Implement string interning

src/ast.rs

@@ -1,4 +1,4 @@
-use std::rc::Rc;
+use crate::stringstore::{StringStore, Sid};
 
 /// Types for binary operators
 #[derive(Debug, PartialEq, Eq, Clone)]
@@ -81,9 +81,9 @@ pub enum Expression {
     /// Integer literal (64-bit)
     I64(i64),
     /// String literal
-    String(Rc<String>),
+    String(Sid),
     /// Variable
-    Var(String),
+    Var(Sid),
     /// Binary operation. Consists of type, left hand side and right hand side
     BinOp(BinOpType, Box<Expression>, Box<Expression>),
     /// Unary operation. Consists of type and operand
@@ -120,6 +120,12 @@ pub enum Statement {
 
 pub type BlockScope = Vec<Statement>;
 
+#[derive(Clone, Default)]
+pub struct Ast {
+    pub stringstore: StringStore,
+    pub main: BlockScope,
+}
+
 impl BinOpType {
     /// Get the precedence for a binary operator. Higher value means the OP is stronger binding.
     /// For example Multiplication is stronger than addition, so Mul has higher precedence than Add.

src/interpreter.rs

@@ -1,15 +1,13 @@
-use std::{fmt::Display, rc::Rc};
-
 use crate::{
     ast::{BlockScope, BinOpType, Expression, If, Statement, UnOpType},
     lexer::lex,
-    parser::parse,
+    parser::parse, stringstore::{Sid, StringStore},
 };
 
 #[derive(Debug, PartialEq, Eq, Clone)]
 pub enum Value {
     I64(i64),
-    String(Rc<String>),
+    String(Sid),
 }
 
 #[derive(Default)]
@@ -17,7 +15,9 @@ pub struct Interpreter {
     capture_output: bool,
     output: Vec<Value>,
     // Variable table stores the runtime values of variables
-    vartable: Vec<(String, Value)>,
+    vartable: Vec<(Sid, Value)>,
+
+    stringstore: StringStore,
 }
 
 impl Interpreter {
@@ -33,7 +33,7 @@ impl Interpreter {
         &self.output
     }
 
-    fn get_var(&self, name: &str) -> Option<Value> {
+    fn get_var(&self, name: Sid) -> Option<Value> {
         self.vartable
             .iter()
             .rev()
@@ -41,7 +41,7 @@ impl Interpreter {
             .map(|it| it.1.clone())
     }
 
-    fn get_var_mut(&mut self, name: &str) -> Option<&mut Value> {
+    fn get_var_mut(&mut self, name: Sid) -> Option<&mut Value> {
         self.vartable
             .iter_mut()
             .rev()
@@ -57,13 +57,15 @@ impl Interpreter {
 
         let ast = parse(tokens);
         if print_ast {
-            println!("{:#?}", ast);
+            println!("{:#?}", ast.main);
         }
 
-        self.run(&ast);
+        self.stringstore = ast.stringstore;
+
+        self.run_block(&ast.main);
     }
 
-    pub fn run(&mut self, prog: &BlockScope) {
+    pub fn run_block(&mut self, prog: &BlockScope) {
         let vartable_len = self.vartable.len();
         for stmt in prog {
             match stmt {
@@ -78,7 +80,7 @@ impl Interpreter {
                             break;
                         }
 
-                        self.run(&looop.body);
+                        self.run_block(&looop.body);
 
                         if let Some(adv) = &looop.advancement {
                             self.resolve_expr(&adv);
@@ -92,7 +94,7 @@ impl Interpreter {
                     if self.capture_output {
                         self.output.push(result)
                     } else {
-                        print!("{}", result);
+                        self.print_value(&result);
                     }
                 }
 
@@ -102,9 +104,9 @@ impl Interpreter {
                     body_false,
                 }) => {
                     if matches!(self.resolve_expr(condition), Value::I64(0)) {
-                        self.run(body_false);
+                        self.run_block(body_false);
                     } else {
-                        self.run(body_true);
+                        self.run_block(body_true);
                     }
                 }
             }
@@ -119,14 +121,14 @@ impl Interpreter {
             Expression::String(text) => Value::String(text.clone()),
             Expression::BinOp(bo, lhs, rhs) => self.resolve_binop(bo, lhs, rhs),
             Expression::UnOp(uo, operand) => self.resolve_unop(uo, operand),
-            Expression::Var(name) => self.resolve_var(name),
+            Expression::Var(name) => self.resolve_var(*name),
         }
     }
 
-    fn resolve_var(&mut self, name: &str) -> Value {
+    fn resolve_var(&mut self, name: Sid) -> Value {
         match self.get_var(name) {
             Some(val) => val.clone(),
-            None => panic!("Variable '{}' used but not declared", name),
+            None => panic!("Variable '{}' used but not declared", self.stringstore.lookup(name).unwrap()),
         }
     }
 
@@ -150,7 +152,7 @@ impl Interpreter {
                 return rhs;
             }
             (BinOpType::Assign, Expression::Var(name)) => {
-                match self.get_var_mut(name) {
+                match self.get_var_mut(*name) {
                     Some(val) => *val = rhs.clone(),
                     None => panic!("Runtime Error: Trying to assign value to undeclared variable"),
                 }
@@ -187,15 +189,14 @@ impl Interpreter {
             _ => panic!("Value types are not compatible"),
         }
     }
-}
 
-impl Display for Value {
-    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
-        match self {
-            Value::I64(val) => write!(f, "{}", val),
-            Value::String(text) => write!(f, "{}", text),
+    fn print_value(&self, val: &Value) {
+        match val {
+            Value::I64(val) => print!("{}", val),
+            Value::String(text) => print!("{}", self.stringstore.lookup(*text).unwrap()),
         }
     }
+
 }
 
 #[cfg(test)]

src/lib.rs

@@ -3,6 +3,7 @@ pub mod interpreter;
 pub mod lexer;
 pub mod parser;
 pub mod token;
+pub mod stringstore;
 
 #[cfg(test)]
 mod tests {
@@ -94,3 +95,4 @@ mod tests {
         assert_eq!(interpreter.output(), &expected_output);
     }
 }
+

src/parser.rs

@@ -1,28 +1,42 @@
 use std::iter::Peekable;
 
 use crate::ast::*;
+use crate::stringstore::StringStore;
 use crate::token::Token;
 
 /// Parse the given tokens into an abstract syntax tree
-pub fn parse<T: Iterator<Item = Token>, A: IntoIterator<IntoIter = T>>(tokens: A) -> BlockScope {
-    let mut parser = Parser::new(tokens);
+pub fn parse<T: Iterator<Item = Token>, A: IntoIterator<IntoIter = T>>(tokens: A) -> Ast {
+    let parser = Parser::new(tokens);
     parser.parse()
 }
 
 struct Parser<T: Iterator<Item = Token>> {
     tokens: Peekable<T>,
+    stringstore: StringStore,
 }
 
 impl<T: Iterator<Item = Token>> Parser<T> {
     /// Create a new parser to parse the given Token Stream
-    fn new<A: IntoIterator<IntoIter = T>>(tokens: A) -> Self {
+    pub fn new<A: IntoIterator<IntoIter = T>>(tokens: A) -> Self {
         let tokens = tokens.into_iter().peekable();
-        Self { tokens }
+        let stringstore = StringStore::new();
+        Self {
+            tokens,
+            stringstore,
+        }
+    }
+
+    pub fn parse(mut self) -> Ast {
+        let main = self.parse_scoped_block();
+        Ast {
+            main,
+            stringstore: self.stringstore,
+        }
     }
 
     /// Parse tokens into an abstract syntax tree. This will continuously parse statements until
     /// encountering end-of-file or a block end '}' .
-    fn parse(&mut self) -> BlockScope {
+    fn parse_scoped_block(&mut self) -> BlockScope {
         let mut prog = Vec::new();
 
         loop {
@@ -86,7 +100,7 @@ impl<T: Iterator<Item = Token>> Parser<T> {
             panic!("Error lexing if: Expected '{{'")
         }
 
-        let body_true = self.parse();
+        let body_true = self.parse_scoped_block();
 
         if !matches!(self.next(), Token::RBraces) {
             panic!("Error lexing if: Expected '}}'")
@@ -101,7 +115,7 @@ impl<T: Iterator<Item = Token>> Parser<T> {
                 panic!("Error lexing if: Expected '{{'")
             }
 
-            body_false = self.parse();
+            body_false = self.parse_scoped_block();
 
             if !matches!(self.next(), Token::RBraces) {
                 panic!("Error lexing if: Expected '}}'")
@@ -128,7 +142,7 @@ impl<T: Iterator<Item = Token>> Parser<T> {
 
         match self.next() {
             Token::LBraces => {
-                body = self.parse();
+                body = self.parse_scoped_block();
             }
 
             Token::Semicolon => {
@@ -138,7 +152,7 @@ impl<T: Iterator<Item = Token>> Parser<T> {
                     panic!("Error lexing loop: Expected '{{'")
                 }
 
-                body = self.parse();
+                body = self.parse_scoped_block();
             }
 
             _ => panic!("Error lexing loop: Expected ';' or '{{'"),
@@ -196,9 +210,9 @@ impl<T: Iterator<Item = Token>> Parser<T> {
             Token::I64(val) => Expression::I64(val),
 
             // Literal String
-            Token::String(text) => Expression::String(text.into()),
+            Token::String(text) => Expression::String(self.stringstore.intern_or_lookup(&text)),
 
-            Token::Ident(name) => Expression::Var(name),
+            Token::Ident(name) => Expression::Var(self.stringstore.intern_or_lookup(&name)),
 
             // Parentheses grouping
             Token::LParen => {
@@ -248,10 +262,7 @@ impl<T: Iterator<Item = Token>> Parser<T> {
 #[cfg(test)]
 mod tests {
     use super::{parse, BinOpType, Expression};
-    use crate::{
-        parser::Statement,
-        token::Token,
-    };
+    use crate::{parser::Statement, token::Token};
 
     #[test]
     fn test_parser() {
@@ -287,6 +298,6 @@ mod tests {
         let expected = vec![expected];
 
         let actual = parse(tokens);
-        assert_eq!(expected, actual);
+        assert_eq!(expected, actual.main);
     }
 }

src/stringstore.rs

@@ -0,0 +1,31 @@
+use std::collections::HashMap;
+
+#[derive(Debug, Clone, Copy, PartialEq, Eq, PartialOrd, Ord)]
+pub struct Sid(usize);
+
+#[derive(Clone, Default)]
+pub struct StringStore {
+    strings: Vec<String>,
+    sids: HashMap<String, Sid>,
+}
+
+impl StringStore {
+
+    pub fn new() -> Self {
+        Self { strings: Vec::new(), sids: HashMap::new() }
+    }
+
+    pub fn intern_or_lookup(&mut self, text: &str) -> Sid {
+        self.sids.get(text).copied().unwrap_or_else(|| {
+            let sid = Sid(self.strings.len());
+            self.strings.push(text.to_string());
+            self.sids.insert(text.to_string(), sid);
+            sid
+        })
+    }
+
+    pub fn lookup(&self, sid: Sid) -> Option<&String> {
+        self.strings.get(sid.0)
+    }
+
+}