From 8f20e538807de6481014b8cb583c2385ffeafafc Mon Sep 17 00:00:00 2001
From: Jeff <dev@jeffa.io>
Date: Mon, 9 Sep 2024 23:45:06 -0400
Subject: [PATCH] Add support for characters and bytes

---
 dust-lang/src/lexer.rs                     | 16 +++++++++++
 dust-lang/src/parser.rs                    | 32 +++++++++++++++++++++-
 dust-lang/src/token.rs                     |  9 ++++++
 dust-lang/tests/{math.rs => operations.rs} |  0
 4 files changed, 56 insertions(+), 1 deletion(-)
 rename dust-lang/tests/{math.rs => operations.rs} (100%)

diff --git a/dust-lang/src/lexer.rs b/dust-lang/src/lexer.rs
index 557a6e2..06c09c9 100644
--- a/dust-lang/src/lexer.rs
+++ b/dust-lang/src/lexer.rs
@@ -424,6 +424,22 @@ impl<'src> Lexer<'src> {
                 }
             }
 
+            if c == 'x' {
+                self.next_char();
+
+                while let Some(c) = self.peek_char() {
+                    if c.is_ascii_hexdigit() {
+                        self.next_char();
+                    } else {
+                        break;
+                    }
+                }
+
+                let text = &self.source[start_pos..self.position];
+
+                return Ok((Token::Byte(text), Span(start_pos, self.position)));
+            }
+
             if c.is_ascii_digit() {
                 self.next_char();
             } else {
diff --git a/dust-lang/src/parser.rs b/dust-lang/src/parser.rs
index df83045..f97005b 100644
--- a/dust-lang/src/parser.rs
+++ b/dust-lang/src/parser.rs
@@ -115,6 +115,27 @@ impl<'src> Parser<'src> {
         Ok(())
     }
 
+    fn parse_byte(&mut self, _allow_assignment: bool) -> Result<(), ParseError> {
+        if let Token::Byte(text) = self.previous_token {
+            let byte = u8::from_str_radix(&text[2..], 16)?;
+            let value = Value::byte(byte);
+
+            self.emit_constant(value)?;
+        }
+
+        Ok(())
+    }
+
+    fn parse_character(&mut self, _allow_assignment: bool) -> Result<(), ParseError> {
+        if let Token::Character(character) = self.previous_token {
+            let value = Value::character(character);
+
+            self.emit_constant(value)?;
+        }
+
+        Ok(())
+    }
+
     fn parse_float(&mut self, _allow_assignment: bool) -> Result<(), ParseError> {
         if let Token::Float(text) = self.previous_token {
             let float = text.parse::<f64>().unwrap();
@@ -418,7 +439,16 @@ impl From<&TokenKind> for ParseRule<'_> {
                 infix: None,
                 precedence: Precedence::None,
             },
-            TokenKind::Character => todo!(),
+            TokenKind::Byte => ParseRule {
+                prefix: Some(Parser::parse_byte),
+                infix: None,
+                precedence: Precedence::None,
+            },
+            TokenKind::Character => ParseRule {
+                prefix: Some(Parser::parse_character),
+                infix: None,
+                precedence: Precedence::None,
+            },
             TokenKind::Float => ParseRule {
                 prefix: Some(Parser::parse_float),
                 infix: None,
diff --git a/dust-lang/src/token.rs b/dust-lang/src/token.rs
index 2f2ce04..a0d5fe1 100644
--- a/dust-lang/src/token.rs
+++ b/dust-lang/src/token.rs
@@ -11,6 +11,7 @@ pub enum Token<'src> {
 
     // Hard-coded values
     Boolean(&'src str),
+    Byte(&'src str),
     Character(char),
     Float(&'src str),
     Identifier(&'src str),
@@ -70,6 +71,7 @@ impl<'src> Token<'src> {
         match self {
             Token::Eof => 0,
             Token::Boolean(text) => text.len(),
+            Token::Byte(_) => 3,
             Token::Character(_) => 3,
             Token::Float(text) => text.len(),
             Token::Identifier(text) => text.len(),
@@ -128,6 +130,7 @@ impl<'src> Token<'src> {
             Token::Bool => TokenOwned::Bool,
             Token::Boolean(boolean) => TokenOwned::Boolean(boolean.to_string()),
             Token::Break => TokenOwned::Break,
+            Token::Byte(byte) => TokenOwned::Byte(byte.to_string()),
             Token::Character(character) => TokenOwned::Character(*character),
             Token::Colon => TokenOwned::Colon,
             Token::Comma => TokenOwned::Comma,
@@ -182,6 +185,7 @@ impl<'src> Token<'src> {
             Token::Bool => TokenKind::Bool,
             Token::Boolean(_) => TokenKind::Boolean,
             Token::Break => TokenKind::Break,
+            Token::Byte(_) => TokenKind::Byte,
             Token::Character(_) => TokenKind::Character,
             Token::Colon => TokenKind::Colon,
             Token::Comma => TokenKind::Comma,
@@ -238,6 +242,7 @@ impl<'src> Display for Token<'src> {
             Token::Bool => write!(f, "bool"),
             Token::Boolean(value) => write!(f, "{}", value),
             Token::Break => write!(f, "break"),
+            Token::Byte(value) => write!(f, "{}", value),
             Token::Character(value) => write!(f, "'{}'", value),
             Token::Colon => write!(f, ":"),
             Token::Comma => write!(f, ","),
@@ -296,6 +301,7 @@ pub enum TokenOwned {
 
     // Hard-coded values
     Boolean(String),
+    Byte(String),
     Character(char),
     Float(String),
     Integer(String),
@@ -357,6 +363,7 @@ impl Display for TokenOwned {
             TokenOwned::Bool => Token::Bool.fmt(f),
             TokenOwned::Boolean(boolean) => Token::Boolean(boolean).fmt(f),
             TokenOwned::Break => Token::Break.fmt(f),
+            TokenOwned::Byte(byte) => Token::Byte(byte).fmt(f),
             TokenOwned::Character(character) => Token::Character(*character).fmt(f),
             TokenOwned::Colon => Token::Colon.fmt(f),
             TokenOwned::Comma => Token::Comma.fmt(f),
@@ -413,6 +420,7 @@ pub enum TokenKind {
 
     // Hard-coded values
     Boolean,
+    Byte,
     Character,
     Float,
     Integer,
@@ -474,6 +482,7 @@ impl Display for TokenKind {
             TokenKind::Bool => Token::Bool.fmt(f),
             TokenKind::Boolean => write!(f, "boolean value"),
             TokenKind::Break => Token::Break.fmt(f),
+            TokenKind::Byte => write!(f, "byte value"),
             TokenKind::Character => write!(f, "character value"),
             TokenKind::Colon => Token::Colon.fmt(f),
             TokenKind::Comma => Token::Comma.fmt(f),
diff --git a/dust-lang/tests/math.rs b/dust-lang/tests/operations.rs
similarity index 100%
rename from dust-lang/tests/math.rs
rename to dust-lang/tests/operations.rs