
+#[derive(Debug)]
 pub enum Token {
 	OpenParen,
 	CloseParen,
@@ -6,16 +9,143 @@ pub enum Token {
 	Integer(i32)
 }
 
-pub trait Tokenizable {
+
-	fn tokenize(&self) -> Vec<Token>;
+pub struct TokenizerError {
+	line: u64,
+	column: u64,
+
+	line_end: u64,
+	column_end: u64,
+
+	message: String,
 }
 
-impl Tokenizable for String {
+impl fmt::Display for TokenizerError {
-	fn tokenize(&self) -> Vec<Token> {
+	fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
-		let tokens = Vec::new();
+		if self.line != self.line_end || self.column != self.column_end {
-
+			return write!(f, "Error from {}:{} to {}:{}, '{}'", self.line, self.column, self.line_end, self.column_end, self.message);
-
+		}
-
+		write!(f, "Error at {}:{}, '{}'", self.line, self.column, self.message)
-		tokens
+	}
+}
+
+pub struct Tokenizer {
+	code: String,
+	line: u64,
+	column: u64,
+
+	reading_string: bool,
+	escape_next_char: bool,
+
+	reading_identifier: bool,
+	skipping_comment: bool,
+
+	storage: Vec<char>,
+}
+
+impl Tokenizer {
+	pub fn new(code: String) -> Tokenizer {
+		Self {
+			code,
+			line: 1,
+			column: 1,
+
+			reading_string: false,
+			escape_next_char: false,
+
+			reading_identifier: false,
+			skipping_comment: false,
+
+			storage: Vec::new()
+		}
+	}
+
+	pub fn tokenize(&mut self) -> Result<Vec<Token>, TokenizerError> {
+		let mut tokens = Vec::new();
+
+		for char in self.code.chars() {
+			let line = self.line;
+			let column = self.column;
+
+			self.line += 1;
+			if char == '\n' {
+				self.column += 1;
+				self.line = 1;
+			}
+
+			if self.skipping_comment {
+				match char {
+					'\n' => {
+						self.skipping_comment = false;
+					},
+					_ => continue
+				}
+			}
+
+			if self.reading_identifier {
+				if !char.is_alphabetic() {
+					self.reading_identifier = false;
+
+					tokens.push(Token::Identifier(self.storage.iter().collect()));
+					self.storage.clear();
+				}
+
+				self.storage.push(char);
+				continue;
+			}
+
+			if self.reading_string {
+				if char == '"' && !self.escape_next_char {
+					self.reading_string = false;
+
+					tokens.push(Token::String(self.storage.iter().collect()));
+					self.storage.clear();
+					continue;
+				}
+
+				if char == '\n' {
+					return Err(TokenizerError {
+						line: line - self.storage.len() as u64 - 1,
+						column,
+
+						line_end: line,
+						column_end: column,
+
+						message: String::from("Expected \", got \\n"),
+					})
+				}
+
+				if char == '\\' {
+					self.escape_next_char = true;
+					continue;
+				}
+
+				self.storage.push(char);
+
+				continue;
+			}
+
+			match char {
+				';' => self.skipping_comment = true,
+				'(' => tokens.push(Token::OpenParen),
+				')' => tokens.push(Token::CloseParen),
+				'"' => {
+					self.reading_string = true;
+					self.storage.clear();
+				},
+				c => {
+					if c.is_alphabetic() {
+						self.reading_identifier = true;
+						self.storage.clear();
+						self.storage.push(c);
+						continue;
+					}
+
+					print!("{}", char)
+				}
+			}
+		}
+
+		Ok(tokens)
 	}
 }

src/main.rs

@@ -1,4 +1,14 @@
+use lisp_stuff::*;
+
 fn main() {
     let source = std::fs::read_to_string("src/test.lisp").unwrap();
-    println!("{}", source);
+    let mut tokenizer = Tokenizer::new(source);
+	let tokens = match tokenizer.tokenize() {
+		Ok(tokens) => tokens,
+		Err(e) => {
+			println!("{}", e);
+			Vec::new()
+		}
+	};
+	println!("{:?}", tokens);
 }

src/test.lisp

@@ -1,6 +1,4 @@
 ; This is a comment
-(print "Hello, World")
-(print (add 1 2))
 
-; OpenParen, Identifier, String, CloseParen
+(print "Hello, World") ; OpenParen, Identifier, String, CloseParen
-; OpenParen, Identifier, OpenParen, Identifier, Int, Int, CloseParen, CloseParen
+; (print (add 1 2)) ; OpenParen, Identifier, OpenParen, Identifier, Int, Int, CloseParen, CloseParen