mu

rust/src/scanner.rs

@@ -4,8 +4,8 @@ pub struct Scanner<'a> {
     line: usize,
 }
 
-fn is_alpha(c: &char) -> bool {
-    c.is_ascii_alphabetic() || *c == '_'
+fn is_alpha(c: char) -> bool {
+    c.is_ascii_alphabetic() || c == '_'
 }
 
 impl<'a> Scanner<'a> {
@@ -17,32 +17,44 @@ impl<'a> Scanner<'a> {
         }
     }
 
-    fn advance(&mut self) -> Option<char> {
+    fn advance(&mut self) -> char {
         self.current += 1;
-        self.source.get(self.current - 1).map(|&x| x as char)
+        self.source[self.current - 1] as char
     }
 
-    fn advance_if<F: Fn(&char) -> bool>(&mut self, f: F) -> bool {
-        if matches!(self.peek(), Some(c) if f(&c)) {
-            self.current += 1;
-            true
-        } else {
-            false
+    fn is_match(&mut self, expected: char) -> bool {
+        if self.peek() != Some(expected) {
+            return false;
         }
+
+        self.current += 1;
+        true
     }
 
     fn skip_whitespace(&mut self) {
         loop {
-            while self.advance_if(|&c| c == ' ' || c == '\r' || c == '\t') {}
-
-            if self.peek() == Some('/') && self.peek_next() == Some('/') {
-                while self.advance_if(|&c| c != '\n') {}
-            }
-
-            if self.advance_if(|&c| c == '\n') {
-                self.line += 1;
-            } else {
-                return;
+            match self.peek() {
+                Some(' ') | Some('\r') | Some('\t') => {
+                    self.advance();
+                    break;
+                }
+                Some('\n') => {
+                    self.line += 1;
+                    self.advance();
+                    break;
+                }
+                Some('/') => {
+                    if matches!(self.peek_next(), Some(c) if c == '/') {
+                        while matches!(self.peek(), Some(c) if c != '\n') {
+                            self.advance();
+                        }
+                    } else {
+                        return;
+                    }
+                }
+                _ => {
+                    return;
+                }
             }
         }
     }
@@ -56,36 +68,47 @@ impl<'a> Scanner<'a> {
     }
 
     fn string(&mut self) -> Token<'a> {
-        loop {
-            match self.advance() {
-                Some('"') => break,
-                Some('\n') => {
-                    self.line += 1;
-                }
-                Some(_) => {}
-                None => return Token::error(self, "Unterminated string."),
+        while matches!(self.peek(), Some(c) if c != '"') {
+            if self.peek() == Some('\n') {
+                self.line += 1;
             }
+            self.advance();
+        }
+
+        if self.peek().is_none() {
+            // Maybe this should return an Err?
+            return Token::error(&self, "Unterminated string.");
         }
 
-        Token::new(self, TokenKind::String)
+        self.advance();
+        // Token::new(&self, TokenKind::String)
+        self.make_token(TokenKind::String)
     }
 
     fn number(&mut self) -> Token<'a> {
-        while self.advance_if(char::is_ascii_digit) {}
+        while matches!(self.peek(), Some(c) if c.is_ascii_digit()) {
+            self.advance();
+        }
 
         if self.peek() == Some('.') && matches!(self.peek_next(), Some(c) if c.is_ascii_digit()) {
             self.advance();
 
-            while self.advance_if(char::is_ascii_digit) {}
+            while matches!(self.peek(), Some(c) if c.is_ascii_digit()) {
+                self.advance();
+            }
         }
 
-        Token::new(self, TokenKind::Number)
+        // Token::new(&self, TokenKind::Number)
+        self.make_token(TokenKind::Number)
     }
 
     fn identifier(&mut self) -> Token<'a> {
-        while self.advance_if(|c| is_alpha(c) || c.is_ascii_digit()) {}
+        while matches!(self.peek(), Some(c) if is_alpha(c) || c.is_ascii_digit()) {
+            self.advance();
+        }
 
-        Token::new(self, self.identifier_type())
+        // Token::new(&self, self.identifier_type())
+        self.make_token(self.identifier_type())
     }
 
     fn identifier_type(&self) -> TokenKind {
@@ -115,6 +138,14 @@ impl<'a> Scanner<'a> {
             _ => TokenKind::Identifier,
         }
     }
+
+    fn make_token(&self, kind: TokenKind) -> Token<'a> {
+        Token {
+            kind: TokenKind::And,
+            value: std::str::from_utf8(&self.source[..self.current]).unwrap(),
+            line: self.line,
+        }
+    }
 }
 
 impl<'a> Iterator for Scanner<'a> {
@@ -122,60 +153,54 @@ impl<'a> Iterator for Scanner<'a> {
 
     fn next(&mut self) -> Option<Self::Item> {
         self.skip_whitespace();
-
-        self.advance()
-            .map(|c| match c {
-                c if is_alpha(&c) => self.identifier(),
-                c if c.is_ascii_digit() => self.number(),
-                '(' => Token::new(self, TokenKind::LeftParen),
-                ')' => Token::new(self, TokenKind::RightParen),
-                '{' => Token::new(self, TokenKind::LeftBrace),
-                '}' => Token::new(self, TokenKind::RightBrace),
-                ';' => Token::new(self, TokenKind::Semicolon),
-                ',' => Token::new(self, TokenKind::Comma),
-                '.' => Token::new(self, TokenKind::Dot),
-                '-' => Token::new(self, TokenKind::Minus),
-                '+' => Token::new(self, TokenKind::Plus),
-                '/' => Token::new(self, TokenKind::Slash),
-                '*' => Token::new(self, TokenKind::Star),
-                '!' => {
-                    if self.advance_if(|&c| c == '=') {
-                        Token::new(self, TokenKind::BangEqual)
-                    } else {
-                        Token::new(self, TokenKind::Bang)
-                    }
-                }
-                '=' => {
-                    if self.advance_if(|&c| c == '=') {
-                        Token::new(self, TokenKind::EqualEqual)
-                    } else {
-                        Token::new(self, TokenKind::Equal)
-                    }
-                }
-                '<' => {
-                    if self.advance_if(|&c| c == '=') {
-                        Token::new(self, TokenKind::LessEqual)
-                    } else {
-                        Token::new(self, TokenKind::Less)
-                    }
-                }
-                '>' => {
-                    if self.advance_if(|&c| c == '=') {
-                        Token::new(self, TokenKind::GreaterEqual)
-                    } else {
-                        Token::new(self, TokenKind::Greater)
-                    }
-                }
-                '"' => self.string(),
-                _ => Token::error(self, "Unexpected character."),
-            })
-            .or({
-                Some(Token {
-                    kind: TokenKind::Eof,
-                    value: "",
-                    line: self.line,
-                })
-            })
+        self.peek()?;
+
+        Some(match self.advance() {
+            c if is_alpha(c) => self.identifier(),
+            c if c.is_ascii_digit() => self.number(),
+            '(' => Token::new(self, TokenKind::LeftParen),
+            // '(' => self.make_token(TokenKind::LeftParen),
+        //     ')' => self.make_token(TokenKind::RightParen),
+        //     '{' => self.make_token(TokenKind::LeftBrace),
+        //     '}' => self.make_token(TokenKind::RightBrace),
+        //     ';' => self.make_token(TokenKind::Semicolon),
+        //     ',' => self.make_token(TokenKind::Comma),
+        //     '.' => self.make_token(TokenKind::Dot),
+        //     '-' => self.make_token(TokenKind::Minus),
+        //     '+' => self.make_token(TokenKind::Plus),
+        //     '/' => self.make_token(TokenKind::Slash),
+        //     '*' => self.make_token(TokenKind::Star),
+        //     '!' => {
+        //         if self.is_match('=') {
+        //             self.make_token(TokenKind::BangEqual)
+        //         } else {
+        //             self.make_token(TokenKind::Bang)
+        //         }
+        //     }
+        //     '=' => {
+        //         if self.is_match('=') {
+        //             self.make_token(TokenKind::EqualEqual)
+        //         } else {
+        //             self.make_token(TokenKind::Equal)
+        //         }
+        //     }
+        //     '<' => {
+        //         if self.is_match('=') {
+        //             self.make_token(TokenKind::LessEqual)
+        //         } else {
+        //             self.make_token(TokenKind::Less)
+        //         }
+        //     }
+        //     '>' => {
+        //         if self.is_match('=') {
+        //             self.make_token(TokenKind::GreaterEqual)
+        //         } else {
+        //             self.make_token(TokenKind::Greater)
+        //         }
+        //     }
+        //     '"' => self.string(),
+            _ => Token::error(self, "Unexpected character."),
+        })
     }
 }
 
@@ -282,7 +307,7 @@ mod tests {
         assert_source_scan("fa", Some(TokenKind::Identifier));
 
         assert_source_scan("@", Some(TokenKind::Error));
-        assert_source_scan("", Some(TokenKind::Eof));
+        assert_source_scan("", None);
     }
 
     #[test]
@@ -290,14 +315,14 @@ mod tests {
         let mut scanner = Scanner::new("()");
         assert_scan(&mut scanner, Some(TokenKind::LeftParen));
         assert_scan(&mut scanner, Some(TokenKind::RightParen));
-        assert_scan(&mut scanner, Some(TokenKind::Eof));
+        assert_scan(&mut scanner, None);
     }
 
     #[test]
     fn test_whitespace() {
         assert_source_scan(" foo", Some(TokenKind::Identifier));
         assert_source_scan("\tfoo", Some(TokenKind::Identifier));
-        assert_source_scan("//   \n", Some(TokenKind::Eof));
+        assert_source_scan("//   \n", None);
     }
 
     fn assert_source_scan(source: &str, kind: Option<TokenKind>) {