Merge pull request 'Merge compiler and emulator progress from last few months into main.' (#11) from compiler into main

Reviewed-on: #11

.gitignore

@@ -1,3 +1,3 @@
 /target
 **/*.env
-book
+Cargo.lock

.zed/settings.json

@@ -0,0 +1,15 @@
+// Folder-specific settings
+//
+// For a full list of overridable settings, and general information on folder-specific settings,
+// see the documentation: https://zed.dev/docs/configuring-zed#settings-files
+{
+  "lsp": {
+    "rust-analyzer": {
+      "initialization_options": {
+        "check": {
+          "command": "clippy", // rust-analyzer.check.command (default: "check")
+        },
+      },
+    },
+  },
+}

.zed/tasks.json

@@ -0,0 +1,37 @@
+[
+  {
+    "label": "Run Emulator",
+    "command": "cargo run --bin emulator",
+    "use_new_terminal": true,
+  },
+  {
+    "label": "Run Compiler",
+    "command": "cargo run --bin compiler",
+    "use_new_terminal": true,
+  },
+  {
+    "label": "Run Assembler",
+    "command": "cargo run --bin assembler",
+    "use_new_terminal": true,
+  },
+  {
+    "label": "Run Build System (dsx-build)",
+    "command": "cargo run --bin dsx-build",
+    "use_new_terminal": true,
+  },
+  {
+    "label": "Build All (Release)",
+    "command": "cargo build --release",
+    "use_new_terminal": false,
+  },
+  {
+    "label": "Run Tests",
+    "command": "cargo test",
+    "use_new_terminal": true,
+  },
+  {
+    "label": "Profile Emulator with perf",
+    "command": "cargo build --profile profiling; perf record -g -F 999 target/profiling/emulator; perf script -F +pid | save test.perf",
+    "use_new_terminal": true,
+  },
+]

Cargo.toml

@@ -1,7 +1,7 @@
 cargo-features = ["codegen-backend"]
 
 [workspace]
-members = ["emulator", "common", "assembler", "dsa_editor"]
+members = ["emulator", "common", "assembler", "dsa_editor", "compiler", "dsx-build"]
 resolver = "3"
 
 [workspace.package]

assembler/Cargo.toml

@@ -5,7 +5,7 @@ edition.workspace = true
 authors.workspace = true
 
 [[bin]]
-name = "assembler_runner"
+name = "assembler"
 path = "src/main.rs"
 
 [lib]

assembler/src/assembler/codegen.rs

@@ -14,8 +14,7 @@ pub fn codegen(nodes: Vec<Node>) -> Result<Vec<Instruction>, AssembleError> {
         instructions.push(build_instruction(&node)?);
     }
 
-    println!("------------------------");
-    log("Compilation Success ✅");
+    log("Assembly Successful ✅");
 
     Ok(instructions)
 }
@@ -224,19 +223,31 @@ fn build_shift_instruction(
     opcode: Opcode,
     args: &[crate::assembler::model::Token],
 ) -> Result<Instruction, AssembleError> {
-    let Some(reg_token) = args.first() else {
+    let Some(src_reg) = args.first() else {
         return Err(AssembleError::MissingArgument(0));
     };
-    let Some(amount_token) = args.get(1) else {
+    let Some(r_shamt) = args.get(1) else {
+        return Err(AssembleError::MissingArgument(0));
+    };
+    let Some(i_shamt) = args.get(2) else {
+        return Err(AssembleError::MissingArgument(1));
+    };
+    let Some(dest_reg) = args.get(3) else {
         return Err(AssembleError::MissingArgument(1));
     };
 
-    let reg = expect_token!(reg_token, Register)?;
-    let amount = expect_token!(amount_token, Immediate)? as u8;
+    let src = expect_token!(src_reg, Register)?;
+    let r_shamt = expect_token!(r_shamt, Register)?;
+    let i_shamt = expect_token!(i_shamt, Immediate)? as u8;
+    let dest = expect_token!(dest_reg, Register)?;
 
     match opcode {
-        Opcode::Shl => Ok(Instruction::ShiftLeft(args!(R, sr1: reg, shamt: amount))),
-        Opcode::Shr => Ok(Instruction::ShiftRight(args!(R, sr1: reg, shamt: amount))),
+        Opcode::Shl => Ok(Instruction::ShiftLeft(
+            args!(R, sr1: src, sr2: r_shamt, shamt: i_shamt, dr: dest),
+        )),
+        Opcode::Shr => Ok(Instruction::ShiftRight(
+            args!(R, sr1: src, sr2: r_shamt, shamt: i_shamt, dr: dest),
+        )),
         _ => unreachable!(),
     }
 }

assembler/src/assembler/lexer.rs

@@ -65,7 +65,7 @@ pub fn lexer(mut program: String, module: u64) -> Result<Vec<Token>, AssembleErr
         }
     }
 
-    println!("{:#?}", tokens);
+    // println!("{:#?}", tokens);
 
     Ok(tokens)
 }

assembler/src/assembler/mod.rs

@@ -9,13 +9,9 @@ use std::{
     thread,
 };
 
+pub use common::logging::log;
 use common::prelude::Instruction;
 
-// TODO: Use an actual logging or tracing library for pretty (scoped) output.
-fn log(message: &str) {
-    println!("\x1b[32mINFO:\x1b[0m {message}");
-}
-
 // Module declarations
 #[macro_use]
 pub mod macros;
@@ -138,7 +134,11 @@ fn assemble(src: &Path) -> Result<Vec<Instruction>, AssembleError> {
     create_sections(&mut nodes)?;
     resolve_symbols(&mut nodes)?;
 
+    log("Generating assembly output...");
+
     let instructions = codegen(nodes)?;
+
+    log("Compilation Successful");
     Ok(instructions)
 }
 
@@ -187,10 +187,7 @@ fn prepare_dependency(
 
     let deps = Parser::get_dependencies(&nodes, path)?;
 
-    log(&format!(
-        "{:20} {:20}",
-        "Expanding PseudoInstructions", filename
-    ));
+    log(&format!("{:20} {:20}", "Expanding Pseudo-ops", filename));
 
     // add a section instruction
     nodes.insert(
@@ -198,9 +195,9 @@ fn prepare_dependency(
         node!(None, Opcode::Segment, Token::Immediate(file_hash as u32)),
     );
 
-    for n in &nodes {
-        println!("{n}");
-    }
+    // for n in &nodes {
+    //     println!("{n}");
+    // }
 
     program.add_module(nodes);
 

assembler/src/assembler/model.rs

@@ -184,11 +184,11 @@ pub enum Token {
 impl fmt::Display for Token {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         match self {
-            Self::Symbol(symbol) => write!(f, "{}", symbol),
-            Self::Register(register) => write!(f, "{}", register),
-            Self::Immediate(immediate) => write!(f, "{}", immediate),
-            Self::StringLit(string_lit) => write!(f, "{}", string_lit),
-            Self::Opcode(opcode) => write!(f, "{}", opcode),
+            Self::Symbol(symbol) => write!(f, "{symbol}"),
+            Self::Register(register) => write!(f, "{register}",),
+            Self::Immediate(immediate) => write!(f, "{immediate}",),
+            Self::StringLit(string_lit) => write!(f, "{string_lit}",),
+            Self::Opcode(opcode) => write!(f, "{opcode}",),
         }
     }
 }

assembler/src/assembler/parser.rs

@@ -1,5 +1,6 @@
 use std::path::{Path, PathBuf};
 
+use crate::assembler::TokenType;
 use crate::{assembler::AssembleError, expect_token, expect_type, node};
 
 use crate::assembler::model::{Node, Opcode, Token};
@@ -100,6 +101,7 @@ impl Parser {
         let opcode = expect_token!(self.next()?, Opcode)?;
         let args: Vec<Token>;
 
+        #[allow(clippy::match_same_arms)]
         match opcode {
             // R-type instructions
             Opcode::Mov | Opcode::Movs => {
@@ -112,22 +114,25 @@ impl Parser {
                 let base = expect_type!(self.next()?, Register, Symbol)?;
                 let dest = expect_type!(self.next()?, Register)?;
 
-                let mut offset = Token::Immediate(0);
-                if let Ok(next) = self.peek_next()
-                    && expect_type!(next, Immediate).is_ok() {
-                        offset = self.next()?;
+                let offset = match self.peek_next() {
+                    Ok(next) if expect_type!(next.clone(), Immediate).is_ok() => {
+                        self.next()?
                     }
+                    _ => Token::Immediate(0),
+                };
 
                 args = vec![base, dest, offset];
             }
             Opcode::Stb | Opcode::Sth | Opcode::Stw => {
                 let base = expect_type!(self.next()?, Register)?;
                 let dest = expect_type!(self.next()?, Register, Symbol)?;
-                let mut offset = Token::Immediate(0);
-                if let Ok(next) = self.peek_next()
-                    && expect_type!(next, Immediate).is_ok() {
-                        offset = self.next()?;
+
+                let offset = match self.peek_next() {
+                    Ok(next) if expect_type!(next.clone(), Immediate).is_ok() => {
+                        self.next()?
                     }
+                    _ => Token::Immediate(0),
+                };
                 args = vec![base, dest, offset];
             }
 
@@ -146,15 +151,49 @@ impl Parser {
             }
 
             Opcode::Not | Opcode::Cmp => {
-                let reg1 = expect_type!(self.next()?, Register, Symbol)?;
-                let reg2 = expect_type!(self.next()?, Register, Symbol)?;
-                args = vec![reg1, reg2];
+                let src = expect_type!(self.next()?, Register, Symbol)?;
+                let dest = expect_type!(self.next()?, Register, Symbol)?;
+                args = vec![src, dest];
             }
-
             Opcode::Shl | Opcode::Shr => {
-                let reg = expect_type!(self.next()?, Register, Symbol)?;
-                let num = expect_type!(self.next()?, Immediate)?;
-                args = vec![reg, num];
+                let src = expect_type!(self.next()?, Register, Symbol)?;
+
+                // First operand after src: could be immediate or register
+                let first = self.next()?;
+
+                let (r_shamt, i_shamt) = match first {
+                    Token::Register(_) => (
+                        first,
+                        if let Ok(tok) = self.peek_next() {
+                            if expect_type!(tok, Immediate).is_ok() {
+                                self.next()?
+                            } else {
+                                Token::Immediate(0)
+                            }
+                        } else {
+                            Token::Immediate(0)
+                        },
+                    ),
+                    Token::Immediate(_) => (Token::Register(Register::Zero), first),
+                    _ => {
+                        return Err(AssembleError::UnexpectedToken(
+                            first,
+                            TokenType::Immediate,
+                        ));
+                    }
+                };
+
+                let dest = if let Ok(tok) = self.peek_next() {
+                    if expect_type!(tok, Register).is_ok() {
+                        self.next()?
+                    } else {
+                        src.clone() // Default to src if no dest specified
+                    }
+                } else {
+                    src.clone() // Default to src if no dest specified
+                };
+
+                args = vec![src, r_shamt, i_shamt, dest];
             }
 
             Opcode::Inc | Opcode::Dec => {

assembler/src/lib.rs

@@ -24,5 +24,27 @@ pub mod prelude {
     pub use crate::tooling::project;
 }
 
+use std::{fs, path::Path};
+
 use num_cpus as _;
 use threadpool as _;
+
+use crate::prelude::CompilerEngine;
+
+pub fn assemble_file(input: &str, output: &str) -> Result<(), std::io::Error> {
+    let mut engine = CompilerEngine::new();
+    engine.start_compilation(Path::new(input));
+    let result = engine.wait_for_result().expect("assembler failed.");
+
+    let buffer: Vec<u8> = result
+        .iter()
+        .flat_map(|instruction| instruction.encode().to_be_bytes())
+        .collect();
+
+    if let Err(e) = fs::write(output, buffer) {
+        eprintln!("Failed to write to output file: {e}");
+        std::process::exit(1);
+    }
+
+    Ok(())
+}

assembler/src/main.rs

@@ -3,6 +3,7 @@ use num_cpus as _;
 use threadpool as _;
 
 use assembler::{
+    assemble_file,
     prelude::*,
     tooling::{brainf, project},
 };
@@ -46,19 +47,5 @@ fn main() {
 
     let input_path = &args[2];
     let output_path = &args[4];
-    let src = PathBuf::from(input_path);
-
-    // Initialize the compiler engine
-    let mut compiler = CompilerEngine::new();
-    compiler.start_compilation(&src);
-
-    // Or block until done
-    let result = compiler.wait_for_result().unwrap();
-
-    for instruction in result {
-        if let Err(e) = fs::write(output_path, instruction.encode().to_be_bytes()) {
-            eprintln!("Failed to write to output file: {e}");
-            std::process::exit(1);
-        }
-    }
+    assemble_file(input_path, output_path).unwrap();
 }

common/src/instructions.rs

@@ -40,7 +40,7 @@ pub enum InstructionType {
     Immediate,
 }
 
-#[derive(Copy, Clone, Debug, PartialEq, Eq)]
+#[derive(Copy, Clone, Debug, PartialEq, Eq, Default)]
 #[non_exhaustive]
 pub enum Register {
     // general purpose registers
@@ -69,7 +69,9 @@ pub enum Register {
     Idr,
     Mmr,
     Zero,
-    NoReg,
+
+    #[default]
+    Null, // Invalid - Triggers a fault if accessed
 
     // system registers - can't be written to by instructions.
     Mar,
@@ -104,12 +106,6 @@ impl Register {
     }
 }
 
-impl Default for Register {
-    fn default() -> Self {
-        Self::NoReg
-    }
-}
-
 impl TryFrom<u8> for Register {
     type Error = RegisterParseError;
 
@@ -144,7 +140,7 @@ impl TryFrom<u8> for Register {
             0x14 => Self::Idr,
             0x15 => Self::Mmr,
             0x16 => Self::Zero,
-            0x17 => Self::NoReg,
+            0x17 => Self::Null,
             0x18 => Self::Mar,
             0x19 => Self::Mdr,
             0x1A => Self::Sts,
@@ -183,7 +179,7 @@ impl TryFrom<&str> for Register {
             "idr" => Ok(Self::Idr),
             "mmr" => Ok(Self::Mmr),
             "zero" => Ok(Self::Zero),
-            "null" => Ok(Self::NoReg),
+            "null" => Ok(Self::Null),
             "pcx" => Ok(Self::Pcx),
             _ => Err(RegisterParseError::InvalidName(value.to_string())),
         }
@@ -216,7 +212,7 @@ impl std::fmt::Display for Register {
             Self::Idr => write!(f, "idr"),
             Self::Mmr => write!(f, "mmr"),
             Self::Zero => write!(f, "zero"),
-            Self::NoReg => write!(f, "noreg"),
+            Self::Null => write!(f, "null"),
             Self::Mar => write!(f, "mar"),
             Self::Mdr => write!(f, "mdr"),
             Self::Sts => write!(f, "sts"),

common/src/instructions/encode.rs

@@ -8,9 +8,9 @@ pub trait Encode {
 /// Encodes a zero argument instruction.
 fn encode_no_args(opcode: u8) -> u32 {
     let opcode = u32::from(opcode);
-    let sr1 = Register::NoReg as u32;
-    let sr2 = Register::NoReg as u32;
-    let dr = Register::NoReg as u32;
+    let sr1 = Register::Null as u32;
+    let sr2 = Register::Null as u32;
+    let dr = Register::Null as u32;
     let shamt = 0;
 
     (opcode << 26) | (sr1 << 21) | (sr2 << 16) | (dr << 11) | (shamt << 6)

common/src/instructions/encode/tests.rs

@@ -2,7 +2,7 @@ use crate::prelude::*;
 
 #[test]
 fn test_encode_nop() {
-    let no_reg = Register::NoReg as u32;
+    let no_reg = Register::Null as u32;
     let no_op = u32::from(Instruction::Nop.opcode());
 
     let expected = (no_op << 26) | (no_reg << 21) | (no_reg << 16) | (no_reg << 11);
@@ -15,7 +15,7 @@ fn test_encode_nop() {
 fn test_encode_mov() {
     let rg0 = Register::Rg0 as u32;
     let rg1 = Register::Rg1 as u32;
-    let no_reg = Register::NoReg as u32;
+    let no_reg = Register::Null as u32;
 
     let instruction = Instruction::Mov(RTypeArgs::new(
         Some(Register::Rg0),
@@ -53,7 +53,7 @@ fn test_encode_load_byte() {
 #[test]
 fn test_encode_shift_left_shamt() {
     let rg0 = Register::Rg0 as u32;
-    let no_reg = Register::NoReg as u32;
+    let no_reg = Register::Null as u32;
 
     let shift_amount = 5;
 
@@ -80,7 +80,7 @@ fn test_encode_shift_left_shamt() {
 fn test_encode_shift_left_reg() {
     let rg0 = Register::Rg0 as u32;
     let rg1 = Register::Rg1 as u32;
-    let no_reg = Register::NoReg as u32;
+    let no_reg = Register::Null as u32;
 
     let instruction = Instruction::ShiftLeft(RTypeArgs::new(
         Some(Register::Rg0),

common/src/lib.rs

@@ -13,6 +13,7 @@
 )]
 
 pub mod instructions;
+pub mod logging;
 
 pub mod prelude {
     //! A collection of types you should definitely import when working with this crate.

common/src/logging.rs

@@ -0,0 +1,4 @@
+// TODO: Use an actual logging or tracing library for pretty (scoped) output.
+pub fn log(message: &str) {
+    println!("\x1b[32mINFO:\x1b[0m {message}");
+}

compiler/Cargo.toml

@@ -0,0 +1,10 @@
+[package]
+name = "compiler"
+version.workspace = true
+edition.workspace = true
+authors.workspace = true
+
+[dependencies]
+chrono = "0.4.43"
+common = { path = "../common" }
+uuid = { version = "1.20.0", features = ["v4"] }

compiler/bacon.toml

@@ -0,0 +1,129 @@
+# This is a configuration file for the bacon tool
+#
+# Complete help on configuration: https://dystroy.org/bacon/config/
+#
+# You may check the current default at
+#   https://github.com/Canop/bacon/blob/main/defaults/default-bacon.toml
+
+default_job = "check"
+
+[jobs.check]
+command = ["cargo", "check", "--color", "always"]
+need_stdout = false
+
+[jobs.check-all]
+command = ["cargo", "check", "--all-targets", "--color", "always"]
+need_stdout = false
+
+# Run clippy on the default target
+[jobs.clippy]
+command = [
+    "cargo", "clippy",
+    "--color", "always",
+]
+need_stdout = false
+
+# Run clippy on all targets
+# To disable some lints, you may change the job this way:
+#    [jobs.clippy-all]
+#    command = [
+#        "cargo", "clippy",
+#        "--all-targets",
+#        "--color", "always",
+#    	 "--",
+#    	 "-A", "clippy::bool_to_int_with_if",
+#    	 "-A", "clippy::collapsible_if",
+#    	 "-A", "clippy::derive_partial_eq_without_eq",
+#    ]
+# need_stdout = false
+[jobs.clippy-all]
+command = [
+    "cargo", "clippy",
+    "--all-targets",
+    "--color", "always",
+]
+need_stdout = false
+
+# This job lets you run
+# - all tests: bacon test
+# - a specific test: bacon test -- config::test_default_files
+# - the tests of a package: bacon test -- -- -p config
+[jobs.test]
+command = [
+    "cargo", "test", "--color", "always",
+    "--", "--color", "always", # see https://github.com/Canop/bacon/issues/124
+]
+need_stdout = true
+
+[jobs.nextest]
+command = [
+    "cargo", "nextest", "run",
+    "--color", "always",
+    "--hide-progress-bar", "--failure-output", "final"
+]
+need_stdout = true
+analyzer = "nextest"
+
+[jobs.doc]
+command = ["cargo", "doc", "--color", "always", "--no-deps"]
+need_stdout = false
+
+# If the doc compiles, then it opens in your browser and bacon switches
+# to the previous job
+[jobs.doc-open]
+command = ["cargo", "doc", "--color", "always", "--no-deps", "--open"]
+need_stdout = false
+on_success = "back" # so that we don't open the browser at each change
+
+# You can run your application and have the result displayed in bacon,
+# if it makes sense for this crate.
+# Don't forget the `--color always` part or the errors won't be
+# properly parsed.
+[jobs.run]
+command = [
+    "cargo", "run",
+    "--color", "always",
+    "--",
+    "../resources/dsa/example.dsc",
+    "../resources/dsa/example.dsa"
+    # put launch parameters for your program behind a `--` separator
+]
+need_stdout = true
+allow_warnings = true
+background = true
+
+# Run your long-running application (eg server) and have the result displayed in bacon.
+# For programs that never stop (eg a server), `background` is set to false
+# to have the cargo run output immediately displayed instead of waiting for
+# program's end.
+# 'on_change_strategy' is set to `kill_then_restart` to have your program restart
+# on every change (an alternative would be to use the 'F5' key manually in bacon).
+# If you often use this job, it makes sense to override the 'r' key by adding
+# a binding `r = job:run-long` at the end of this file .
+[jobs.run-long]
+command = [
+    "cargo", "run",
+    "--color", "always",
+    # put launch parameters for your program behind a `--` separator
+]
+need_stdout = true
+allow_warnings = true
+background = false
+on_change_strategy = "kill_then_restart"
+
+# This parameterized job runs the example of your choice, as soon
+# as the code compiles.
+# Call it as
+#    bacon ex -- my-example
+[jobs.ex]
+command = ["cargo", "run", "--color", "always", "--example"]
+need_stdout = true
+allow_warnings = true
+
+# You may define here keybindings that would be specific to
+# a project, for example a shortcut to launch a specific job.
+# Shortcuts to internal functions (scrolling, toggling, etc.)
+# should go in your personal global prefs.toml file instead.
+[keybindings]
+# alt-m = "job:my-job"
+c = "job:clippy-all" # comment this to have 'c' run clippy on only the default target

compiler/src/backend/dsa/codegen.rs

@@ -0,0 +1,955 @@
+use std::collections::HashMap;
+use std::sync::atomic::AtomicU32;
+use std::time::SystemTime;
+
+use chrono::{DateTime, Local};
+
+use super::registers::RegisterAllocator;
+use crate::backend::dsa::instruction::{InsBlock as IB, Instruction as I, Label};
+use crate::backend::dsa::registers::Register;
+
+use crate::model::{
+    AssignmentOperator, BinaryOperator, Call, CompilerError, ConstExpr, Declaration,
+    Dependency, Expression, Number, Program, Statement, TypeId, UnaryOperator, Variable,
+};
+
+pub struct CodeGenerator {
+    ast: Program,
+    imports: HashMap<String, I>,
+    globals: HashMap<String, I>,
+    functions: Vec<IB>,
+    symbols: Vec<String>,
+    allocator: RegisterAllocator,
+}
+
+impl CodeGenerator {
+    pub fn new(ast: Program) -> Self {
+        CodeGenerator {
+            ast,
+            imports: HashMap::new(),
+            globals: HashMap::new(),
+            functions: Vec::new(),
+            symbols: Vec::new(),
+            allocator: RegisterAllocator::new(),
+        }
+    }
+
+    pub fn include(&mut self, name: impl Into<String>, path: impl Into<String>) {
+        let name = name.into();
+        self.imports.insert(name.clone(), I::include(name, path));
+    }
+
+    fn is_global(&self, name: &str) -> bool {
+        // Check if this variable is in the globals list
+        self.globals.contains_key(name)
+    }
+
+    pub fn generate(&mut self) -> Result<String, CompilerError> {
+        // always include the print library for debugging!
+        self.include("print", "./lib/io/print.dsa");
+
+        for block in self.ast.clone().declarations {
+            match block {
+                Declaration::Variable {
+                    var: Variable { name, .. },
+                    ..
+                } => self.symbols.push(name),
+                Declaration::Function { name, .. } => self.symbols.push(name),
+                Declaration::Dependency(Dependency { name, .. }) => {
+                    self.symbols.push(name)
+                }
+                Declaration::Struct { .. } => {} /* we can't do any code generation for
+                                                  * a struct yet. we may need to later
+                                                  * once these become class-like
+                                                  * objects with implementations */
+            }
+        }
+
+        for block in self.ast.clone().declarations {
+            self.generate_block(block.clone())?;
+        }
+
+        let assembly = self.generate_layout()?;
+        Ok(assembly
+            .iter()
+            .map(|i| i.to_string())
+            .collect::<Vec<_>>()
+            .join("\n"))
+    }
+
+    fn generate_layout(&mut self) -> Result<IB, CompilerError> {
+        let datetime: DateTime<Local> = SystemTime::now().into();
+
+        let mut block = IB::new();
+
+        block.extend(vec![
+            I::global_comment(format!(
+                "GENERATED BY DSC COMPILER
+                Generated at {}",
+                datetime.format("%Y-%m-%d %H:%M:%S")
+            )),
+            I::Newline,
+            I::global_comment("Imports"),
+        ]);
+
+        block.extend(self.imports.values().cloned().collect::<Vec<_>>());
+
+        block.extend(vec![
+            I::Newline,
+            I::global_comment("Globals & Reserved Memory"),
+        ]);
+
+        block.extend(self.globals.values().cloned().collect::<Vec<_>>());
+
+        block.extend(vec![
+            I::Newline,
+            I::global_comment("Entry Point"),
+            I::db_word("stack", 0x10000),
+            I::db_string("message", "Process Exited with code:"),
+            // init function for stack setup.
+            I::label("_init"),
+            I::ldw_label("stack", Register::Bpr),
+            I::mov(Register::Bpr, Register::Spr),
+            I::push(Register::Zero),
+            I::call("main"),
+            I::call("print::print_newline"),
+            I::lwi_label("message", Register::Rg0),
+            I::push(Register::Rg0),
+            I::call("print::print"),
+            I::pop(Register::Zero),
+            I::call("print::print_hex_word"),
+            I::pop(Register::Zero),
+            I::Hlt,
+            I::Newline,
+            // default return block boilerplate
+            I::global_comment("Return"),
+            I::label("_ret"),
+            I::mov(Register::Bpr, Register::Spr),
+            I::pop(Register::Bpr),
+            I::Return,
+        ]);
+
+        for function in self.functions.iter() {
+            block.extend(function.iter().cloned());
+        }
+
+        Ok(block)
+    }
+
+    fn generate_global(&mut self, name: &str, init: Option<ConstExpr>) {
+        let init = init.unwrap_or(ConstExpr::Number(0));
+        match init {
+            ConstExpr::Number(value) => {
+                self.globals
+                    .insert(name.to_string(), I::db_word(name, value as u32));
+            }
+            ConstExpr::String(str) => {
+                self.globals
+                    .insert(name.to_string(), I::db_string(name, str));
+            }
+        }
+    }
+
+    fn generate_block(&mut self, block: Declaration) -> Result<(), CompilerError> {
+        match block {
+            Declaration::Variable { var, init, .. } => {
+                self.generate_global(&var.name, init)
+            }
+            Declaration::Function {
+                name,
+                params,
+                body,
+                return_type,
+            } => {
+                let func = self.generate_function(&name, &params, &body, return_type);
+                self.functions.push(func);
+            }
+            Declaration::Dependency(Dependency { name, path }) => {
+                self.include(name, path);
+            }
+            Declaration::Struct { .. } => {} /* can't do any codegen for these yet,
+                                              * they're just types. */
+        };
+
+        Ok(())
+    }
+
+    // Example: Generate code for a function
+    fn generate_function(
+        &mut self,
+        name: &str,
+        params: &[Variable],
+        body: &[Statement],
+        return_type: TypeId,
+    ) -> IB {
+        let mut code = IB::new();
+
+        // Reset allocator for new function
+        self.allocator.reset();
+
+        let fmtparams = params
+            .iter()
+            .map(|p| format!("{}: {}", p.name, p.type_id))
+            .collect::<Vec<String>>()
+            .join(", ");
+
+        code.extend(vec![
+            I::global_comment(format!("fn {name}({fmtparams}) -> {return_type}")),
+            I::label(name),
+            I::push(Register::Bpr),
+            I::mov(Register::Spr, Register::Bpr),
+        ]);
+
+        // Allocate parameters to registers or stack locations
+        for (i, param) in params.iter().enumerate() {
+            let offset = 8 + (i as i32 * 4); // Parameters start at bpr+8
+            // Track that this parameter is at a stack location
+            let (reg, load_code) = self.allocator.alloc_var(&param.name).unwrap();
+            code.append(load_code);
+            code.push(I::ldw_reg_offset(Register::Bpr, reg, offset));
+        }
+
+        // Generate code for function body
+        for stmt in body {
+            let stmt_code = self.generate_statement(stmt, &mut code).unwrap();
+            code.append(stmt_code);
+        }
+
+        // automatically return at function end
+        if let Some(x) = code.iter().last()
+            && let I::Jmp { target: Label(val) } = x
+            && val == "_ret"
+        {
+        } else {
+            code.push(I::jmp("_ret"));
+        }
+
+        code.insert(0, I::Newline);
+
+        code
+    }
+
+    // Example: Generate code for a statement
+    fn generate_statement(
+        &mut self,
+        stmt: &Statement,
+        func_body: &mut IB,
+    ) -> Result<IB, CompilerError> {
+        let mut code = IB::new();
+
+        match stmt {
+            Statement::Declaration { var, value } => {
+                if let Some(expr) = value {
+                    // Evaluate expression
+                    let (result_reg, expr_code) =
+                        self.generate_expression(expr, true, func_body)?;
+                    code.append(expr_code);
+
+                    // Store result in variable
+                    let store_code = self.allocator.store_var(&var.name, &result_reg);
+                    code.append(store_code);
+
+                    // Free temporary register
+                    self.allocator.free_temp(result_reg);
+                } else {
+                    // Just declaring variable without initialization
+                    self.allocator.alloc_var(&var.name)?;
+                }
+            }
+
+            Statement::Break => unimplemented!("need scope tracking first!"),
+            Statement::Continue => unimplemented!("need scope tracking first!"),
+            Statement::Defer(_func) => unimplemented!("we need scope tracking first!"),
+
+            Statement::PtrWrite { ptr, value } => {
+                let (result_reg, expr_code) =
+                    self.generate_expression(value, true, func_body)?;
+                code.append(expr_code);
+
+                let (ptr_reg, ptr_code) =
+                    self.generate_expression(ptr, true, func_body)?;
+                code.append(ptr_code);
+
+                code.push(I::stw_reg(result_reg, ptr_reg));
+
+                self.allocator.free_temp(result_reg);
+                self.allocator.free_temp(ptr_reg);
+            }
+
+            Statement::Assign {
+                varname,
+                value,
+                operator,
+            } => {
+                // Evaluate expression
+                let (result_reg, expr_code) =
+                    self.generate_expression(value, true, func_body)?;
+                code.append(expr_code);
+
+                if *operator == AssignmentOperator::Assign {
+                    // Check if this is a global variable
+                    if self.is_global(varname) {
+                        // Store to global label
+                        code.push(I::stw_label(result_reg, varname.clone()))
+                    } else {
+                        // Store result in local variable
+                        let store_code = self.allocator.store_var(varname, &result_reg);
+                        code.append(store_code);
+                    }
+
+                    // Free temporary register
+                    self.allocator.free_temp(result_reg);
+
+                    return Ok(code);
+                }
+
+                // for more complex assignment cases we need an intermediate register.
+                let (temp_reg, temp_code) = self.allocator.alloc_temp()?;
+                code.append(temp_code);
+
+                // fetch the value of the variable
+                let var_reg = if self.is_global(varname) {
+                    let instruction = I::ldw_label(varname.clone(), temp_reg);
+                    code.push(instruction);
+                    temp_reg
+                } else {
+                    let (rg, block) = self.allocator.load_var(varname)?;
+                    code.append(block);
+                    rg
+                };
+
+                let assign_code = match operator {
+                    AssignmentOperator::Assign => {
+                        unreachable!("assignment was already checked earlier.")
+                    }
+                    AssignmentOperator::AddAssign => {
+                        I::add(var_reg, result_reg, temp_reg)
+                    }
+                    AssignmentOperator::SubAssign => {
+                        I::sub(var_reg, result_reg, temp_reg)
+                    }
+                    AssignmentOperator::MulAssign => {
+                        return Err(CompilerError::Unimplemented(
+                            "TODO: implement multiplication for assignment".to_string(),
+                        ));
+                    }
+                    AssignmentOperator::DivAssign => {
+                        return Err(CompilerError::Unimplemented(
+                            "TODO: write proper div function for DSA".to_string(),
+                        ));
+                    }
+                    AssignmentOperator::ModAssign => {
+                        return Err(CompilerError::Unimplemented(
+                            "TODO: write proper mod function for DSA".to_string(),
+                        ));
+                    }
+                    AssignmentOperator::AndAssign => {
+                        I::and(var_reg, result_reg, temp_reg)
+                    }
+                    AssignmentOperator::OrAssign => I::or(var_reg, result_reg, temp_reg),
+                    AssignmentOperator::XorAssign => {
+                        I::xor(var_reg, result_reg, temp_reg)
+                    }
+                    AssignmentOperator::LeftShiftAssign => {
+                        // this is only useful if we optimise out the register allocation
+                        // inside value.
+                        // if let Expression::Number { value, .. } = *value {
+                        //     I::shl(var_reg, value, temp_reg)
+                        // }
+                        I::shl(var_reg, result_reg, 0, temp_reg)
+                    }
+                    AssignmentOperator::RightShiftAssign => {
+                        // this is only useful if we optimise out the register allocation
+                        // if let Expression::Number { value, .. } = *value {
+                        //     I::shr(var_reg, value, temp_reg)
+                        // }
+                        I::shr(var_reg, result_reg, 0, temp_reg)
+                    }
+                };
+                code.push(assign_code);
+
+                // Check if this is a global variable
+                if self.is_global(varname) {
+                    // Store to global label
+                    code.push(I::stw_label(temp_reg, varname.clone()))
+                } else {
+                    // Store result in local variable
+                    let store_code = self.allocator.store_var(varname, &temp_reg);
+                    code.append(store_code);
+                }
+
+                self.allocator.free_temp(result_reg);
+                self.allocator.free_temp(temp_reg);
+            }
+
+            Statement::Return(expr) => {
+                if let Some(e) = expr {
+                    let (result_reg, expr_code) =
+                        self.generate_expression(e, true, func_body)?;
+                    code.append(expr_code);
+                    code.push(I::stw_reg_offset(result_reg, Register::Bpr, 8));
+                    code.push(I::jmp("_ret"));
+                    self.allocator.free_temp(result_reg);
+                }
+            }
+
+            Statement::If {
+                condition,
+                then_stmt,
+                else_stmt,
+            } => {
+                // Generate condition
+                let (cond_reg, cond_code) =
+                    self.generate_expression(condition, true, func_body)?;
+                code.append(cond_code);
+
+                // Compare with zero
+                code.push(I::cmp(cond_reg, Register::Zero));
+                self.allocator.free_temp(cond_reg);
+
+                // Generate unique labels
+                let then_label = format!("_then_{}", self.get_unique_label());
+                let else_label = format!("_else_{}", self.get_unique_label());
+                let end_label = format!("_end_{}", self.get_unique_label());
+
+                // Jump to else if condition is false (equal to zero)
+                code.push(I::jeq(else_label.clone()));
+
+                // Then block
+                code.push(I::label(then_label));
+                for s in then_stmt {
+                    code.append(self.generate_statement(s, func_body)?);
+                }
+
+                if then_stmt.is_empty() {
+                    code.push(I::Nop);
+                }
+
+                code.push(I::jmp(end_label.clone()));
+
+                // Else block
+                code.push(I::label(else_label));
+                for s in else_stmt {
+                    code.append(self.generate_statement(s, func_body)?);
+                }
+
+                if else_stmt.is_empty() {
+                    code.push(I::Nop);
+                }
+
+                code.push(I::label(end_label));
+            }
+
+            Statement::While { condition, body } => {
+                let loop_start = format!("_while_start_{}", self.get_unique_label());
+                let loop_end = format!("_while_end_{}", self.get_unique_label());
+
+                code.push(I::label(&loop_start));
+
+                // Generate condition
+                let (cond_reg, cond_code) =
+                    self.generate_expression(condition, true, func_body)?;
+                code.append(cond_code);
+
+                code.push(I::cmp(cond_reg, Register::Zero));
+                self.allocator.free_temp(cond_reg);
+
+                code.push(I::jeq(loop_end.clone()));
+
+                // Loop body
+                for s in body {
+                    code.append(self.generate_statement(s, func_body)?);
+                }
+
+                code.push(I::jmp(loop_start));
+                code.push(I::label(loop_end));
+            }
+
+            Statement::Loop(body) => {
+                let loop_start = format!("_loop_start_{}", self.get_unique_label());
+                code.push(I::label(&loop_start));
+
+                for s in body {
+                    code.append(self.generate_statement(s, func_body)?);
+                }
+
+                code.push(I::jmp(loop_start));
+            }
+
+            Statement::Expression { expr } => {
+                let (result_reg, expr_code) =
+                    self.generate_expression(expr, false, func_body)?;
+                code.append(expr_code);
+                self.allocator.free_temp(result_reg);
+            }
+
+            Statement::Block(statements) => {
+                for s in statements {
+                    code.append(self.generate_statement(s, func_body)?);
+                }
+            }
+        }
+
+        Ok(code)
+    }
+
+    // Example: Generate code for an expression
+    // Returns (register containing result, assembly code)
+    fn generate_expression(
+        &mut self,
+        expr: &Expression,
+        use_result: bool,
+        func_body: &mut IB,
+    ) -> Result<(Register, IB), CompilerError> {
+        let mut code = IB::new();
+
+        match expr {
+            Expression::Empty => Ok((Register::Null, code)),
+
+            Expression::Number(n) => match n {
+                Number::Signed(value, _) => {
+                    let (reg, alloc_code) = self.allocator.alloc_temp()?;
+                    code.append(alloc_code);
+
+                    // Load immediate value
+                    code.push(I::lwi(*value as u32, reg));
+
+                    Ok((reg, code))
+                }
+                Number::Unsigned(value, _) => {
+                    let (reg, alloc_code) = self.allocator.alloc_temp()?;
+                    code.append(alloc_code);
+
+                    // Load immediate value
+                    code.push(I::lwi(*value as u32, reg));
+
+                    Ok((reg, code))
+                }
+            },
+
+            Expression::CharLiteral(value) => {
+                let (reg, alloc_code) = self.allocator.alloc_temp()?;
+                code.append(alloc_code);
+
+                // Load immediate value
+                code.push(I::comment(format!("char literal '{value}'")));
+                code.push(I::lwi(*value as u32, reg));
+
+                Ok((reg, code))
+            }
+
+            Expression::StringLiteral(value) => {
+                let (reg, alloc_code) = self.allocator.alloc_temp()?;
+                code.append(alloc_code);
+
+                // write string into memory
+                let uuid = self.get_unique_label();
+
+                func_body.insert(0, I::db_string(format!("str_{uuid}"), value));
+
+                // Load pointer to string
+                code.push(I::lwi_label(format!("str_{uuid}"), reg));
+
+                Ok((reg, code))
+            }
+
+            Expression::ArrayLiteral { elements, type_id } => todo!(),
+            Expression::StructLiteral {
+                name,
+                fields,
+                type_id,
+            } => todo!(),
+
+            Expression::Variable { name, .. } => {
+                if self.is_global(&name.name) {
+                    // Allocate a temporary register for the global
+                    let (reg, alloc_code) = self.allocator.alloc_temp()?;
+                    code.append(alloc_code);
+
+                    // Load from global label
+                    code.push(I::ldw_label(name.name.clone(), reg));
+
+                    Ok((reg, code))
+                } else {
+                    // Local variable - use existing allocator logic
+                    let (reg, load_code) = self.allocator.load_var(&name.name)?;
+                    code.append(load_code);
+                    Ok((reg, code))
+                }
+            }
+
+            Expression::Binary {
+                op, left, right, ..
+            } => {
+                // Evaluate left operand
+                let (left_reg, left_code) =
+                    self.generate_expression(left, true, func_body)?;
+                code.append(left_code);
+
+                // Evaluate right operand
+                let (right_reg, right_code) =
+                    self.generate_expression(right, true, func_body)?;
+                code.append(right_code);
+
+                // Allocate result register
+                let (result_reg, result_alloc) = self.allocator.alloc_temp()?;
+                code.append(result_alloc);
+
+                // Generate operation
+                match op {
+                    BinaryOperator::Add => {
+                        code.push(I::add(left_reg, right_reg, result_reg));
+                    }
+                    BinaryOperator::Sub => {
+                        code.push(I::sub(left_reg, right_reg, result_reg));
+                    }
+                    BinaryOperator::Mul => {
+                        self.include("maths", "./lib/maths/core.dsa");
+                        // Call multiply function
+                        code.push(I::push(right_reg));
+                        code.push(I::push(left_reg));
+                        code.push(I::call("maths::multiply"));
+                        code.push(I::pop(result_reg));
+                        code.push(I::pop(Register::Zero));
+                    }
+                    BinaryOperator::Div => {
+                        return Err(CompilerError::Unimplemented(
+                            "TODO: write proper div function for DSA".to_string(),
+                        ));
+                        // self.include("maths", "./lib/maths/core.dsa");
+                        // // Call divide function
+                        // code.push(format!("\tpush {}", right_reg));
+                        // code.push(format!("\tpush {}", left_reg));
+                        // code.push("\tcall maths::divide".to_string());
+                        // code.push(format!("\tpop {}", result_reg));
+                        // code.push("\tpop zero".to_string());
+                    }
+                    BinaryOperator::Mod => {
+                        return Err(CompilerError::Unimplemented(
+                            "TODO: write proper mod function for DSA".to_string(),
+                        ));
+                        // self.include("maths", "./lib/maths/core.dsa");
+                        // // Call modulo function
+                        // code.push(format!("\tpush {}", right_reg));
+                        // code.push(format!("\tpush {}", left_reg));
+                        // code.push("\tcall maths::modulo".to_string());
+                        // code.push(format!("\tpop {}", result_reg));
+                        // code.push("\tpop zero".to_string());
+                    }
+                    BinaryOperator::BitwiseAnd => {
+                        code.push(I::and(left_reg, right_reg, result_reg));
+                    }
+                    BinaryOperator::BitwiseOr => {
+                        code.push(I::or(left_reg, right_reg, result_reg));
+                    }
+                    BinaryOperator::BitwiseXor => {
+                        code.push(I::xor(left_reg, right_reg, result_reg));
+                    }
+                    BinaryOperator::LogicalAnd => {
+                        return Err(CompilerError::Unimplemented(
+                            "assembler/ISA does not yet support logical and!".to_string(),
+                        ));
+                    }
+                    BinaryOperator::LogicalOr => {
+                        return Err(CompilerError::Unimplemented(
+                            "assembler/ISA does not yet support logical or!".to_string(),
+                        ));
+                    }
+                    BinaryOperator::LeftShift => {
+                        code.push(I::shl(left_reg, right_reg, 0, result_reg));
+                    }
+                    BinaryOperator::RightShift => {
+                        code.push(I::shr(left_reg, right_reg, 0, result_reg));
+                    }
+                    // Comparison operators - return 1 (true) or 0 (false)
+                    BinaryOperator::Equal => {
+                        code.push(I::cmp(left_reg, right_reg));
+                        code.push(I::lwi(1, result_reg));
+                        let end_label = format!("_cmp_end_{}", self.get_unique_label());
+                        code.push(I::jeq(end_label.clone()));
+                        code.push(I::lwi(0, result_reg));
+                        code.push(I::label(end_label));
+                    }
+                    BinaryOperator::NotEqual => {
+                        code.push(I::cmp(left_reg, right_reg));
+                        code.push(I::lwi(1, result_reg));
+                        let end_label = format!("_cmp_end_{}", self.get_unique_label());
+                        code.push(I::Jne {
+                            target: Label(end_label.clone()),
+                        });
+                        code.push(I::lwi(0, result_reg));
+                        code.push(I::label(&end_label));
+                    }
+                    BinaryOperator::LessThan => {
+                        code.push(I::cmp(left_reg, right_reg));
+                        code.push(I::lwi(1, result_reg));
+                        let end_label = format!("_cmp_end_{}", self.get_unique_label());
+                        code.push(I::Jlt {
+                            target: Label(end_label.clone()),
+                        });
+                        code.push(I::lwi(0, result_reg));
+                        code.push(I::label(&end_label));
+                    }
+                    BinaryOperator::LessOrEqual => {
+                        code.push(I::cmp(left_reg, right_reg));
+                        code.push(I::lwi(1, result_reg));
+                        let end_label = format!("_cmp_end_{}", self.get_unique_label());
+                        code.push(I::Jle {
+                            target: Label(end_label.clone()),
+                        });
+                        code.push(I::lwi(0, result_reg));
+                        code.push(I::label(&end_label));
+                    }
+                    BinaryOperator::GreaterThan => {
+                        code.push(I::cmp(left_reg, right_reg));
+                        code.push(I::lwi(1, result_reg));
+                        let end_label = format!("_cmp_end_{}", self.get_unique_label());
+                        code.push(I::Jgt {
+                            target: Label(end_label.clone()),
+                        });
+                        code.push(I::lwi(0, result_reg));
+                        code.push(I::label(&end_label));
+                    }
+                    BinaryOperator::GreaterOrEqual => {
+                        code.push(I::cmp(left_reg, right_reg));
+                        code.push(I::lwi(1, result_reg));
+                        let end_label = format!("_cmp_end_{}", self.get_unique_label());
+                        code.push(I::Jge {
+                            target: Label(end_label.clone()),
+                        });
+                        code.push(I::lwi(0, result_reg));
+                        code.push(I::label(&end_label));
+                    } // _ => unimplemented!(),
+                }
+
+                // Free operand registers (allocator will protect variables)
+                self.allocator.free_temp(left_reg);
+                self.allocator.free_temp(right_reg);
+
+                Ok((result_reg, code))
+            }
+
+            Expression::UnaryPostfix { op, operand, .. } => {
+                let (operand_reg, operand_code) =
+                    self.generate_expression(operand, true, func_body)?;
+                code.append(operand_code);
+
+                let (result_reg, result_alloc) = self.allocator.alloc_temp()?;
+                code.append(result_alloc);
+
+                match op {
+                    UnaryOperator::Increment => {
+                        // postfix increment - return old value
+                        code.push(I::mov(operand_reg, result_reg));
+                    }
+                    UnaryOperator::Decrement => {
+                        // postfix decrement - return old value
+                        code.push(I::mov(operand_reg, result_reg));
+                    }
+                    _ => {
+                        return Err(CompilerError::Generic(format!(
+                            "{op} is prefix only!"
+                        )));
+                    }
+                }
+
+                self.allocator.free_temp(operand_reg);
+                Ok((result_reg, code))
+            }
+
+            Expression::Unary { op, operand, .. } => {
+                let (operand_reg, operand_code) =
+                    self.generate_expression(operand, true, func_body)?;
+                code.append(operand_code);
+
+                let (result_reg, result_alloc) = self.allocator.alloc_temp()?;
+                code.append(result_alloc);
+
+                match op {
+                    UnaryOperator::Minus => {
+                        // Negate: result = 0 - operand
+                        code.push(I::sub(Register::Zero, operand_reg, result_reg));
+                    }
+                    UnaryOperator::Plus => {
+                        // Just move
+                        code.push(I::mov(operand_reg, result_reg));
+                    }
+                    UnaryOperator::Dereference => {
+                        code.push(I::ldw_reg(operand_reg, result_reg));
+                    }
+                    UnaryOperator::AddressOf => {
+                        // ensure the referenced variable is on the stack and return its
+                        // address.
+                        let (offset, alloc_code) =
+                            self.allocator.free_register(&operand_reg)?;
+                        code.push(alloc_code);
+                        code.push(I::iadd_dest(
+                            Register::Spr,
+                            offset - self.allocator.get_stack_offset(),
+                            result_reg,
+                        ));
+                    }
+                    UnaryOperator::SizeOf => {
+                        if let Ok(id) = operand.type_id() {
+                            let size = id.size();
+                            code.push(I::lwi(size as u32, result_reg));
+                        }
+                    }
+                    UnaryOperator::Increment => {
+                        // prefix increment
+                        code.push(I::mov(operand_reg, result_reg));
+                        code.push(I::iadd_dest(operand_reg, 1, result_reg));
+                    }
+                    UnaryOperator::Decrement => {
+                        // prefix decrement
+                        code.push(I::mov(operand_reg, result_reg));
+                        code.push(I::iadd_dest(operand_reg, -1, result_reg));
+                    }
+                    UnaryOperator::BitwiseNot => {
+                        code.push(I::not(operand_reg, result_reg));
+                    }
+                    UnaryOperator::LogicalNot => {
+                        return Err(CompilerError::Unimplemented(
+                            "Assembler/ISA does not yet support logical not".to_string(),
+                        ));
+                    }
+                    _ => {
+                        return Err(CompilerError::Generic(format!(
+                            "{op} is postfix only!"
+                        )));
+                    }
+                }
+
+                self.allocator.free_temp(operand_reg);
+                Ok((result_reg, code))
+            }
+
+            Expression::Call {
+                func: Call { name, args },
+                ..
+            } => {
+                // first evaluate all the args we're going to need
+                let mut arg_regs = Vec::new();
+                for arg in args.iter().rev() {
+                    let (arg_reg, arg_code) =
+                        self.generate_expression(arg, true, func_body)?;
+                    code.append(arg_code);
+                    arg_regs.push(arg_reg);
+                }
+
+                // Save caller-saved registers and track which ones we saved
+                let saved_regs = self.allocator.get_caller_saved_registers();
+                for reg in &saved_regs {
+                    // spill variables to stack
+                    code.push(self.allocator.free_register(reg).unwrap().1);
+                }
+
+                // Evaluate and push arguments in reverse order
+                for (i, arg_reg) in arg_regs.iter().enumerate() {
+                    code.push(I::comment(format!("push arg {}", args.len() - 1 - i)));
+                    code.push(I::push(*arg_reg));
+                }
+
+                if self.symbols.contains(&name.name) {
+                    // Call local function
+                    code.push(I::call(name.to_string()));
+                } else if let Some(ns) = name.namespace.clone()
+                    && self.imports.contains_key(&ns)
+                {
+                    code.push(I::call(name.to_string()));
+                } else {
+                    return Err(CompilerError::Undefined(name.clone()));
+                }
+
+                let result_reg: Register;
+
+                if use_result {
+                    let (temp_result_reg, result_alloc) = self.allocator.alloc_temp()?;
+                    result_reg = temp_result_reg;
+
+                    code.append(result_alloc);
+                    code.push(I::pop(result_reg));
+
+                    // Clean up arguments
+                    if args.len() > 1 {
+                        for _ in 0..(args.len() - 1) {
+                            code.push(I::pop(Register::Zero));
+                        }
+                    }
+                } else {
+                    result_reg = Register::Zero;
+
+                    // Clean up arguments
+                    if args.len() > 0 {
+                        for _ in 0..(args.len()) {
+                            code.push(I::pop(Register::Zero));
+                        }
+                    }
+                }
+
+                // Free argument registers
+                for reg in arg_regs {
+                    self.allocator.free_temp(reg);
+                }
+
+                Ok((result_reg, code))
+            }
+
+            Expression::IndexAccess {
+                expr,
+                index,
+                type_id,
+            } => {
+                let (expr_reg, expr_alloc) =
+                    self.generate_expression(expr, true, func_body)?;
+                code.append(expr_alloc);
+
+                let (index_reg, index_alloc) =
+                    self.generate_expression(index, true, func_body)?;
+                code.append(index_alloc);
+
+                let (result_reg, result_alloc) = self.allocator.alloc_temp()?;
+                code.append(result_alloc);
+
+                // add the expr pointer to the index to get the final address.
+                code.push(I::add(expr_reg, index_reg, result_reg));
+                // load the value at the address.
+                code.push(I::ldw_reg(result_reg, result_reg));
+
+                self.allocator.free_temp(expr_reg);
+                self.allocator.free_temp(index_reg);
+
+                Ok((result_reg, code))
+            }
+            Expression::MemberAccess {
+                expr,
+                field_name,
+                type_id,
+            } => Err(CompilerError::Unimplemented(
+                "Structs are not yet implemented!".to_string(),
+            )),
+
+            Expression::TypeCast {
+                expr,
+                target_type,
+                type_id,
+            } => {
+                let (expr_reg, expr_code) =
+                    self.generate_expression(expr, true, func_body)?;
+
+                // not sure if we actually need to do anything here.
+                // for now we just return the previous expression.
+                Ok((expr_reg, expr_code))
+            }
+        }
+    }
+
+    // Helper for generating unique labels
+    fn get_unique_label(&mut self) -> String {
+        // You'd implement a counter here
+        static COUNTER: AtomicU32 = AtomicU32::new(0);
+
+        let val = COUNTER.fetch_add(1, std::sync::atomic::Ordering::SeqCst);
+        (val + 1).to_string()
+    }
+}

compiler/src/backend/dsa/instruction.rs

@@ -0,0 +1,797 @@
+use std::fmt;
+
+use crate::backend::dsa::registers::Register;
+
+pub struct InsBlock {
+    instructions: Vec<Instruction>,
+}
+
+impl InsBlock {
+    pub fn new() -> Self {
+        Self {
+            instructions: vec![],
+        }
+    }
+
+    pub fn insert(&mut self, index: usize, instr: Instruction) {
+        self.instructions.insert(index, instr);
+    }
+
+    pub fn push(&mut self, instr: Instruction) {
+        self.instructions.push(instr);
+    }
+
+    pub fn append(&mut self, mut other: Self) {
+        self.instructions.append(&mut other.instructions);
+    }
+
+    pub fn extend(&mut self, instrs: impl IntoIterator<Item = Instruction>) {
+        self.instructions.extend(instrs);
+    }
+
+    pub fn is_empty(&self) -> bool {
+        self.instructions.is_empty()
+    }
+
+    pub fn len(&self) -> usize {
+        self.instructions.len()
+    }
+
+    pub fn iter(&self) -> impl Iterator<Item = &Instruction> {
+        self.instructions.iter()
+    }
+}
+
+impl From<Vec<Instruction>> for InsBlock {
+    fn from(instructions: Vec<Instruction>) -> Self {
+        Self { instructions }
+    }
+}
+
+impl From<Instruction> for InsBlock {
+    fn from(instr: Instruction) -> Self {
+        Self {
+            instructions: vec![instr],
+        }
+    }
+}
+
+#[derive(Debug, Clone)]
+pub enum Instruction {
+    // Labels and comments
+    Label(Label),
+    Comment {
+        text: String,
+        top_level: bool,
+    },
+    Newline,
+
+    // Data Directives
+    Db {
+        label: String,
+        data: Vec<u8>,
+    },
+    Dh {
+        label: String,
+        data: Vec<u16>,
+    },
+    Dw {
+        label: String,
+        data: Vec<u32>,
+    },
+    DString {
+        // alias for db.
+        label: String,
+        data: String,
+    },
+
+    Resx {
+        label: String,
+        size: u32,
+    },
+
+    // Include
+    Include {
+        name: String,
+        path: String,
+    },
+
+    // Data movement
+    Mov {
+        src: Register,
+        dest: Register,
+    },
+    Movs {
+        src: Register,
+        dest: Register,
+    },
+
+    // Memory operations
+    Ldb {
+        src: MemOperand,
+        dest: Register,
+    },
+    Ldh {
+        src: MemOperand,
+        dest: Register,
+    },
+    Ldw {
+        src: MemOperand,
+        dest: Register,
+    },
+    Stb {
+        src: Register,
+        dest: MemOperand,
+    },
+    Sth {
+        src: Register,
+        dest: MemOperand,
+    },
+    Stw {
+        src: Register,
+        dest: MemOperand,
+    },
+
+    // Immediate loads
+    Lli {
+        imm: Imm,
+        dest: Register,
+    },
+    Lui {
+        imm: Imm,
+        dest: Register,
+    },
+    Lwi {
+        imm: Imm,
+        dest: Register,
+    },
+    LwiLabel {
+        label: String,
+        dest: Register,
+    },
+
+    // Arithmetic
+    Add {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    Sub {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    IAdd {
+        src: Register,
+        imm: Imm,
+        dest: Option<Register>,
+    },
+    ISub {
+        src: Register,
+        imm: Imm,
+        dest: Option<Register>,
+    },
+    Inc {
+        reg: Register,
+    },
+    Dec {
+        reg: Register,
+    },
+
+    // Bitwise
+    And {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    Or {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    Xor {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    Not {
+        src: Register,
+        dest: Register,
+    },
+    Nand {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    Nor {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+    Xnor {
+        src1: Register,
+        src2: Register,
+        dest: Register,
+    },
+
+    // Shifts
+    Shl {
+        src1: Register,
+        r_shamt: Register,
+        i_shamt: u16,
+        dest: Register,
+    },
+    Shr {
+        src1: Register,
+        r_shamt: Register,
+        i_shamt: u16,
+        dest: Register,
+    },
+
+    // Comparison
+    Cmp {
+        reg1: Register,
+        reg2: Register,
+    },
+
+    // Jumps
+    Jmp {
+        target: Label,
+    },
+    Jeq {
+        target: Label,
+    },
+    Jne {
+        target: Label,
+    },
+    Jgt {
+        target: Label,
+    },
+    Jge {
+        target: Label,
+    },
+    Jlt {
+        target: Label,
+    },
+    Jle {
+        target: Label,
+    },
+
+    // Stack
+    Push {
+        reg: Register,
+    },
+    Pop {
+        reg: Register,
+    },
+
+    // Function calls
+    Call {
+        target: String,
+    }, // namespace::function
+    Return,
+
+    // System
+    Hlt,
+    Nop,
+    Int {
+        code: u8,
+    },
+}
+
+pub enum DataDirective {
+    U8(Vec<u8>),
+    U16(Vec<u16>),
+    U32(Vec<u32>),
+    String(String),
+    Char(char),
+}
+
+impl fmt::Display for Instruction {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        match self {
+            Self::Label(l) => write!(f, "{}:", l),
+            Self::Newline => write!(f, ""), /* empty string as newlines are inserted */
+            // automatically.
+            Self::Comment { text, top_level } => write!(
+                f,
+                "{}",
+                text.lines()
+                    .map(|line| format!(
+                        "{}// {}",
+                        if *top_level { "" } else { "    " },
+                        line.trim(),
+                    ))
+                    .collect::<Vec<String>>()
+                    .join("\n")
+            ),
+
+            Self::Include { name, path } => write!(f, "include {name}: \"{}\"", path),
+
+            Self::Db { label, data } => write!(
+                f,
+                "db {}: {}",
+                label,
+                data.iter()
+                    .map(|&b| format!("{:#04X}", b))
+                    .collect::<Vec<String>>()
+                    .join(", ")
+            ),
+            Self::Dh { label, data } => write!(
+                f,
+                "dh {}: {}",
+                label,
+                data.iter()
+                    .map(|&b| format!("{:#06X}", b))
+                    .collect::<Vec<String>>()
+                    .join(", ")
+            ),
+            Self::Dw { label, data } => write!(
+                f,
+                "dw {}: {}",
+                label,
+                data.iter()
+                    .map(|&b| format!("{:#08X}", b))
+                    .collect::<Vec<String>>()
+                    .join(", ")
+            ),
+            Self::DString { label, data } => write!(f, "db {}: \"{}\"", label, data),
+
+            Self::Resx { label, size } => write!(f, "resx {}: {}", label, size),
+
+            Self::Mov { src, dest } => write!(f, "    mov {}, {}", src, dest),
+            Self::Movs { src, dest } => write!(f, "    movs {}, {}", src, dest),
+
+            Self::Ldb { src: addr, dest } => {
+                let (reg, offset) = reg_and_offset(addr);
+                write!(f, "    ldb {}, {}, {}", reg, dest, offset)
+            }
+            Self::Ldh { src: addr, dest } => {
+                let (reg, offset) = reg_and_offset(addr);
+                write!(f, "    ldh {}, {}, {}", reg, dest, offset)
+            }
+            Self::Ldw { src, dest } => {
+                let (reg, offset) = reg_and_offset(src);
+                write!(f, "    ldw {}, {}, {}", reg, dest, offset)
+            }
+            // Self::Ldbs { addr, dest } => {
+            //     write!(f, "    ldbs {}, {}", format_mem_operand(addr), dest)
+            // }
+            // Self::Ldhs { addr, dest } => {
+            //     write!(f, "    ldhs {}, {}", format_mem_operand(addr), dest)
+            // }
+            // Self::Ldws { addr, dest } => {
+            //     write!(f, "    ldws {}, {}", format_mem_operand(addr), dest)
+            // }
+            Self::Stb { src, dest: addr } => {
+                let (reg, offset) = reg_and_offset(addr);
+                write!(f, "    stb {}, {}, {}", src, reg, offset)
+            }
+            Self::Sth { src, dest: addr } => {
+                let (reg, offset) = reg_and_offset(addr);
+                write!(f, "    sth {}, {}, {}", src, reg, offset)
+            }
+            Self::Stw { src, dest: addr } => {
+                let (reg, offset) = reg_and_offset(addr);
+                write!(f, "    stw {}, {}, {}", src, reg, offset)
+            }
+
+            Self::Lli { imm, dest } => write!(f, "    lli {}, {}", imm, dest),
+            Self::Lui { imm, dest } => write!(f, "    lui {}, {}", imm, dest),
+            Self::Lwi { imm, dest } => write!(f, "    lwi {}, {}", imm, dest),
+            Self::LwiLabel { label, dest } => write!(f, "    lwi {}, {}", label, dest),
+
+            // arithmetic
+            Self::Add { src1, src2, dest } => {
+                write!(f, "    add {}, {}, {}", src1, src2, dest)
+            }
+            Self::Sub { src1, src2, dest } => {
+                write!(f, "    sub {}, {}, {}", src1, src2, dest)
+            }
+            Self::And { src1, src2, dest } => {
+                write!(f, "    and {}, {}, {}", src1, src2, dest)
+            }
+            Self::Or { src1, src2, dest } => {
+                write!(f, "    or {}, {}, {}", src1, src2, dest)
+            }
+            Self::Nand { src1, src2, dest } => {
+                write!(f, "    nand {}, {}, {}", src1, src2, dest)
+            }
+            Self::Xor { src1, src2, dest } => {
+                write!(f, "    xor {}, {}, {}", src1, src2, dest)
+            }
+            Self::Nor { src1, src2, dest } => {
+                write!(f, "    nor {}, {}, {}", src1, src2, dest)
+            }
+            Self::Not { src, dest } => {
+                write!(f, "    not {} {}", src, dest)
+            }
+            Self::Xnor { src1, src2, dest } => {
+                write!(f, "    xnor {}, {}, {}", src1, src2, dest)
+            }
+            Self::IAdd { src, imm, dest } => {
+                if let Some(d) = dest {
+                    write!(f, "    addi {}, {}, {}", src, imm, d)
+                } else {
+                    write!(f, "    addi {}, {}", src, imm)
+                }
+            }
+            Self::ISub { src, imm, dest } => {
+                if let Some(d) = dest {
+                    write!(f, "    subi {}, {}, {}", src, imm, d)
+                } else {
+                    write!(f, "    subi {}, {}", src, imm)
+                }
+            }
+
+            // shift instructions
+            Self::Shl {
+                src1,
+                r_shamt,
+                i_shamt,
+                dest,
+            } => {
+                write!(f, "    shl {}, {}, {}, {}", src1, r_shamt, i_shamt, dest)
+            }
+            Self::Shr {
+                src1,
+                r_shamt,
+                i_shamt,
+                dest,
+            } => {
+                write!(f, "    shl {}, {}, {}, {}", src1, r_shamt, i_shamt, dest)
+            }
+
+            // increment instructions
+            Self::Inc { reg } => write!(f, "    inc {}", reg),
+            Self::Dec { reg } => write!(f, "    dec {}", reg),
+
+            Self::Cmp { reg1, reg2 } => write!(f, "    cmp {}, {}", reg1, reg2),
+
+            // jump instructions
+            Self::Jmp { target } => write!(f, "    jmp {}", target),
+            Self::Jeq { target } => write!(f, "    jeq {}", target),
+            Self::Jne { target } => write!(f, "    jne {}", target),
+            Self::Jgt { target } => write!(f, "    jgt {}", target),
+            Self::Jge { target } => write!(f, "    jge {}", target),
+            Self::Jlt { target } => write!(f, "    jlt {}", target),
+            Self::Jle { target } => write!(f, "    jle {}", target),
+
+            // stack pseudoinstructions
+            Self::Push { reg } => write!(f, "    push {}", reg),
+            Self::Pop { reg } => write!(f, "    pop {}", reg),
+
+            // call & return pseudoinstructions
+            Self::Call { target } => write!(f, "    call {}", target),
+            Self::Return => write!(f, "    return"),
+
+            // misc instructions
+            Self::Int { code } => write!(f, "    int {}", code),
+            Self::Hlt => write!(f, "    hlt"),
+            Self::Nop => write!(f, "    nop"),
+        }
+    }
+}
+
+impl Instruction {
+    // data directives
+    pub fn db_string(label: impl Into<String>, data: impl Into<String>) -> Self {
+        Self::DString {
+            label: label.into(),
+            data: data.into(),
+        }
+    }
+
+    pub fn db_word(label: impl Into<String>, data: u32) -> Self {
+        Self::Dw {
+            label: label.into(),
+            data: vec![data],
+        }
+    }
+
+    pub fn db_bytes(label: impl Into<String>, data: &[u8]) -> Self {
+        Self::Db {
+            label: label.into(),
+            data: data.to_vec(),
+        }
+    }
+
+    // Movement
+    pub fn mov<R1, R2>(src: R1, dest: R2) -> Self
+    where
+        R1: Into<Register>,
+        R2: Into<Register>,
+    {
+        Self::Mov {
+            src: src.into(),
+            dest: dest.into(),
+        }
+    }
+
+    // Memory loads
+    pub fn ldw_reg<R>(base: R, dest: Register) -> Self
+    where
+        R: Into<Register>,
+    {
+        Self::Ldw {
+            src: MemOperand::RegIndirect(base.into()),
+            dest,
+        }
+    }
+
+    pub fn ldw_reg_offset<R>(base: R, dest: Register, offset: i32) -> Self
+    where
+        R: Into<Register>,
+    {
+        Self::Ldw {
+            src: MemOperand::RegOffset(base.into(), offset),
+            dest,
+        }
+    }
+
+    pub fn ldw_label(label: impl Into<Label>, dest: Register) -> Self {
+        Self::Ldw {
+            src: MemOperand::Label(label.into()),
+            dest,
+        }
+    }
+
+    // Memory stores
+    pub fn stw_reg<R>(src: Register, base: R) -> Self
+    where
+        R: Into<Register>,
+    {
+        Self::Stw {
+            src,
+            dest: MemOperand::RegIndirect(base.into()),
+        }
+    }
+
+    pub fn stw_reg_offset<R>(src: Register, base: R, offset: i32) -> Self
+    where
+        R: Into<Register>,
+    {
+        Self::Stw {
+            src,
+            dest: MemOperand::RegOffset(base.into(), offset),
+        }
+    }
+
+    pub fn stw_label(src: Register, label: impl Into<Label>) -> Self {
+        Self::Stw {
+            src,
+            dest: MemOperand::Label(label.into()),
+        }
+    }
+
+    // Arithmetic
+    pub fn add(src1: Register, src2: Register, dest: Register) -> Self {
+        Self::Add { src1, src2, dest }
+    }
+
+    pub fn sub(src1: Register, src2: Register, dest: Register) -> Self {
+        Self::Sub { src1, src2, dest }
+    }
+
+    pub fn and(src1: Register, src2: Register, dest: Register) -> Self {
+        Self::And { src1, src2, dest }
+    }
+
+    pub fn or(src1: Register, src2: Register, dest: Register) -> Self {
+        Self::Or { src1, src2, dest }
+    }
+
+    pub fn xor(src1: Register, src2: Register, dest: Register) -> Self {
+        Self::Xor { src1, src2, dest }
+    }
+
+    pub fn not(src: Register, dest: Register) -> Self {
+        Self::Not { src, dest }
+    }
+
+    pub fn shl(src1: Register, r_shamt: Register, i_shamt: u16, dest: Register) -> Self {
+        Self::Shl {
+            src1,
+            r_shamt,
+            i_shamt,
+            dest,
+        }
+    }
+
+    pub fn shr(src1: Register, r_shamt: Register, i_shamt: u16, dest: Register) -> Self {
+        Self::Shr {
+            src1,
+            r_shamt,
+            i_shamt,
+            dest,
+        }
+    }
+
+    pub fn iadd(src: Register, value: i64) -> Self {
+        let imm = Imm(value.unsigned_abs() as u32);
+
+        if value < 0 {
+            Self::ISub {
+                src,
+                imm,
+                dest: None,
+            }
+        } else {
+            Self::IAdd {
+                src,
+                imm,
+                dest: None,
+            }
+        }
+    }
+
+    pub fn iadd_dest(src: Register, value: i32, dest: Register) -> Self {
+        let imm = Imm(value.unsigned_abs());
+
+        if value < 0 {
+            Self::ISub {
+                src,
+                imm,
+                dest: Some(dest),
+            }
+        } else {
+            Self::IAdd {
+                src,
+                imm,
+                dest: Some(dest),
+            }
+        }
+    }
+
+    pub fn inc(reg: Register) -> Self {
+        Self::Inc { reg }
+    }
+
+    pub fn dec(reg: Register) -> Self {
+        Self::Dec { reg }
+    }
+
+    // Immediate loads
+    pub fn lwi(value: u32, dest: Register) -> Self {
+        if value > 0xFFFF {
+            Self::Lwi {
+                imm: Imm(value),
+                dest,
+            }
+        } else {
+            Self::Lli {
+                imm: Imm(value),
+                dest,
+            }
+        }
+    }
+
+    pub fn lwi_label(label: impl Into<String>, dest: Register) -> Self {
+        Self::LwiLabel {
+            label: label.into(),
+            dest,
+        }
+    }
+
+    // Control flow
+    pub fn label(name: impl Into<String>) -> Self {
+        Self::Label(Label(name.into()))
+    }
+
+    pub fn jmp(target: impl Into<Label>) -> Self {
+        Self::Jmp {
+            target: target.into(),
+        }
+    }
+
+    pub fn jeq(target: impl Into<Label>) -> Self {
+        Self::Jeq {
+            target: target.into(),
+        }
+    }
+
+    pub fn cmp(reg1: Register, reg2: Register) -> Self {
+        Self::Cmp { reg1, reg2 }
+    }
+
+    // Stack
+    pub fn push(reg: Register) -> Self {
+        Self::Push { reg }
+    }
+
+    pub fn pop(reg: Register) -> Self {
+        Self::Pop { reg }
+    }
+
+    // Functions
+    pub fn call(target: impl Into<String>) -> Self {
+        Self::Call {
+            target: target.into(),
+        }
+    }
+
+    pub fn int(code: u8) -> Self {
+        Self::Int { code }
+    }
+
+    pub fn ret() -> Self {
+        Self::Return
+    }
+
+    // Utilities
+    pub fn comment(text: impl Into<String>) -> Self {
+        Self::Comment {
+            text: text.into(),
+            top_level: false,
+        }
+    }
+
+    pub fn global_comment(text: impl Into<String>) -> Self {
+        Self::Comment {
+            text: text.into(),
+            top_level: true,
+        }
+    }
+
+    pub fn include(name: impl Into<String>, path: impl Into<String>) -> Self {
+        Self::Include {
+            name: name.into(),
+            path: path.into(),
+        }
+    }
+}
+
+// Convenience trait for Label conversion
+impl From<String> for Label {
+    fn from(s: String) -> Self {
+        Label(s)
+    }
+}
+
+impl From<&str> for Label {
+    fn from(s: &str) -> Self {
+        Label(s.to_string())
+    }
+}
+
+fn reg_and_offset(op: &MemOperand) -> (String, i32) {
+    match op {
+        MemOperand::RegIndirect(reg) => (reg.to_string(), 0),
+        MemOperand::RegOffset(reg, offset) => (reg.to_string(), *offset),
+        MemOperand::Label(label) => (label.to_string(), 0),
+        MemOperand::LabelOffset(label, offset) => (label.to_string(), *offset),
+    }
+}
+
+/// Memory operand for loads/stores
+#[derive(Debug, Clone)]
+pub enum MemOperand {
+    /// Register indirect: [reg]
+    RegIndirect(Register),
+    /// Register with offset: [reg + offset]
+    RegOffset(Register, i32),
+    /// Label: [label]
+    Label(Label),
+    /// Label with offset: [label + offset]
+    LabelOffset(Label, i32),
+}
+
+/// Immediate value (16-bit or 32-bit)
+#[derive(Debug, Clone, Copy, PartialEq, Eq)]
+pub struct Imm(pub u32);
+
+impl fmt::Display for Imm {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "{}", self.0)
+    }
+}
+
+/// Label reference
+#[derive(Debug, Clone, PartialEq, Eq)]
+pub struct Label(pub String);
+
+impl fmt::Display for Label {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "{}", self.0)
+    }
+}

compiler/src/backend/dsa/mod.rs

@@ -0,0 +1,12 @@
+use crate::model::{CompilerError, Program};
+
+mod codegen;
+mod instruction;
+mod registers;
+mod scope;
+mod variable;
+
+pub fn generate_code(ast: &Program) -> Result<String, CompilerError> {
+    let mut codegen = codegen::CodeGenerator::new(ast.clone());
+    codegen.generate()
+}

compiler/src/backend/dsa/registers.rs

@@ -0,0 +1,560 @@
+use std::{collections::HashMap, fmt};
+
+use crate::{
+    backend::dsa::instruction::{InsBlock, Instruction},
+    model::CompilerError,
+};
+
+/// Register allocator for DSA assembly generation
+/// Manages general-purpose registers (rg0-rgf) and handles stack spilling
+pub struct RegisterAllocator {
+    /// Available general-purpose registers
+    /// Maps variable names to their current location (register or stack offset)
+    variable_locations: HashMap<String, Location>,
+
+    /// Maps registers to the variables they currently hold
+    register_contents: HashMap<Register, String>,
+
+    /// Current stack offset for local variables (relative to bpr)
+    /// Starts at -4 (going downward from base pointer)
+    stack_offset: i32,
+
+    /// Track which registers are currently in use
+    in_use: Vec<(Register, bool)>,
+}
+
+#[derive(Debug, Clone)]
+pub struct Location {
+    register: Option<Register>,
+    stack: Option<i32>,
+}
+
+impl Location {
+    pub fn stack(offset: i32) -> Self {
+        Location {
+            register: None,
+            stack: Some(offset),
+        }
+    }
+
+    pub fn register(register: Register) -> Self {
+        Location {
+            register: Some(register),
+            stack: None,
+        }
+    }
+}
+
+impl RegisterAllocator {
+    pub fn new() -> Self {
+        // Initialize with available GP registers (rg0-rgf = 16 registers)
+        let in_use = vec![
+            Register::Rg0,
+            Register::Rg1,
+            Register::Rg2,
+            Register::Rg3,
+            Register::Rg4,
+            Register::Rg5,
+            Register::Rg6,
+            Register::Rg7,
+            Register::Rg8,
+            Register::Rg9,
+            Register::Rga,
+            Register::Rgb,
+            Register::Rgc,
+            Register::Rgd,
+            Register::Rge,
+            Register::Rgf,
+        ]
+        .iter()
+        .map(|&reg| (reg, false))
+        .collect();
+
+        RegisterAllocator {
+            // available_registers: registers,
+            variable_locations: HashMap::new(),
+            register_contents: HashMap::new(),
+            stack_offset: -4, // Start at -4 (first local below saved bpr)
+            in_use,
+        }
+    }
+
+    /// Allocate a temporary register for expression evaluation
+    /// Returns the register name and optionally assembly code to save it
+    pub fn alloc_temp(&mut self) -> Result<(Register, InsBlock), CompilerError> {
+        // Try to find an unused register
+
+        // println!("finding! {:#?}", self.in_use);
+
+        if let Some(reg) = self.find_free_register() {
+            self.in_use[reg as usize].1 = true;
+            return Ok((reg, InsBlock::new()));
+        }
+
+        // All registers in use - need to spill one
+        // Choose the first register with a variable we can spill
+        // Find a register to spill
+
+        // let reg_to_spill = self
+        //     .available_registers
+        //     .iter()
+        //     .find(|reg| self.register_contents.contains_key(*reg))
+        //     .cloned();
+
+        // if let Some(reg) = reg_to_spill {
+        //     // Spill this variable to stack
+        //     let spill_code = self.spill_register(&reg)?;
+        //     code.extend(spill_code);
+
+        //     self.in_use.insert(reg.clone(), true);
+        //     return Ok((reg, code));
+        // }
+
+        todo!("an efficient stack spilling algorithm. needs scope awareness.");
+
+        Err(CompilerError::Generic(
+            "All registers are used up yet there are no variables to spill to the stack"
+                .to_string(),
+        ))
+    }
+
+    // fn set_in_use(&mut self, reg: Register, in_use: bool) {
+    //     self.in_use[reg as usize].1 = in_use;
+    // }
+
+    /// Free a temporary register after use
+    /// NOTE: This will NOT free registers that contain variables!
+    /// Variables persist throughout their scope and must not be freed
+    pub fn free_temp(&mut self, reg: Register) {
+        // Check if this register contains a variable
+        if self.register_contents.contains_key(&reg) {
+            // This register holds a variable - don't free it!
+            // Variables are only freed when they go out of scope via free_var()
+            return;
+        }
+
+        // This is a true temporary - safe to free
+        if !matches!(reg, Register::Zero | Register::Null) {
+            self.in_use[reg as usize].1 = false;
+        }
+    }
+
+    pub fn free_var(&mut self, var: &str) {
+        // Check if this variable is in a register
+        if let Some(location) = self.variable_locations.get(var).cloned() {
+            if let Some(reg) = location.register
+                && !matches!(reg, Register::Zero | Register::Null)
+            {
+                self.register_contents.remove(&reg);
+                self.in_use[reg as usize].1 = false;
+            }
+
+            self.variable_locations.remove(var);
+        }
+    }
+
+    /// Allocate a register for a named variable
+    /// Returns the register and any necessary assembly code
+    pub fn alloc_var(
+        &mut self,
+        var_name: &str,
+    ) -> Result<(Register, InsBlock), CompilerError> {
+        if let Some(mut location) = self.variable_locations.get(var_name).cloned() {
+            // if the var is in a register we can use it already.
+            if let Some(reg) = location.register {
+                return Ok((reg, InsBlock::new()));
+            }
+
+            // if the variable is on the stack only, we need to get it in a register.
+            if let Some(offset) = location.stack {
+                // Variable was pushed, need to calculate actual position and update its
+                // location.
+                let (reg, mut code) = self.alloc_temp()?;
+
+                // acknowledge var is now in a reg as well.
+                location.register = Some(reg);
+
+                // Load from bpr + offset (offset is negative)
+                // code.push(format!("\tsubi bpr {} {}", -(offset + 4), reg));
+
+                code.push(Instruction::ldw_reg_offset(
+                    Register::Spr,
+                    reg,
+                    offset - self.stack_offset,
+                ));
+
+                // Update location to register
+                self.variable_locations
+                    .insert(var_name.to_string(), location);
+                self.register_contents.insert(reg, var_name.to_string());
+
+                return Ok((reg, code));
+            }
+        }
+
+        // Variable doesn't have a location yet, allocate a new register
+        let (reg, code) = self.alloc_temp()?;
+        self.variable_locations
+            .insert(var_name.to_string(), Location::register(reg));
+        self.register_contents.insert(reg, var_name.to_string());
+
+        Ok((reg, code))
+    }
+
+    /// Get the current location of a variable
+    pub fn _get_var_location(&self, var_name: &str) -> Option<&Location> {
+        self.variable_locations.get(var_name)
+    }
+
+    /// Load a variable into a register (allocating if necessary)
+    /// Returns the register and assembly code to load it
+    pub fn load_var(
+        &mut self,
+        var_name: &str,
+    ) -> Result<(Register, InsBlock), CompilerError> {
+        self.alloc_var(var_name)
+    }
+
+    /// Store a value from a register into a variable
+    /// Updates tracking and returns any necessary assembly code
+    pub fn store_var(&mut self, var_name: &str, source_reg: &Register) -> InsBlock {
+        let mut block = InsBlock::new();
+
+        // Check if variable already has a location
+        if let Some(location) = self.variable_locations.get(var_name) {
+            // if the variable exists in a register we write to that.
+            match location.register {
+                Some(reg) if reg == *source_reg => {
+                    block.push(Instruction::mov(*source_reg, reg));
+                    return block;
+                }
+                _ => (),
+            }
+
+            // if the variable exists on the stack but not a register we write here.
+            if let Some(offset) = location.stack {
+                block.push(Instruction::stw_reg_offset(
+                    *source_reg,
+                    Register::Spr,
+                    offset - self.stack_offset,
+                ));
+                return block;
+            }
+        }
+
+        // Variable doesn't exist yet, we can just use the same reg.
+        // if we can avoid a move, absolutely do that.
+
+        // if this is true then there's no permanent variable here so it's safe to use.
+        if !self.register_contents.contains_key(source_reg) {
+            self.variable_locations
+                .insert(var_name.to_string(), Location::register(*source_reg));
+            self.register_contents
+                .insert(*source_reg, var_name.to_string());
+            self.in_use[*source_reg as usize].1 = true;
+
+            return block;
+        }
+
+        // if current register isn't free, (eg is another variable) we assign somewhere
+        // else.
+        if let Some(free_reg) = self.find_free_register() {
+            self.variable_locations
+                .insert(var_name.to_string(), Location::register(free_reg));
+            self.register_contents
+                .insert(free_reg, var_name.to_string());
+            self.in_use[free_reg as usize].1 = true;
+
+            block.push(Instruction::mov(*source_reg, free_reg));
+            return block;
+        }
+
+        // No free registers - allocate on stack
+        // code.push(format!("\tstw {}, bpr, {}", source_reg, self.stack_offset));
+        // self.variable_locations
+        //     .insert(var_name.to_string(), Location::Stack(self.stack_offset));
+        // self.stack_offset -= 4; // Move to next stack slot
+        //
+        todo!("an efficient stack spilling algorithm. needs scope awareness.");
+    }
+
+    /// spill a register to the stack (WITHOUT FREEING)
+    /// DO NOT USE this if it's for a pointer!!!!
+    pub fn _spill_register(&mut self, reg: &Register) -> Result<InsBlock, CompilerError> {
+        let mut code = InsBlock::new();
+
+        // check if the variable is declared.
+        if let Some(var_name) = self.register_contents.get(reg).cloned()
+            && let Some(location) = self.variable_locations.get_mut(&var_name)
+        {
+            // check if var is on the stack
+            if let Some(offset) = location.stack {
+                code.push(Instruction::stw_reg_offset(
+                    *reg,
+                    Register::Spr,
+                    offset - self.stack_offset,
+                ));
+                return Ok(code);
+            }
+
+            // Track that we pushed one word
+            self.stack_offset -= 4;
+
+            // if the variable is not on the stack:
+            // push register to stack (spr decrements automatically)
+            let offset = self.stack_offset;
+            code.push(Instruction::push(*reg));
+
+            // Update variable location - it's now at current spr
+            // Note: We track offset from bpr for consistency
+            location.stack = Some(offset);
+
+            Ok(code)
+        } else {
+            Err(CompilerError::Generic(format!(
+                "Register {} does not contain a variable to spill!",
+                reg
+            )))
+        }
+    }
+
+    /// free a register by spilling it to the stack.
+    /// Returns assembly code to perform the spill
+    pub fn free_register(
+        &mut self,
+        reg: &Register,
+    ) -> Result<(i32, Instruction), CompilerError> {
+        // check if the variable is declared.
+        if let Some(var_name) = self.register_contents.get(reg).cloned()
+            && let Some(location) = self.variable_locations.get_mut(&var_name)
+        {
+            // check if var name is on the stack
+            if let Some(offset) = location.stack {
+                // store current register value in stack location
+                let code = Instruction::stw_reg_offset(
+                    *reg,
+                    Register::Spr,
+                    offset - self.stack_offset,
+                );
+
+                // free the register.
+                location.register = None;
+                self.register_contents.remove(reg);
+                return Ok((offset, code));
+            }
+
+            // Track that we pushed one word
+            self.stack_offset -= 4;
+
+            let offset = self.stack_offset;
+            let code = Instruction::push(*reg);
+
+            // Update variable location
+            // Note: We track offset from bpr for consistency
+            location.stack = Some(offset);
+            location.register = None;
+            self.register_contents.remove(reg);
+
+            Ok((offset, code))
+        } else {
+            Err(CompilerError::Generic(format!(
+                "Register {} does not contain a variable to spill!",
+                reg
+            )))
+        }
+    }
+
+    /// Find a free register (not currently in use)
+    fn find_free_register(&self) -> Option<Register> {
+        self.in_use
+            .iter()
+            .filter(|(_, in_use)| !*in_use)
+            .map(|(reg, _)| *reg)
+            .next()
+    }
+
+    /// Spill all registers to stack (useful before function calls)
+    pub fn _spill_all(&mut self) -> InsBlock {
+        let mut code = InsBlock::new();
+
+        let regs_to_spill: Vec<Register> =
+            self.register_contents.keys().cloned().collect();
+
+        for reg in regs_to_spill {
+            if let Ok(spill_code) = self.free_register(&reg) {
+                code.push(spill_code.1);
+            }
+        }
+
+        code
+    }
+
+    /// Get the total stack offset
+    pub fn get_stack_offset(&self) -> i32 {
+        self.stack_offset
+    }
+
+    /// Get the total stack space needed for local variables
+    pub fn _get_stack_size(&self) -> i32 {
+        -self.stack_offset // Convert negative offset to positive size
+    }
+
+    /// Reset allocator for a new function
+    pub fn reset(&mut self) {
+        self.variable_locations.clear();
+        self.register_contents.clear();
+        self.stack_offset = -4;
+        self.in_use = vec![
+            Register::Rg0,
+            Register::Rg1,
+            Register::Rg2,
+            Register::Rg3,
+            Register::Rg4,
+            Register::Rg5,
+            Register::Rg6,
+            Register::Rg7,
+            Register::Rg8,
+            Register::Rg9,
+            Register::Rga,
+            Register::Rgb,
+            Register::Rgc,
+            Register::Rgd,
+            Register::Rge,
+            Register::Rgf,
+        ]
+        .iter()
+        .map(|&reg| (reg, false))
+        .collect();
+    }
+
+    /// Get list of registers that contain variables and are in use
+    /// These need to be saved before function calls
+    pub fn get_caller_saved_registers(&self) -> Vec<Register> {
+        self.register_contents
+            .iter()
+            .filter(|(reg, _)| {
+                self.in_use
+                    .get(**reg as usize)
+                    .unwrap_or(&(Register::Null, false))
+                    .1
+            })
+            .map(|(reg, _)| *reg)
+            .collect()
+    }
+}
+
+#[derive(Debug, Copy, Clone, PartialEq, Eq, Hash)]
+pub enum Register {
+    // general purpose
+    Rg0 = 0,
+    Rg1 = 1,
+    Rg2 = 2,
+    Rg3 = 3,
+    Rg4 = 4,
+    Rg5 = 5,
+    Rg6 = 6,
+    Rg7 = 7,
+    Rg8 = 8,
+    Rg9 = 9,
+    Rga = 10,
+    Rgb = 11,
+    Rgc = 12,
+    Rgd = 13,
+    Rge = 14,
+    Rgf = 15,
+
+    // special
+    Bpr,
+    Spr,
+    Ret,
+    Acc,
+
+    // read only
+    Pcx,
+    Zero,
+
+    // null
+    Null,
+}
+
+impl Register {
+    pub fn get_gp() -> [Register; 16] {
+        [
+            Register::Rg0,
+            Register::Rg1,
+            Register::Rg2,
+            Register::Rg3,
+            Register::Rg4,
+            Register::Rg5,
+            Register::Rg6,
+            Register::Rg7,
+            Register::Rg8,
+            Register::Rg9,
+            Register::Rga,
+            Register::Rgb,
+            Register::Rgc,
+            Register::Rgd,
+            Register::Rge,
+            Register::Rgf,
+        ]
+    }
+
+    pub fn is_gp(&self) -> bool {
+        (*self as u8) < 16
+    }
+
+    pub fn from_index(idx: usize) -> Register {
+        match idx {
+            0 => Register::Rg0,
+            1 => Register::Rg1,
+            2 => Register::Rg2,
+            3 => Register::Rg3,
+            4 => Register::Rg4,
+            5 => Register::Rg5,
+            6 => Register::Rg6,
+            7 => Register::Rg7,
+            8 => Register::Rg8,
+            9 => Register::Rg9,
+            10 => Register::Rga,
+            11 => Register::Rgb,
+            12 => Register::Rgc,
+            13 => Register::Rgd,
+            14 => Register::Rge,
+            15 => Register::Rgf,
+            _ => unreachable!("this function shouldn't ever be called with idx>15"),
+        }
+    }
+}
+
+impl fmt::Display for Register {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        match self {
+            Self::Rg0 => write!(f, "rg0"),
+            Self::Rg1 => write!(f, "rg1"),
+            Self::Rg2 => write!(f, "rg2"),
+            Self::Rg3 => write!(f, "rg3"),
+            Self::Rg4 => write!(f, "rg4"),
+            Self::Rg5 => write!(f, "rg5"),
+            Self::Rg6 => write!(f, "rg6"),
+            Self::Rg7 => write!(f, "rg7"),
+            Self::Rg8 => write!(f, "rg8"),
+            Self::Rg9 => write!(f, "rg9"),
+            Self::Rga => write!(f, "rga"),
+            Self::Rgb => write!(f, "rgb"),
+            Self::Rgc => write!(f, "rgc"),
+            Self::Rgd => write!(f, "rgd"),
+            Self::Rge => write!(f, "rge"),
+            Self::Rgf => write!(f, "rgf"),
+
+            Self::Acc => write!(f, "acc"),
+            Self::Ret => write!(f, "ret"),
+            Self::Bpr => write!(f, "bpr"),
+            Self::Spr => write!(f, "spr"),
+
+            Self::Zero => write!(f, "zero"),
+            Self::Pcx => write!(f, "pcx"),
+
+            Self::Null => write!(f, "null"),
+        }
+    }
+}

compiler/src/backend/dsa/scope.rs

@@ -0,0 +1,287 @@
+use std::{cell::RefCell, collections::HashMap, ops::Deref, rc::Rc};
+
+use uuid::Uuid;
+
+use crate::{
+    backend::dsa::{
+        instruction::{InsBlock, Instruction},
+        registers::{Register, RegisterAllocator},
+        variable::Variable,
+    },
+    model::CompilerError,
+};
+
+pub struct Allocator {
+    stack_offset: i32,
+    in_use: [(Register, bool); 16],
+}
+
+pub struct TempReg(Register);
+pub struct AssignedReg(Register);
+pub struct StackSlot(i32);
+
+impl Deref for TempReg {
+    type Target = Register;
+
+    fn deref(&self) -> &Self::Target {
+        &self.0
+    }
+}
+
+impl Deref for AssignedReg {
+    type Target = Register;
+
+    fn deref(&self) -> &Self::Target {
+        &self.0
+    }
+}
+
+impl Deref for StackSlot {
+    type Target = i32;
+
+    fn deref(&self) -> &Self::Target {
+        &self.0
+    }
+}
+
+impl Allocator {
+    pub fn new() -> Self {
+        let mut in_use = [(Register::Null, false); 16];
+        in_use.copy_from_slice(&Register::get_gp().map(|r| (r, false))[0..16]);
+
+        Self {
+            stack_offset: 0,
+            in_use,
+        }
+    }
+
+    pub fn get_stack_offset(&self) -> i32 {
+        self.stack_offset
+    }
+
+    pub fn destroy_scope(&mut self, scope: &mut Scope) {
+        self.stack_offset = scope.entry_stack_offset;
+
+        for var in scope.variables.drain() {
+            if let Some(assigned) = var.1.register {
+                self.free_assigned(&assigned);
+            }
+        }
+    }
+
+    // what we need:
+
+    // - create var in register from temporary register. free temp and use it.
+    //
+    // - create var on stack from struct/array literal. return stack offset to write to.
+    //
+    // - spill var from register to stack. return stack offset to write to.
+    //
+    // - read/write var from stack+offset into register to use while preserving the stack
+    //   slot.
+    //
+    // - read / write bytes from the stack+offset in a larger variable into a register.
+
+    pub fn read_var(&mut self, var: &mut Variable) -> Result<InsBlock, CompilerError> {
+        if let Some(slot) = &mut var.stack_slot {
+            if var.register.is_none() {
+                var.register = Some(self.allocate_var()?);
+            }
+
+            if let Some(reg) = &var.register {
+                return Ok(InsBlock::from(Instruction::ldw_reg_offset(
+                    **reg,
+                    Register::Spr,
+                    **slot - self.stack_offset,
+                )));
+            }
+
+            unreachable!()
+        }
+
+        Err(CompilerError::Generic(format!(
+            "Tried to write var {} to stack but var was not assigned a reg and/or stack slot",
+            var.name
+        )))
+    }
+
+    pub fn write_var(&mut self, var: &mut Variable) -> Result<InsBlock, CompilerError> {
+        if let Some(slot) = &var.stack_slot {
+            if let Some(reg) = &var.register {
+                return Ok(InsBlock::from(Instruction::stw_reg_offset(
+                    **reg,
+                    Register::Spr,
+                    **slot - self.stack_offset,
+                )));
+            }
+        }
+
+        Err(CompilerError::Generic(format!(
+            "Tried to write var {} to stack but var was not assigned a reg and/or stack slot",
+            var.name
+        )))
+    }
+
+    pub fn spill_var(&mut self, var: &mut Variable) -> Result<InsBlock, CompilerError> {
+        if let Some(slot) = &var.stack_slot {
+            let block = self.write_var(var)?;
+            if let Some(reg) = &var.register {
+                self.free_assigned(reg);
+                var.register = None;
+            }
+
+            return Ok(block);
+        }
+
+        // var doesn't have a stack slot so we need to create one
+        if let Some(reg) = &var.register {
+            let slot = self.allocate_stack_slot(var.size);
+            let block = InsBlock::from(Instruction::push(**reg));
+
+            self.free_assigned(reg);
+            var.register = None;
+            var.stack_slot = Some(slot);
+            return Ok(block);
+        }
+
+        return Err(CompilerError::Generic(
+            "spill_var called on a variable without a register".to_string(),
+        ));
+    }
+
+    pub fn allocate_stack_slot(&mut self, size: usize) -> StackSlot {
+        self.stack_offset -= size as i32;
+        let offset = self.stack_offset;
+        StackSlot(offset)
+    }
+
+    pub fn allocate_var(&mut self) -> Result<AssignedReg, CompilerError> {
+        if let Some(reg) = self.find_free_register() {
+            Ok(AssignedReg(reg))
+        } else {
+            Err(CompilerError::Generic(
+                "No free registers available".to_string(),
+            ))
+        }
+    }
+
+    pub fn allocate_temp(&mut self) -> Result<TempReg, CompilerError> {
+        // allocates a temporary register
+        if let Some(reg) = self.find_free_register() {
+            Ok(TempReg(reg))
+        } else {
+            todo!("an efficient stack spilling algorithm. needs scope awareness.");
+        }
+    }
+
+    pub fn free_temp(&mut self, temp: &TempReg) {
+        // frees a temporary register.
+        self.in_use[**temp as usize].1 = false;
+    }
+
+    fn free_assigned(&mut self, reg: &AssignedReg) {
+        // frees a register.
+        self.in_use[**reg as usize].1 = false;
+    }
+
+    // if we have register(s) free, return the first one.
+    fn find_free_register(&mut self) -> Option<Register> {
+        self.in_use.iter_mut().find_map(|(reg, used)| {
+            if !*used {
+                *used = true;
+                Some(*reg)
+            } else {
+                None
+            }
+        })
+    }
+}
+
+pub struct FunctionContext {
+    name: String,
+    allocator: RefCell<Allocator>,
+}
+
+impl FunctionContext {
+    pub fn new(name: String) -> Self {
+        Self {
+            name,
+            allocator: RefCell::new(Allocator::new()),
+        }
+    }
+
+    pub fn get_stack_offset(&self) -> i32 {
+        self.allocator.borrow().get_stack_offset()
+    }
+}
+
+/// scope object
+pub struct Scope<'a> {
+    /// outer scope, for a function this will be the global scope.
+    parent: Option<&'a mut Scope<'a>>,
+
+    context: Rc<FunctionContext>,
+
+    /// is the scope a function body or just a loop?
+    /// depending on the type, ending a scope will have different behaviour
+    r#type: ScopeType,
+
+    /// variables
+    variables: HashMap<Uuid, Variable>,
+
+    entry_stack_offset: i32,
+}
+
+impl<'a> Scope<'a> {
+    pub fn new(parent: &'a mut Scope<'a>, r#type: ScopeType) -> Scope<'a> {
+        Self {
+            entry_stack_offset: parent.context.get_stack_offset(),
+            context: Rc::clone(&parent.context),
+            parent: Some(parent),
+            r#type,
+            variables: HashMap::new(),
+        }
+    }
+
+    pub fn close(&mut self) -> Result<(), CompilerError> {
+        // closing a scope means we need to drop all variables in scope and free
+        // registers.
+        for (name, var) in self.variables.iter() {
+            todo!()
+            // if let Some(reg) = var.allocated_register {}
+
+            // if let Some(offset) = var.bpr_offset {
+            //     self.stack_offset -= offset;
+            // }
+        }
+
+        Ok(())
+    }
+
+    pub fn alloc_temp_reg(&mut self) -> Result<(Register, InsBlock), CompilerError> {
+        todo!()
+    }
+
+    pub fn alloc_var_reg(&mut self) -> Result<(Register, InsBlock), CompilerError> {
+        todo!()
+    }
+
+    pub fn alloc_var_stack(&mut self) -> Result<(Register, InsBlock), CompilerError> {
+        todo!()
+    }
+
+    pub fn free_var_stack(&mut self) -> Result<(Register, InsBlock), CompilerError> {
+        todo!()
+    }
+
+    pub fn free_temp_reg(&mut self) -> Result<(Register, InsBlock), CompilerError> {
+        todo!()
+    }
+}
+
+#[derive(PartialEq, Copy, Clone, Debug)]
+pub enum ScopeType {
+    Function,
+    IfBlock,
+    LoopBlock,
+}

compiler/src/backend/dsa/variable.rs

@@ -0,0 +1,93 @@
+use std::{collections::HashMap, hash::Hash, rc::Rc};
+
+use uuid::Uuid;
+
+use crate::{
+    backend::dsa::{
+        instruction::InsBlock,
+        registers::Register,
+        scope::{AssignedReg, FunctionContext, Scope, StackSlot},
+    },
+    model::{CompilerError, TypeId},
+};
+
+pub struct Variable {
+    pub name: String,
+    pub uuid: Uuid,
+
+    /// the type of the variable.
+    r#type: TypeId,
+
+    /// size taken up in bytes.
+    /// if size > 4, value must be stored on the stack.
+    pub size: usize,
+
+    pub stack_slot: Option<StackSlot>,
+    pub register: Option<AssignedReg>,
+}
+
+impl Variable {
+    pub fn new_uninit(name: String, r#type: TypeId) -> Self {
+        Self {
+            name,
+            uuid: Uuid::new_v4(),
+            size: r#type.size(),
+            r#type,
+            stack_slot: None,
+            register: None,
+        }
+    }
+
+    pub fn new(
+        name: String,
+        r#type: TypeId,
+        scope: &'_ mut Scope,
+    ) -> Result<Self, CompilerError> {
+        let mut var = Self::new_uninit(name, r#type);
+        var.alloc_default(scope);
+
+        Ok(var)
+    }
+
+    fn alloc_default(&mut self, scope: &'_ mut Scope) {
+        if self.size > 4 {
+            self.alloc_stack(scope).unwrap();
+        } else {
+            self.alloc_register(scope).unwrap();
+        }
+    }
+
+    fn alloc_register(
+        &mut self,
+        scope: &'_ mut Scope,
+    ) -> Result<Register, CompilerError> {
+        if self.size > 4 {
+            return Err(CompilerError::Generic(format!(
+                "Type {} cannot be allocated a register as it has a size of {} bytes",
+                self.r#type, self.size
+            )));
+        }
+
+        todo!("integrate with register alloc logic")
+
+        // self.allocated_register = Some(...)
+    }
+
+    fn alloc_stack(&mut self, scope: &'_ mut Scope) -> Result<usize, CompilerError> {
+        todo!("integrate with stack alloc logic")
+
+        // self.bpr_offset = Some(...)
+    }
+
+    pub fn load(&mut self, scope: &'_ mut Scope) -> Result<Register, CompilerError> {
+        todo!("load var from stack to reg (if possible)")
+    }
+
+    pub fn drop(&mut self, scope: &'_ mut Scope) -> Result<(), CompilerError> {
+        Ok(())
+    }
+
+    pub fn spill(&mut self, scope: &'_ mut Scope) -> Result<(), CompilerError> {
+        todo!()
+    }
+}

compiler/src/backend/mod.rs

@@ -0,0 +1,13 @@
+use crate::model::{CompilerError, Program};
+
+mod dsa;
+
+pub fn compiler_backend(ext: &str, ast: &Program) -> Result<String, CompilerError> {
+    match ext {
+        "dsa" => Ok(dsa::generate_code(ast)?),
+        _ => Err(CompilerError::Generic(format!(
+            "File type {} not supported",
+            ext
+        ))),
+    }
+}

compiler/src/frontend/c/lexer.rs

@@ -0,0 +1,336 @@
+// ============================================================================
+// Token Types
+// ============================================================================
+
+#[derive(Debug, Clone, PartialEq)]
+pub enum TokenType {
+    // Keywords
+    Int,
+    If,
+    Else,
+    While,
+    Return,
+    Include,
+
+    // Identifiers and literals
+    Identifier(String),
+    Number(i32),
+    String(String),
+    Char(char),
+
+    // Operators
+    Plus,
+    Minus,
+    Star,
+    Slash,
+    Assign,
+    Eq,
+    Ne,
+    Lt,
+    Gt,
+    Le,
+    Ge,
+
+    // Delimiters
+    LParen,
+    RParen,
+    LBrace,
+    RBrace,
+    Semicolon,
+    Comma,
+    Colon,
+    Namespace,
+
+    Eof,
+}
+
+#[allow(unused)]
+pub enum Type {
+    Int32,
+    Int16,
+    Int8,
+    Uint32,
+    Uint16,
+    Uint8,
+    Char,
+}
+
+#[derive(Debug, Clone)]
+pub struct Token {
+    pub token_type: TokenType,
+    pub line: usize,
+    pub col: usize,
+}
+
+impl Token {
+    pub fn new(token_type: TokenType, line: usize, col: usize) -> Self {
+        Self {
+            token_type,
+            line,
+            col,
+        }
+    }
+}
+
+// ============================================================================
+// Lexer
+// ============================================================================
+
+pub struct Lexer {
+    source: Vec<char>,
+    pos: usize,
+    line: usize,
+    col: usize,
+}
+
+impl Lexer {
+    pub fn new(source: &str) -> Self {
+        Self {
+            source: source.chars().collect(),
+            pos: 0,
+            line: 1,
+            col: 1,
+        }
+    }
+
+    fn error(&self, msg: &str) -> String {
+        format!(
+            "Lexer error at line {}, col {}: {}",
+            self.line, self.col, msg
+        )
+    }
+
+    fn peek(&self, offset: usize) -> Option<char> {
+        self.source.get(self.pos + offset).copied()
+    }
+
+    fn advance(&mut self) -> Option<char> {
+        if self.pos >= self.source.len() {
+            return None;
+        }
+        let ch = self.source[self.pos];
+        self.pos += 1;
+        if ch == '\n' {
+            self.line += 1;
+            self.col = 1;
+        } else {
+            self.col += 1;
+        }
+        Some(ch)
+    }
+
+    fn skip_whitespace(&mut self) {
+        while let Some(ch) = self.peek(0) {
+            if ch.is_whitespace() {
+                self.advance();
+            } else {
+                break;
+            }
+        }
+    }
+
+    fn skip_comment(&mut self) {
+        if self.peek(0) == Some('/') && self.peek(1) == Some('/') {
+            while let Some(ch) = self.peek(0) {
+                if ch == '\n' {
+                    break;
+                }
+                self.advance();
+            }
+        }
+    }
+
+    fn read_number(&mut self) -> i32 {
+        let mut num_str = String::new();
+        while let Some(ch) = self.peek(0) {
+            if ch.is_ascii_digit() {
+                num_str.push(ch);
+                self.advance();
+            } else {
+                break;
+            }
+        }
+        num_str.parse().unwrap_or(0)
+    }
+
+    fn read_identifier(&mut self) -> String {
+        let mut ident = String::new();
+        while let Some(ch) = self.peek(0) {
+            if ch.is_alphanumeric() || ch == '_' {
+                ident.push(ch);
+                self.advance();
+            } else {
+                break;
+            }
+        }
+        ident
+    }
+
+    fn read_string(&mut self) -> Result<String, String> {
+        let mut string = String::new();
+        self.advance(); // Consume the opening quote
+
+        while let Some(ch) = self.peek(0) {
+            if ch == '"' {
+                self.advance(); // Consume the closing quote
+                return Ok(string);
+            } else if ch == '\\' {
+                self.advance(); // Consume the backslash
+                if let Some(escaped_char) = self.peek(0) {
+                    string.push(escaped_char);
+                    self.advance();
+                }
+            } else {
+                string.push(ch);
+                self.advance();
+            }
+        }
+
+        Err(String::from("Unexpected EOF"))
+    }
+
+    fn read_char(&mut self) -> Result<char, String> {
+        self.advance(); // Consume the opening quote
+
+        if let Some(ch) = self.peek(0) {
+            self.advance();
+            if self.peek(0) == Some('\'') {
+                self.advance();
+                return Ok(ch);
+            } else {
+                Err(String::from("expected closing quote"))
+            }
+        } else {
+            Err(String::from("expected character"))
+        }
+    }
+
+    pub fn tokenize(&mut self) -> Result<Vec<Token>, String> {
+        let mut tokens = Vec::new();
+
+        loop {
+            self.skip_whitespace();
+            self.skip_comment();
+
+            if self.pos >= self.source.len() {
+                break;
+            }
+
+            let line = self.line;
+            let col = self.col;
+            let ch = self.peek(0).unwrap();
+
+            let token_type = if ch.is_ascii_digit() {
+                let num = self.read_number();
+                TokenType::Number(num)
+            } else if ch == '"' {
+                let string = self.read_string()?;
+                TokenType::String(string)
+            } else if ch == '\'' {
+                let char = self.read_char()?;
+                TokenType::Char(char)
+            } else if ch.is_alphabetic() || ch == '_' {
+                let ident = self.read_identifier();
+                match ident.as_str() {
+                    "int" => TokenType::Int,
+                    "if" => TokenType::If,
+                    "else" => TokenType::Else,
+                    "while" => TokenType::While,
+                    "return" => TokenType::Return,
+                    "include" => TokenType::Include,
+                    _ => TokenType::Identifier(ident),
+                }
+            } else {
+                match ch {
+                    ':' if self.peek(1) == Some(':') => {
+                        self.advance();
+                        self.advance();
+                        TokenType::Namespace
+                    }
+                    ':' => {
+                        self.advance();
+                        TokenType::Colon
+                    }
+                    '=' if self.peek(1) == Some('=') => {
+                        self.advance();
+                        self.advance();
+                        TokenType::Eq
+                    }
+                    '!' if self.peek(1) == Some('=') => {
+                        self.advance();
+                        self.advance();
+                        TokenType::Ne
+                    }
+                    '<' if self.peek(1) == Some('=') => {
+                        self.advance();
+                        self.advance();
+                        TokenType::Le
+                    }
+                    '>' if self.peek(1) == Some('=') => {
+                        self.advance();
+                        self.advance();
+                        TokenType::Ge
+                    }
+                    '+' => {
+                        self.advance();
+                        TokenType::Plus
+                    }
+                    '-' => {
+                        self.advance();
+                        TokenType::Minus
+                    }
+                    '*' => {
+                        self.advance();
+                        TokenType::Star
+                    }
+                    '/' => {
+                        self.advance();
+                        TokenType::Slash
+                    }
+                    '=' => {
+                        self.advance();
+                        TokenType::Assign
+                    }
+                    '<' => {
+                        self.advance();
+                        TokenType::Lt
+                    }
+                    '>' => {
+                        self.advance();
+                        TokenType::Gt
+                    }
+                    '(' => {
+                        self.advance();
+                        TokenType::LParen
+                    }
+                    ')' => {
+                        self.advance();
+                        TokenType::RParen
+                    }
+                    '{' => {
+                        self.advance();
+                        TokenType::LBrace
+                    }
+                    '}' => {
+                        self.advance();
+                        TokenType::RBrace
+                    }
+                    ';' => {
+                        self.advance();
+                        TokenType::Semicolon
+                    }
+                    ',' => {
+                        self.advance();
+                        TokenType::Comma
+                    }
+                    _ => return Err(self.error(&format!("Unexpected character: {}", ch))),
+                }
+            };
+
+            tokens.push(Token::new(token_type, line, col));
+        }
+
+        tokens.push(Token::new(TokenType::Eof, self.line, self.col));
+        Ok(tokens)
+    }
+}

compiler/src/frontend/c/mod.rs

@@ -0,0 +1,25 @@
+use common::logging::log;
+
+use crate::model::{CompilerError, Program};
+use parser::Parser;
+
+pub mod lexer;
+pub mod parser;
+
+pub fn generate_ast(input: &str) -> Result<Program, CompilerError> {
+    log("Tokenising Input...");
+
+    let mut lexer = lexer::Lexer::new(&input);
+    let tokens = lexer.tokenize().map_err(|e| CompilerError::Generic(e))?;
+    // println!("{tokens:?}");
+
+    log(&format!("Parsing {} Tokens...", tokens.len()));
+
+    let mut parser = Parser::new(tokens);
+    let ast = match parser.parse() {
+        Ok(ast) => ast,
+        Err(e) => return Err(CompilerError::Generic(e)),
+    };
+
+    Ok(ast)
+}

compiler/src/frontend/c/parser.rs

@@ -0,0 +1,482 @@
+// ============================================================================
+// AST Node Types
+// ============================================================================
+
+use crate::model::{
+    BinaryOperator, Block, ConstExpr, Declaration, Dependency, Expression, Name, Program,
+    Statement, TypeId, UnaryOperator, Variable,
+};
+
+use super::lexer::{Token, TokenType};
+
+// ============================================================================
+// Parser
+// ============================================================================
+
+pub struct Parser {
+    tokens: Vec<Token>,
+    pos: usize,
+}
+
+impl Parser {
+    pub fn new(tokens: Vec<Token>) -> Self {
+        Self { tokens, pos: 0 }
+    }
+
+    fn error(&self, msg: &str) -> String {
+        let token = self.current();
+        format!(
+            "Parser error at line {}, col {}: {}",
+            token.line, token.col, msg
+        )
+    }
+
+    fn current(&self) -> &Token {
+        self.tokens
+            .get(self.pos)
+            .unwrap_or_else(|| self.tokens.last().unwrap())
+    }
+
+    fn peek(&self, offset: usize) -> &Token {
+        self.tokens
+            .get(self.pos + offset)
+            .unwrap_or_else(|| self.tokens.last().unwrap())
+    }
+
+    fn advance(&mut self) -> &Token {
+        if self.pos < self.tokens.len() - 1 {
+            self.pos += 1;
+        }
+        self.current()
+    }
+
+    fn expect(&mut self, expected: TokenType) -> Result<Token, String> {
+        let token = self.current().clone();
+        if std::mem::discriminant(&token.token_type) != std::mem::discriminant(&expected)
+        {
+            return Err(self.error(&format!(
+                "Expected {:?}, got {:?}",
+                expected, token.token_type
+            )));
+        }
+        self.advance();
+        Ok(token)
+    }
+
+    pub fn parse(&mut self) -> Result<Program, String> {
+        let mut declarations = Vec::new();
+
+        while !matches!(self.current().token_type, TokenType::Eof) {
+            declarations.push(self.parse_declaration()?);
+        }
+
+        Ok(Program { declarations })
+    }
+
+    fn parse_declaration(&mut self) -> Result<Declaration, String> {
+        // check for an import
+        if let TokenType::Include = self.current().token_type {
+            self.advance();
+
+            let name =
+                if let TokenType::Identifier(id) = self.current().clone().token_type {
+                    Some(id)
+                } else {
+                    None
+                }
+                .ok_or(String::from("Expected identifier"))?;
+
+            self.advance();
+            self.expect(TokenType::Colon)?;
+
+            let path = if let TokenType::String(id) = self.current().clone().token_type {
+                Some(id)
+            } else {
+                None
+            }
+            .ok_or(String::from("Expected string literal"))?;
+
+            self.advance();
+            return Ok(Declaration::Dependency(Dependency { name, path }));
+        }
+
+        self.expect(TokenType::Int)?;
+
+        let name = match &self.current().token_type {
+            TokenType::Identifier(s) => s.clone(),
+            _ => return Err(self.error("Expected identifier")),
+        };
+        self.advance();
+
+        match &self.current().token_type {
+            TokenType::LParen => {
+                // Function declaration
+                self.advance();
+                let mut params = Vec::<Variable>::new();
+
+                if !matches!(self.current().token_type, TokenType::RParen) {
+                    self.expect(TokenType::Int)?;
+
+                    match &self.current().token_type {
+                        TokenType::Identifier(s) => {
+                            params.push(Variable {
+                                name: s.clone(),
+                                type_id: TypeId::U32,
+                            });
+                            self.advance();
+                        }
+                        _ => return Err(self.error("Expected parameter name")),
+                    }
+
+                    while matches!(self.current().token_type, TokenType::Comma) {
+                        self.advance();
+                        self.expect(TokenType::Int)?;
+
+                        match &self.current().token_type {
+                            TokenType::Identifier(s) => {
+                                params.push(Variable {
+                                    name: s.clone(),
+                                    type_id: TypeId::U32,
+                                });
+                                self.advance();
+                            }
+                            _ => return Err(self.error("Expected parameter name")),
+                        }
+                    }
+                }
+
+                self.expect(TokenType::RParen)?;
+                let body = self.parse_block()?;
+
+                Ok(Declaration::Function {
+                    name,
+                    params,
+                    body,
+                    return_type: TypeId::U32,
+                })
+            }
+            _ => {
+                // Variable declaration
+                let init = if matches!(self.current().token_type, TokenType::Assign) {
+                    self.advance();
+
+                    if let TokenType::Number(n) = self.current().token_type {
+                        self.advance();
+                        Some(ConstExpr::Number(n))
+                    } else {
+                        return Err(self
+                            .error("Expected constant in global variable declaration"));
+                    }
+                } else {
+                    None
+                };
+
+                self.expect(TokenType::Semicolon)?;
+                Ok(Declaration::Variable {
+                    var: Variable {
+                        name,
+                        type_id: TypeId::U32,
+                    },
+                    init,
+                    is_const: false,
+                })
+            }
+        }
+    }
+
+    fn parse_block(&mut self) -> Result<Block, String> {
+        self.expect(TokenType::LBrace)?;
+        let mut statements = Vec::new();
+
+        while !matches!(self.current().token_type, TokenType::RBrace) {
+            statements.push(self.parse_statement()?);
+        }
+
+        self.expect(TokenType::RBrace)?;
+        Ok(statements)
+    }
+
+    fn parse_statement(&mut self) -> Result<Statement, String> {
+        match &self.current().token_type {
+            TokenType::LBrace => Ok(Statement::Block(self.parse_block()?)),
+            TokenType::If => self.parse_if_stmt(),
+            TokenType::While => self.parse_while_stmt(),
+            TokenType::Return => self.parse_return_stmt(),
+            TokenType::Identifier(name) => {
+                let name = name.clone();
+
+                // peek ahead for open paren (func call expr)
+                if matches!(self.peek(1).token_type, TokenType::LParen) {
+                    let expr = self.parse_expression()?; // a function call expr
+                    self.expect(TokenType::Semicolon)?;
+                    return Ok(Statement::Expression { expr });
+                }
+
+                self.advance(); // advance past identifier
+
+                // assignment expression
+                if matches!(self.current().token_type, TokenType::Assign) {
+                    self.advance();
+                    let expr = self.parse_expression()?;
+
+                    self.expect(TokenType::Semicolon)?;
+                    Ok(Statement::Assign {
+                        varname: name,
+                        value: expr,
+                    })
+                }
+                // var expression
+                else {
+                    self.expect(TokenType::Semicolon)?;
+                    Ok(Statement::Expression {
+                        expr: Expression::Variable {
+                            name: Name {
+                                name,
+                                namespace: None,
+                            },
+                            expr_type: None,
+                        },
+                    })
+                }
+            }
+            TokenType::Int => {
+                // Local variable declaration
+                self.advance();
+                let name = match &self.current().token_type {
+                    TokenType::Identifier(s) => s.clone(),
+                    _ => return Err(self.error("Expected variable name")),
+                };
+                self.advance();
+
+                let init = if matches!(self.current().token_type, TokenType::Assign) {
+                    self.advance();
+                    Some(self.parse_expression()?)
+                } else {
+                    None
+                };
+
+                self.expect(TokenType::Semicolon)?;
+
+                // Convert to assignment expression statement
+                let expr = if let Some(init_expr) = init {
+                    Statement::Assign {
+                        varname: name,
+                        value: init_expr,
+                    }
+                } else {
+                    Statement::Assign {
+                        varname: name,
+                        value: Expression::Empty,
+                    }
+                };
+
+                Ok(expr)
+            }
+            _ => {
+                let expr = if matches!(self.current().token_type, TokenType::Semicolon) {
+                    Expression::Empty
+                } else {
+                    self.parse_expression()?
+                };
+
+                self.expect(TokenType::Semicolon)?;
+                Ok(Statement::Expression { expr })
+            }
+        }
+    }
+
+    fn parse_if_stmt(&mut self) -> Result<Statement, String> {
+        self.expect(TokenType::If)?;
+        self.expect(TokenType::LParen)?;
+        let condition = self.parse_expression()?;
+        self.expect(TokenType::RParen)?;
+        let then_stmt = self.parse_block()?;
+
+        let else_stmt = if matches!(self.current().token_type, TokenType::Else) {
+            self.advance();
+            self.parse_block()?
+        } else {
+            Vec::new()
+        };
+
+        Ok(Statement::If {
+            condition,
+            then_stmt,
+            else_stmt,
+        })
+    }
+
+    fn parse_while_stmt(&mut self) -> Result<Statement, String> {
+        self.expect(TokenType::While)?;
+        self.expect(TokenType::LParen)?;
+        let condition = self.parse_expression()?;
+        self.expect(TokenType::RParen)?;
+        let body = self.parse_block()?;
+
+        Ok(Statement::While { condition, body })
+    }
+
+    fn parse_return_stmt(&mut self) -> Result<Statement, String> {
+        self.expect(TokenType::Return)?;
+
+        let expr = if matches!(self.current().token_type, TokenType::Semicolon) {
+            None
+        } else {
+            Some(self.parse_expression()?)
+        };
+
+        self.expect(TokenType::Semicolon)?;
+        Ok(Statement::Return(expr))
+    }
+
+    fn parse_expression(&mut self) -> Result<Expression, String> {
+        self.parse_comparison()
+    }
+
+    fn parse_comparison(&mut self) -> Result<Expression, String> {
+        let mut expr = self.parse_additive()?;
+
+        while let Some(op) = match &self.current().token_type {
+            TokenType::Eq => Some(BinaryOperator::Eq),
+            TokenType::Ne => Some(BinaryOperator::Ne),
+            TokenType::Lt => Some(BinaryOperator::Lt),
+            TokenType::Gt => Some(BinaryOperator::Gt),
+            TokenType::Le => Some(BinaryOperator::Le),
+            TokenType::Ge => Some(BinaryOperator::Ge),
+            _ => None,
+        } {
+            self.advance();
+            let right = Box::new(self.parse_additive()?);
+            expr = Expression::Binary {
+                op,
+                left: Box::new(expr),
+                right,
+                type_id: None,
+            };
+        }
+
+        Ok(expr)
+    }
+
+    fn parse_additive(&mut self) -> Result<Expression, String> {
+        let mut expr = self.parse_multiplicative()?;
+
+        while let Some(op) = match &self.current().token_type {
+            TokenType::Plus => Some(BinaryOperator::Add),
+            TokenType::Minus => Some(BinaryOperator::Sub),
+            _ => None,
+        } {
+            self.advance();
+            let right = Box::new(self.parse_multiplicative()?);
+            expr = Expression::Binary {
+                op,
+                left: Box::new(expr),
+                right,
+                type_id: None,
+            };
+        }
+
+        Ok(expr)
+    }
+
+    fn parse_multiplicative(&mut self) -> Result<Expression, String> {
+        let mut expr = self.parse_unary()?;
+
+        while let Some(op) = match &self.current().token_type {
+            TokenType::Star => Some(BinaryOperator::Mul),
+            TokenType::Slash => Some(BinaryOperator::Div),
+            _ => None,
+        } {
+            self.advance();
+            let right = Box::new(self.parse_unary()?);
+            expr = Expression::Binary {
+                op,
+                left: Box::new(expr),
+                right,
+                type_id: None,
+            };
+        }
+
+        Ok(expr)
+    }
+
+    fn parse_unary(&mut self) -> Result<Expression, String> {
+        let op = match &self.current().token_type {
+            TokenType::Plus => Some(UnaryOperator::Plus),
+            TokenType::Minus => Some(UnaryOperator::Minus),
+            _ => None,
+        };
+
+        if let Some(op) = op {
+            self.advance();
+            let operand = Box::new(self.parse_unary()?);
+            return Ok(Expression::Unary {
+                op,
+                operand,
+                type_id: None,
+            });
+        }
+
+        self.parse_primary()
+    }
+
+    fn parse_primary(&mut self) -> Result<Expression, String> {
+        match &self.current().token_type.clone() {
+            TokenType::Number(n) => {
+                let value = *n;
+                self.advance();
+                Ok(Expression::Number {
+                    value: value as isize,
+                    type_id: None,
+                })
+            }
+            TokenType::Identifier(name) => {
+                let name = name.clone();
+                self.advance();
+
+                if matches!(self.current().token_type, TokenType::LParen) {
+                    // Function call
+                    self.advance();
+                    let mut args = Vec::new();
+
+                    if !matches!(self.current().token_type, TokenType::RParen) {
+                        args.push(self.parse_expression()?);
+
+                        while matches!(self.current().token_type, TokenType::Comma) {
+                            self.advance();
+                            args.push(self.parse_expression()?);
+                        }
+                    }
+
+                    self.expect(TokenType::RParen)?;
+                    Ok(Expression::Call {
+                        name: Name {
+                            name,
+                            namespace: None,
+                        },
+                        args,
+                        type_id: None,
+                    })
+                } else {
+                    Ok(Expression::Variable {
+                        name: Name {
+                            name,
+                            namespace: None,
+                        },
+                        expr_type: None,
+                    })
+                }
+            }
+            TokenType::LParen => {
+                self.advance();
+                let expr = self.parse_expression()?;
+                self.expect(TokenType::RParen)?;
+                Ok(expr)
+            }
+            _ => Err(self.error(&format!(
+                "Unexpected token: {:?}",
+                self.current().token_type
+            ))),
+        }
+    }
+}

compiler/src/frontend/dsc/mod.rs

@@ -0,0 +1,38 @@
+use common::logging::log;
+
+use crate::model::{CompilerError, Program};
+use parser::{ParseResult, Parser};
+// use semantic_analyser::Analyser;
+
+pub mod lexer;
+pub mod parser;
+// pub mod semantic_analyser;
+
+pub fn generate_ast(input: &str) -> Result<Program, CompilerError> {
+    log("Tokenising Input...");
+
+    let lexer = lexer::Lexer::new(&input);
+    let tokens = lexer.collect::<Vec<_>>();
+    println!("{tokens:#?}");
+
+    log(&format!("Parsing {} Tokens...", tokens.len()));
+
+    let mut parser = Parser::new(tokens);
+    let ast = match parser.parse() {
+        ParseResult::Accept(ast) => ast,
+        ParseResult::Reject(e) => return Err(e),
+        ParseResult::Deny => {
+            return Err(CompilerError::Generic("Parser used ::Deny".to_string()));
+        }
+    };
+    // println!("{ast:#?}");
+
+    log("Analyzing AST...");
+    log("Checking Type Information...");
+
+    // let mut analyser = Analyser::new();
+    // analyser.analyse(ast.clone()).unwrap();
+
+    log("Type Checking Complete...");
+    Ok(ast)
+}

compiler/src/frontend/dsc/parser.rs

@@ -0,0 +1,987 @@
+use super::lexer::Token;
+use crate::model::{
+    AssignmentOperator, BinaryOperator, Block, Call, CompilerError, ConstExpr,
+    Declaration, Dependency, Expression, Number, Program, Statement, TypeId,
+    UnaryOperator, Variable,
+};
+use crate::{expect_tt, expect_value};
+use std::ops::{ControlFlow, FromResidual, Try};
+
+#[derive(Debug, Clone)]
+pub enum ParseResult<T, E> {
+    Accept(T),
+    Deny,
+    Reject(E),
+}
+
+pub struct Parser {
+    tokens: Vec<Token>,
+    idx: usize,
+}
+
+impl Parser {
+    pub fn new(tokens: Vec<Token>) -> Self {
+        Self { tokens, idx: 0 }
+    }
+
+    pub fn parse(&mut self) -> ParseResult<Program, CompilerError> {
+        let mut declarations = Vec::new();
+
+        while let ParseResult::Accept(_) = self.peek_next() {
+            declarations.push(self.parse_declaration()?);
+        }
+
+        ParseResult::Accept(Program { declarations })
+    }
+
+    fn parse_declaration(&mut self) -> ParseResult<Declaration, CompilerError> {
+        if expect_tt!(self.peek_next()?, Fn).accepted() {
+            return self.parse_func();
+        }
+
+        if expect_tt!(self.peek_next()?, Struct).accepted() {
+            return self.parse_struct();
+        }
+
+        if expect_tt!(self.peek_next()?, Include).accepted() {
+            // expect include keyword
+            let _ = self.next();
+
+            // expect namespace identifier
+            let name = expect_value!(self.next()?, Identifier)?;
+
+            // expect colon
+            let _ = expect_tt!(self.next()?, Colon)?;
+
+            // expect string literal (module path)
+            let path = expect_value!(self.next()?, String)?;
+
+            // expect semicolon
+            let _ = expect_tt!(self.next()?, Semicolon)?;
+
+            return ParseResult::Accept(Declaration::Dependency(Dependency {
+                name: name.name,
+                path,
+            }));
+        }
+
+        if expect_tt!(self.peek_next()?, Const, Static).accepted() {
+            let is_const = match self.next()? {
+                Token::Const => true,
+                Token::Static => false,
+                _ => {
+                    return ParseResult::Reject(CompilerError::Generic(String::from(
+                        "This can't happen!",
+                    )));
+                }
+            };
+
+            let var = self.parse_var_decl()?;
+
+            let _ = expect_tt!(self.next()?, Assign)?;
+
+            let value = self.next()?;
+            let init = match value {
+                Token::String(x) => Some(ConstExpr::String(x)),
+                Token::SignedInt(x, _) => Some(ConstExpr::Number(x)),
+                Token::UnsignedInt(x, _) => Some(ConstExpr::Number(x as i32)),
+                _ => {
+                    return ParseResult::Reject(CompilerError::UnexpectedToken(
+                        value.tt().to_string(),
+                    ));
+                }
+            };
+
+            let _ = expect_tt!(self.next()?, Semicolon)?;
+
+            return ParseResult::Accept(Declaration::Variable {
+                var,
+                init,
+                is_const,
+            });
+        }
+
+        ParseResult::Reject(CompilerError::UnexpectedEndOfInput)
+    }
+
+    fn parse_struct(&mut self) -> ParseResult<Declaration, CompilerError> {
+        let _ = expect_tt!(self.next()?, Struct)?;
+        let name = expect_value!(self.next()?, Identifier)?;
+
+        let _ = expect_tt!(self.next()?, LeftBrace)?;
+
+        let mut fields = Vec::new();
+        while expect_tt!(self.peek_next()?, Identifier).accepted() {
+            let arg = self.parse_var_decl()?;
+            fields.push(arg);
+
+            if expect_tt!(self.peek_next()?, Comma).accepted() {
+                self.next()?;
+            } else {
+                break;
+            }
+        }
+
+        let _ = expect_tt!(self.next()?, RightBrace)?;
+        ParseResult::Accept(Declaration::Struct { name, fields })
+    }
+
+    fn parse_func(&mut self) -> ParseResult<Declaration, CompilerError> {
+        // expect function keyword
+        let _ = expect_tt!(self.next()?, Fn);
+        // expect function name
+        let name = expect_value!(self.next()?, Identifier)?;
+
+        // expect left paren
+        let _ = expect_tt!(self.next()?, LeftParen)?;
+
+        let mut params = Vec::new();
+        while expect_tt!(self.peek_next()?, Identifier).accepted() {
+            let arg = self.parse_var_decl()?;
+            params.push(arg);
+
+            if expect_tt!(self.peek_next()?, Comma).accepted() {
+                self.next()?;
+            } else {
+                break;
+            }
+        }
+
+        // expect right paren
+        let _ = expect_tt!(self.next()?, RightParen)?;
+
+        // see if we can parse the return type!
+        let mut return_type = TypeId::Void;
+        if expect_tt!(self.peek_next()?, RightArrow).accepted() {
+            let _ = self.next();
+            return_type = self.parse_type()?;
+        }
+
+        // expect vald block
+        if expect_tt!(self.peek_next()?, LeftBrace).accepted() {
+            ParseResult::Accept(Declaration::Function {
+                name: name.name,
+                params,
+                return_type,
+                body: self.parse_block()?,
+            })
+        } else {
+            ParseResult::Reject(CompilerError::UnexpectedToken(
+                self.peek_next()?.tt().to_string(),
+            ))
+        }
+    }
+
+    fn parse_block(&mut self) -> ParseResult<Block, CompilerError> {
+        // expect left brace
+        let _ = expect_tt!(self.next()?, LeftBrace)?;
+
+        let mut block = Vec::new();
+        while !expect_tt!(self.peek_next()?, RightBrace).accepted() {
+            block.push(self.parse_statement()?);
+        }
+
+        // expect right brace
+        let _ = expect_tt!(self.next()?, RightBrace)?;
+
+        ParseResult::Accept(block)
+    }
+
+    fn parse_statement(&mut self) -> ParseResult<Statement, CompilerError> {
+        // handle if statements
+        if expect_tt!(self.peek_next()?, If).accepted() {
+            self.next()?;
+
+            let condition = self.parse_expression()?;
+
+            let then_stmt = self.parse_block()?;
+
+            if !expect_tt!(self.peek_next()?, Else).accepted() {
+                return ParseResult::Accept(Statement::If {
+                    condition,
+                    then_stmt,
+                    else_stmt: vec![],
+                });
+            }
+
+            let _ = expect_tt!(self.next()?, Else)?;
+
+            let else_stmt = self.parse_block()?;
+
+            return ParseResult::Accept(Statement::If {
+                condition,
+                then_stmt,
+                else_stmt,
+            });
+        }
+
+        // handle while loops
+        if expect_tt!(self.peek_next()?, While).accepted() {
+            self.next()?;
+
+            // expect valid expression
+            let expr = self.parse_expression()?;
+
+            // expect valid block after
+            let block = self.parse_block()?;
+
+            // return result
+            return ParseResult::Accept(Statement::While {
+                condition: expr,
+                body: block,
+            });
+        }
+
+        // handle indefinite loops
+        if expect_tt!(self.peek_next()?, Loop).accepted() {
+            self.next()?;
+
+            // parse the inner block
+            return ParseResult::Accept(Statement::Loop(self.parse_block()?));
+        }
+
+        if expect_tt!(self.peek_next()?, Return).accepted() {
+            self.next()?;
+
+            // handle case where nothing is returned
+            if expect_tt!(self.peek_next()?, Semicolon).accepted() {
+                return ParseResult::Accept(Statement::Return(None));
+            }
+
+            let expr = self.parse_expression()?;
+            expect_tt!(self.next()?, Semicolon)?;
+
+            return ParseResult::Accept(Statement::Return(Some(expr)));
+        }
+
+        if expect_tt!(self.peek_next()?, Break).accepted() {
+            self.next()?;
+
+            // expect semicolon
+            expect_tt!(self.next()?, Semicolon)?;
+
+            // return result
+            return ParseResult::Accept(Statement::Break);
+        }
+
+        if expect_tt!(self.peek_next()?, Continue).accepted() {
+            self.next()?;
+
+            // expect semicolon
+            expect_tt!(self.next()?, Semicolon)?;
+
+            // return result
+            return ParseResult::Accept(Statement::Continue);
+        }
+
+        // handle writes to pointers!
+        if expect_tt!(self.peek_next()?, Star).accepted() {
+            self.next()?;
+
+            let left = if expect_tt!(self.peek_next()?, Identifier).accepted() {
+                let identifier = expect_value!(self.next()?, Identifier)?;
+
+                Expression::Variable {
+                    name: identifier,
+                    expr_type: None,
+                }
+            } else if expect_tt!(self.peek_next()?, LeftParen).accepted() {
+                self.next()?;
+
+                let expr = self.parse_expression()?;
+
+                let _ = expect_tt!(self.next()?, RightParen).accepted();
+
+                expr
+            } else {
+                return ParseResult::Reject(CompilerError::UnexpectedToken(
+                    self.peek_next()?.tt().to_string(),
+                ));
+            };
+
+            let _ = expect_tt!(self.next()?, Assign)?;
+
+            let right = self.parse_expression()?;
+
+            // expect semicolon
+            expect_tt!(self.next()?, Semicolon)?;
+
+            // return result
+            return ParseResult::Accept(Statement::PtrWrite {
+                ptr: left,
+                value: right,
+            });
+        }
+
+        // handle let statements (declarations)
+        if expect_tt!(self.peek_next()?, Let).accepted() {
+            self.next();
+
+            // expect variable name and type.
+            let name = self.parse_var_decl()?;
+
+            // handle uninitialised variable case
+            if expect_tt!(self.peek_next()?, Semicolon).accepted() {
+                self.next();
+                return ParseResult::Accept(Statement::Declaration {
+                    var: name,
+                    value: None,
+                });
+            }
+
+            // handle initialised case
+            // expect equals
+            let _ = expect_tt!(self.next()?, Assign)?;
+
+            // expect a valid expression
+            let expr = self.parse_expression()?;
+
+            let _ = expect_tt!(self.next()?, Semicolon);
+
+            // return statement
+            return ParseResult::Accept(Statement::Declaration {
+                var: name,
+                value: Some(expr),
+            });
+        }
+
+        // handle an in-place function call
+        if let ParseResult::Accept(name) = expect_value!(self.peek_next()?, Identifier)
+            && let ParseResult::Accept(operator) = expect_tt!(
+                self.peek(1)?,
+                Assign,
+                PlusEqual,
+                MinusEqual,
+                StarEqual,
+                SlashEqual,
+                PercentEqual,
+                AndEqual,
+                OrEqual,
+                XorEqual,
+                ShlEqual,
+                ShrEqual
+            )
+        {
+            // consume name token
+            self.next()?;
+
+            // pattern match to find operator
+            let operator = match operator {
+                Token::Assign => AssignmentOperator::Assign,
+                Token::PlusEqual => AssignmentOperator::AddAssign,
+                Token::MinusEqual => AssignmentOperator::SubAssign,
+                Token::StarEqual => AssignmentOperator::MulAssign,
+                Token::SlashEqual => AssignmentOperator::DivAssign,
+                Token::PercentEqual => AssignmentOperator::ModAssign,
+                Token::AndEqual => AssignmentOperator::AndAssign,
+                Token::OrEqual => AssignmentOperator::OrAssign,
+                Token::XorEqual => AssignmentOperator::XorAssign,
+                Token::ShlEqual => AssignmentOperator::LeftShiftAssign,
+                Token::ShrEqual => AssignmentOperator::RightShiftAssign,
+                _ => {
+                    return ParseResult::Reject(CompilerError::UnexpectedToken(
+                        self.peek_next()?.tt().to_string(),
+                    ));
+                }
+            };
+
+            // consume operator token
+            self.next()?;
+
+            let value = self.parse_expression()?;
+
+            let _ = expect_tt!(self.next()?, Semicolon);
+
+            return ParseResult::Accept(Statement::Assign {
+                varname: name.name,
+                operator,
+                value,
+            });
+        }
+
+        // parse an expression and a semicolon
+        let expr = self.parse_expression()?;
+        let _ = expect_tt!(self.next()?, Semicolon)?;
+
+        ParseResult::Accept(Statement::Expression { expr })
+    }
+
+    fn parse_expression(&mut self) -> ParseResult<Expression, CompilerError> {
+        self.parse_logical_or()
+    }
+
+    fn parse_logical_or(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_logical_and()?;
+
+        let op = match self.peek_next()? {
+            Token::LogicalOr => BinaryOperator::LogicalOr,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_logical_or()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_logical_and(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_bitwise_or()?;
+
+        let op = match self.peek_next()? {
+            Token::LogicalAnd => BinaryOperator::LogicalAnd,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_logical_and()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_bitwise_or(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_bitwise_xor()?;
+
+        let op = match self.peek_next()? {
+            Token::Pipe => BinaryOperator::BitwiseOr,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_bitwise_or()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_bitwise_xor(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_bitwise_and()?;
+
+        let op = match self.peek_next()? {
+            Token::Caret => BinaryOperator::BitwiseXor,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_bitwise_xor()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_bitwise_and(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_comparison()?;
+
+        let op = match self.peek_next()? {
+            Token::Ampersand => BinaryOperator::BitwiseAnd,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_bitwise_and()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_comparison(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_shift()?;
+
+        let op = match self.peek_next()? {
+            Token::EqualEqual => BinaryOperator::Equal,
+            Token::BangEqual => BinaryOperator::NotEqual,
+            Token::Less => BinaryOperator::LessThan,
+            Token::Greater => BinaryOperator::GreaterThan,
+            Token::LessEqual => BinaryOperator::LessOrEqual,
+            Token::GreaterEqual => BinaryOperator::GreaterOrEqual,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_comparison()?),
+            type_id: Some(TypeId::Bool),
+        })
+    }
+
+    fn parse_shift(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_additive()?;
+
+        let op = match self.peek_next()? {
+            Token::LeftShift => BinaryOperator::LeftShift,
+            Token::RightShift => BinaryOperator::RightShift,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_shift()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_additive(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_multiplicative()?;
+
+        let op = match self.peek_next()? {
+            Token::Plus => BinaryOperator::Add,
+            Token::Minus => BinaryOperator::Sub,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_additive()?),
+            type_id: Some(TypeId::U32),
+        })
+    }
+
+    fn parse_multiplicative(&mut self) -> ParseResult<Expression, CompilerError> {
+        let left = self.parse_unary()?;
+
+        let op = match self.peek_next()? {
+            Token::Star => BinaryOperator::Mul,
+            Token::Slash => BinaryOperator::Div,
+            _ => return ParseResult::Accept(left),
+        };
+
+        self.next()?;
+        ParseResult::Accept(Expression::Binary {
+            op,
+            left: Box::new(left),
+            right: Box::new(self.parse_multiplicative()?),
+            type_id: None,
+        })
+    }
+
+    fn parse_unary(&mut self) -> ParseResult<Expression, CompilerError> {
+        let op = match self.peek_next()? {
+            // prefix inc/dec
+            Token::PlusPlus => UnaryOperator::Increment,
+            Token::MinusMinus => UnaryOperator::Decrement,
+
+            // arithmetic
+            Token::Plus => UnaryOperator::Plus,
+            Token::Minus => UnaryOperator::Minus,
+
+            // pointer
+            Token::Star => UnaryOperator::Dereference,
+            Token::Ampersand => UnaryOperator::AddressOf,
+
+            // boolean
+            Token::Bang => UnaryOperator::LogicalNot,
+            Token::Tilde => UnaryOperator::BitwiseNot,
+
+            Token::SizeOf => UnaryOperator::SizeOf,
+            _ => {
+                let expr = self.parse_primary()?;
+                return self.parse_postfix(expr);
+            }
+        };
+
+        self.next()?;
+        let operand = Box::new(self.parse_unary()?);
+        ParseResult::Accept(Expression::Unary {
+            op,
+            operand,
+            type_id: None,
+        })
+    }
+
+    fn parse_postfix(
+        &mut self,
+        mut expr: Expression,
+    ) -> ParseResult<Expression, CompilerError> {
+        loop {
+            match self.peek_next()? {
+                // Type cast: expr as Type
+                Token::As => {
+                    self.next()?; // consume 'as'
+                    let target_type = self.parse_type()?;
+                    expr = Expression::TypeCast {
+                        expr: Box::new(expr),
+                        target_type,
+                        type_id: None,
+                    };
+                }
+
+                // Postfix increment/decrement
+                Token::PlusPlus => {
+                    self.next()?;
+                    expr = Expression::UnaryPostfix {
+                        op: UnaryOperator::Increment,
+                        operand: Box::new(expr),
+                        type_id: None,
+                    };
+                }
+                Token::MinusMinus => {
+                    self.next()?;
+                    expr = Expression::UnaryPostfix {
+                        op: UnaryOperator::Decrement,
+                        operand: Box::new(expr),
+                        type_id: None,
+                    };
+                }
+
+                // Array indexing: expr[index]
+                Token::LeftBracket => {
+                    self.next()?; // consume '['
+                    let index = Box::new(self.parse_expression()?);
+
+                    let _ = expect_tt!(self.next()?, RightBracket)?;
+
+                    expr = Expression::IndexAccess {
+                        expr: Box::new(expr),
+                        index,
+                        type_id: None,
+                    };
+                }
+
+                // Function call: expr(args...)
+                Token::LeftParen => {
+                    self.next()?; // consume '('
+                    let mut args = Vec::new();
+
+                    if !matches!(self.peek_next()?, Token::RightParen) {
+                        loop {
+                            args.push(self.parse_expression()?);
+                            if !matches!(self.peek_next()?, Token::Comma) {
+                                break;
+                            }
+                            self.next()?; // consume comma
+                        }
+                    }
+
+                    let _ = expect_tt!(self.next()?, RightParen)?;
+
+                    if let Expression::Variable { name, .. } = expr {
+                        expr = Expression::Call {
+                            func: Call { name, args },
+                            type_id: None,
+                        };
+                    }
+                }
+
+                // Member access: expr.member (if you support structs)
+                Token::Dot => {
+                    self.next()?;
+                    let field_name = expect_value!(self.next()?, Identifier)?;
+                    expr = Expression::MemberAccess {
+                        expr: Box::new(expr),
+                        field_name,
+                        type_id: None,
+                    };
+                }
+
+                // No more postfix operations
+                _ => break,
+            }
+        }
+
+        ParseResult::Accept(expr)
+    }
+
+    fn parse_primary(&mut self) -> ParseResult<Expression, CompilerError> {
+        match self.peek_next()? {
+            Token::UnsignedInt(value, type_id) => {
+                self.next()?;
+                ParseResult::Accept(Expression::Number(Number::Unsigned(value, type_id)))
+            }
+            Token::SignedInt(value, type_id) => {
+                self.next()?;
+                ParseResult::Accept(Expression::Number(Number::Signed(value, type_id)))
+            }
+            Token::String(value) => {
+                self.next()?;
+                ParseResult::Accept(Expression::StringLiteral(value))
+            }
+            Token::Char(value) => {
+                self.next()?;
+                ParseResult::Accept(Expression::CharLiteral(value))
+            }
+
+            Token::Identifier(name) => {
+                self.next()?;
+
+                // if the next token isn't the beginning of a struct literal this is just
+                // an identifier.
+                if !expect_tt!(self.peek_next()?, LeftBrace).accepted() {
+                    return ParseResult::Accept(Expression::Variable {
+                        name,
+                        expr_type: None,
+                    });
+                }
+
+                let _ = self.next()?;
+
+                let mut fields = Vec::new();
+                while !expect_tt!(self.peek_next()?, RightBrace).accepted() {
+                    let name = expect_value!(self.next()?, Identifier)?;
+                    let _ = expect_tt!(self.next()?, Colon)?;
+                    let expr = self.parse_expression()?;
+
+                    fields.push((name, expr));
+
+                    if expect_tt!(self.peek_next()?, Comma).accepted() {
+                        self.next()?;
+                    } else {
+                        break;
+                    }
+                }
+
+                let _ = expect_tt!(self.next()?, RightBrace)?;
+
+                ParseResult::Accept(Expression::StructLiteral {
+                    name,
+                    fields,
+                    type_id: None,
+                })
+            }
+            Token::LeftBracket => {
+                self.next()?; // consume '['
+                let mut elements = Vec::new();
+
+                if !matches!(self.peek_next()?, Token::RightBracket) {
+                    loop {
+                        elements.push(self.parse_expression()?);
+                        if !matches!(self.peek_next()?, Token::Comma) {
+                            break;
+                        }
+                        self.next()?; // consume comma
+                    }
+                }
+
+                expect_tt!(self.next()?, RightBracket)?;
+                ParseResult::Accept(Expression::ArrayLiteral {
+                    elements,
+                    type_id: None,
+                })
+            }
+            Token::LeftParen => {
+                self.next()?;
+                let expr = self.parse_expression()?;
+                let _ = expect_tt!(self.next()?, RightParen)?;
+                ParseResult::Accept(expr)
+            }
+            _ => ParseResult::Reject(CompilerError::UnexpectedToken(
+                self.peek_next()?.tt().to_string(),
+            )),
+        }
+    }
+
+    fn parse_var_decl(&mut self) -> ParseResult<Variable, CompilerError> {
+        let name = expect_value!(self.next()?, Identifier)?;
+
+        let _ = expect_tt!(self.next()?, Colon)?;
+
+        let type_id = self.parse_type()?;
+
+        ParseResult::Accept(Variable {
+            name: name.name,
+            type_id,
+        })
+    }
+
+    fn parse_type(&mut self) -> ParseResult<TypeId, CompilerError> {
+        println!("yes {:?}", self.peek_next()?);
+
+        // parse primitive or named type
+        if expect_tt!(self.peek_next()?, Identifier).accepted() {
+            return self.parse_type_identifier();
+        }
+
+        // parse array type
+        if expect_tt!(self.peek_next()?, LeftBracket).accepted() {
+            let _ = self.next()?;
+
+            let internal_type = self.parse_type()?;
+            let _ = expect_tt!(self.next()?, Semicolon)?;
+
+            let size = expect_value!(self.next()?, UnsignedInt)?;
+
+            let _ = expect_tt!(self.next()?, RightBracket)?;
+
+            return ParseResult::Accept(TypeId::Array {
+                r#type: Box::new(internal_type),
+                size: size as usize,
+            });
+        }
+
+        // parse tuple type
+        if expect_tt!(self.peek_next()?, LeftParen).accepted() {
+            let _ = self.next()?;
+
+            let mut types = Vec::new();
+            while !expect_tt!(self.peek_next()?, RightParen).accepted() {
+                types.push(self.parse_type()?);
+                if !expect_tt!(self.peek_next()?, Comma).accepted() {
+                    break;
+                }
+                let _ = self.next()?;
+            }
+            let _ = expect_tt!(self.next()?, RightParen)?;
+
+            return ParseResult::Accept(TypeId::Tuple(types));
+        }
+
+        ParseResult::Reject(CompilerError::Generic(format!(
+            "Parsing type but no valid type was detected: {:?}",
+            self.peek_next()?
+        )))
+    }
+
+    fn parse_type_identifier(&mut self) -> ParseResult<TypeId, CompilerError> {
+        // get the type name incl namespace
+        let name = expect_value!(self.next()?, Identifier)?;
+
+        let type_id = match name.name.as_str() {
+            "u32" => TypeId::U32,
+            "u16" => TypeId::U16,
+            "u8" => TypeId::U8,
+            "i32" => TypeId::I32,
+            "i16" => TypeId::I16,
+            "i8" => TypeId::I8,
+            "void" => TypeId::Void,
+            "char" => TypeId::Char,
+            "str" => TypeId::Ptr(Box::new(TypeId::Char)),
+            _ => {
+                let mut generics = Vec::new();
+                if expect_tt!(self.peek_next()?, Less).accepted() {
+                    let _ = self.next()?;
+
+                    // loop until we find the closing '>'
+                    while !expect_tt!(self.peek_next()?, Greater).accepted() {
+                        generics.push(self.parse_type()?);
+                        if !expect_tt!(self.peek_next()?, Comma).accepted() {
+                            break;
+                        }
+                        let _ = self.next()?;
+                    }
+                    let _ = expect_tt!(self.next()?, Greater)?;
+                }
+
+                TypeId::UnknownCustom { name, generics }
+            }
+        };
+
+        ParseResult::Accept(type_id)
+    }
+
+    fn next(&mut self) -> ParseResult<Token, CompilerError> {
+        if self.idx >= self.tokens.len() {
+            ParseResult::Reject(CompilerError::UnexpectedEndOfInput)
+        } else {
+            let token = self.tokens[self.idx].clone();
+            self.idx += 1;
+            ParseResult::Accept(token)
+        }
+    }
+
+    fn peek_next(&self) -> ParseResult<Token, CompilerError> {
+        if self.idx >= self.tokens.len() {
+            ParseResult::Reject(CompilerError::UnexpectedEndOfInput)
+        } else {
+            ParseResult::Accept(self.tokens[self.idx].clone())
+        }
+    }
+
+    fn peek(&self, offset: usize) -> ParseResult<Token, CompilerError> {
+        if self.idx + offset >= self.tokens.len() {
+            ParseResult::Reject(CompilerError::UnexpectedEndOfInput)
+        } else {
+            ParseResult::Accept(self.tokens[self.idx + offset].clone())
+        }
+    }
+}
+
+impl<T, E> ParseResult<T, E> {
+    pub fn accepted(&self) -> bool {
+        matches!(self, ParseResult::Accept(_))
+    }
+}
+
+pub enum ParseResultResidual<T> {
+    Deny,
+    Reject(T),
+}
+
+impl<T, E> Try for ParseResult<T, E> {
+    type Output = T;
+    type Residual = ParseResultResidual<E>;
+
+    fn from_output(output: T) -> Self {
+        ParseResult::Accept(output)
+    }
+
+    fn branch(self) -> ControlFlow<Self::Residual, Self::Output> {
+        match self {
+            ParseResult::Accept(v) => ControlFlow::Continue(v),
+            ParseResult::Deny => ControlFlow::Break(ParseResultResidual::Deny),
+            ParseResult::Reject(e) => ControlFlow::Break(ParseResultResidual::Reject(e)),
+        }
+    }
+}
+
+impl<T, E> FromResidual for ParseResult<T, E> {
+    fn from_residual(residual: ParseResultResidual<E>) -> Self {
+        match residual {
+            ParseResultResidual::Deny => ParseResult::Deny,
+            ParseResultResidual::Reject(e) => ParseResult::Reject(e),
+        }
+    }
+}
+
+#[macro_export]
+macro_rules! expect_tt {
+    ($token:expr, $($variant:ident),+) => {{
+        let token = $token.clone();
+        let tt = token.tt().to_string();
+
+        let mut vs = String::new();
+        $(
+            let s = stringify!($variant);
+            vs.push_str(s);
+            vs.push_str("|");
+        )+
+
+        match tt.as_str() {
+            $(
+                stringify!($variant) => ParseResult::Accept(token),
+            )+
+            _ => {
+                // let expected = format!("[{}]", vec![$(stringify!($variant)),+].join(" | "));
+                ParseResult::Reject(CompilerError::UnexpectedToken(tt))
+            }
+        }
+    }};
+}
+
+#[macro_export]
+macro_rules! expect_value {
+    ($expr:expr, $variant:ident) => {{
+        let tok = $expr;
+        match tok.clone() {
+            Token::$variant(first, ..) => ParseResult::Accept(first),
+            _ => {
+                ParseResult::Reject(CompilerError::UnexpectedToken(tok.tt().to_string()))
+            }
+        }
+    }};
+}

compiler/src/frontend/dsc/semantic_analyser.rs

@@ -0,0 +1,226 @@
+use std::collections::HashMap;
+
+use crate::model::{
+    BinaryOperator, // You'll need to add this to your imports
+    CompilerError,
+    Declaration,
+    Dependency,
+    Expression,
+    Program,
+    TypeId,
+    UnaryOperator,
+};
+
+pub struct Analyser {
+    symbol_table: HashMap<String, Declaration>,
+}
+
+const NUMERIC_TYPES: &[TypeId] = &[
+    TypeId::U32,
+    TypeId::I32,
+    TypeId::I16,
+    TypeId::U16,
+    TypeId::I8,
+    TypeId::U8,
+];
+
+impl Analyser {
+    pub fn new() -> Self {
+        Self {
+            symbol_table: HashMap::new(),
+        }
+    }
+
+    pub fn analyse(&mut self, ast: Program) -> Result<(), CompilerError> {
+        // build table of global symbols.
+        for dec in ast.declarations {
+            let name = match dec.clone() {
+                Declaration::Function { name, .. } => name,
+                Declaration::Variable { var, .. } => var.name,
+                Declaration::Dependency(Dependency { name, .. }) => name,
+            };
+
+            self.symbol_table.insert(name, dec);
+        }
+
+        Ok(())
+    }
+
+    fn match_type(
+        actual: TypeId,
+        expected: Option<TypeId>,
+    ) -> Result<TypeId, CompilerError> {
+        match expected {
+            Some(id) => {
+                if id != actual {
+                    Err(CompilerError::TypeMismatch(id, actual))
+                } else {
+                    Ok(actual)
+                }
+            }
+            None => Ok(actual),
+        }
+    }
+
+    fn get_type(
+        &mut self, // Changed from &self to &mut self since we modify expr
+        expr: &mut Expression,
+        expected_type: Option<TypeId>,
+    ) -> Result<TypeId, CompilerError> {
+        match expr {
+            // Correct IFF we're expecting a void type
+            Expression::Empty => Self::match_type(TypeId::Void, expected_type),
+
+            // Correct IFF we're expecting a char type
+            Expression::CharLiteral(_) => Self::match_type(TypeId::Char, expected_type),
+
+            // Correct IFF we're expecting a string slice type
+            Expression::StringLiteral(_) => {
+                Self::match_type(TypeId::Ptr(Box::new(TypeId::Char)), expected_type)
+            }
+
+            Expression::Variable { name, expr_type } => {
+                let actual = expr_type.clone().ok_or(CompilerError::UnknownType)?;
+                Self::match_type(actual, expected_type)
+            }
+
+            Expression::Number { value, type_id } => {
+                // If we already know the TypeId
+                if let Some(id) = type_id {
+                    return Self::match_type(id.clone(), expected_type);
+                }
+
+                // If we're expecting a type id, check it's numeric.
+                // TODO: add checks to make sure it's valid for its size eg u8 cant be
+                // more than 255
+                if let Some(expected) = expected_type {
+                    if NUMERIC_TYPES.contains(&expected) {
+                        *type_id = Some(expected.clone());
+                        return Ok(expected);
+                    } else {
+                        return Err(CompilerError::TypeMismatch(expected, TypeId::U32));
+                    }
+                }
+
+                // Default to i32 if no type information is available
+                *type_id = Some(TypeId::I32);
+                Ok(TypeId::I32)
+            }
+
+            Expression::Binary {
+                op,
+                left,
+                right,
+                type_id,
+            } => {
+                // For binary operations, both operands should have compatible types
+                // and the result type depends on the operation
+                let left_type = self.get_type(left, None)?;
+                let right_type = self.get_type(right, Some(left_type.clone()))?;
+
+                // For numeric operations, result has the same type as operands
+                if NUMERIC_TYPES.contains(&left_type)
+                    && NUMERIC_TYPES.contains(&right_type)
+                {
+                    *type_id = Some(left_type);
+                    Self::match_type(left_type, expected_type)
+                } else {
+                    Err(CompilerError::TypeMismatch(left_type, right_type))
+                }
+            }
+
+            Expression::Unary {
+                op,
+                operand,
+                type_id,
+            } => {
+                match op {
+                    UnaryOperator::Plus | UnaryOperator::Minus => {
+                        // Unary +/- require numeric operands
+                        let inner_type = self.get_type(operand, None)?;
+
+                        if NUMERIC_TYPES.contains(&inner_type) {
+                            *type_id = Some(inner_type.clone());
+                            Self::match_type(inner_type, expected_type)
+                        } else {
+                            Err(CompilerError::TypeMismatch(inner_type, TypeId::I32))
+                        }
+                    }
+
+                    UnaryOperator::Dereference => {
+                        // For dereference (*ptr), the operand must be a pointer
+                        // and the result type is what the pointer points to
+                        let inner_type = self.get_type(operand, None)?;
+
+                        match inner_type {
+                            TypeId::Ptr(inner) => {
+                                let deref_type = *inner;
+                                *type_id = Some(deref_type.clone());
+                                Self::match_type(deref_type, expected_type)
+                            }
+                            _ => Err(CompilerError::Generic(format!(
+                                "Cannot dereference non-pointer type: {:?}",
+                                inner_type
+                            ))),
+                        }
+                    }
+
+                    UnaryOperator::Reference => {
+                        // For reference (&var), we need to determine what we're taking
+                        // a reference to, then wrap it in a Ptr
+                        // If expected_type is Ptr(T), then operand should have type T
+                        let expected_inner = match expected_type.clone() {
+                            Some(TypeId::Ptr(inner)) => Some(*inner),
+                            _ => None,
+                        };
+
+                        let inner_type = self.get_type(operand, expected_inner)?;
+                        let ref_type = TypeId::Ptr(Box::new(inner_type));
+                        *type_id = Some(ref_type.clone());
+                        Self::match_type(ref_type, expected_type)
+                    }
+                }
+            }
+
+            Expression::Call {
+                name,
+                args,
+                type_id,
+            } => match self.symbol_table.get(&name.name) {
+                Some(Declaration::Function {
+                    params,
+                    return_type,
+                    ..
+                }) => {
+                    // check that we've given the right number of arguments.
+                    if args.len() != params.len() {
+                        return Err(CompilerError::Generic(format!(
+                            "Function {} expected {} arguments but received {}",
+                            name.name,
+                            params.len(),
+                            args.len()
+                        )));
+                    }
+
+                    for (arg, param) in args.iter_mut().zip(params.iter()) {
+                        // check that the argument type matches the parameter type.
+                        let provided_type = self.get_type(arg, Some(param.type_id))?;
+                        if provided_type != param.type_id {
+                            return Err(CompilerError::TypeMismatch(
+                                param.type_id,
+                                provided_type,
+                            ));
+                        }
+                    }
+
+                    *type_id = Some(return_type.clone());
+                    Self::match_type(return_type.clone(), expected_type)
+                }
+                _ => Err(CompilerError::Generic(format!(
+                    "Function {} not found in symbol table",
+                    name.name
+                ))),
+            },
+        }
+    }
+}

compiler/src/frontend/mod.rs

@@ -0,0 +1,15 @@
+use crate::model::{CompilerError, Program};
+
+// mod c;
+mod dsc;
+
+pub fn compiler_frontend(ext: &str, data: &str) -> Result<Program, CompilerError> {
+    match ext {
+        "dsc" => Ok(dsc::generate_ast(&data)?),
+        // "c" => Ok(c::generate_ast(&data)?),
+        _ => Err(CompilerError::Generic(format!(
+            "File type {} not supported",
+            ext
+        ))),
+    }
+}

compiler/src/lib.rs

@@ -0,0 +1,72 @@
+#![feature(try_trait_v2)]
+
+use std::path::Path;
+
+use common::logging::log;
+
+use crate::specialised::build_specialised;
+
+mod backend;
+mod frontend;
+mod model;
+mod specialised;
+
+pub fn compile_file(
+    input_path: &Path,
+    output_path: &Path,
+) -> Result<(), Box<dyn std::error::Error>> {
+    let input = std::fs::read_to_string(input_path).expect("Failed to read input file");
+
+    let input_ext = input_path
+        .extension()
+        .and_then(|s| s.to_str())
+        .unwrap_or("");
+
+    // check if we're using a specialised compiler
+    if let Some(output) = build_specialised(input_ext, &input) {
+        let result = match output {
+            Ok(output) => output,
+            Err(err) => return Err(format!("Compilation failed: {err:?}").into()),
+        };
+
+        std::fs::write(output_path, &result).expect("Failed to write output");
+
+        log(&format!(
+            "Compilation Successful ✅ \n\tSource: {}\n\tOutput: {}\n",
+            input_path.display(),
+            output_path.display(),
+        ));
+
+        return Ok(());
+    }
+
+    // Parse the input using the frontend, providing the file extension and data.
+    let ast = match frontend::compiler_frontend(input_ext, &input) {
+        Ok(ast) => ast,
+        Err(err) => return Err(format!("Compilation failed: {err:?}").into()),
+    };
+
+    println!("Parsed AST: {:#?}", ast);
+
+    let output_ext = output_path
+        .extension()
+        .and_then(|s| s.to_str())
+        .unwrap_or("");
+
+    // Generate the output using the backend with the parsed result.
+    let result = match backend::compiler_backend(output_ext, &ast) {
+        Ok(result) => result,
+        Err(err) => return Err(format!("Compilation failed: {err:?}").into()),
+    };
+
+    // println!("{result}");
+    std::fs::write(output_path, &result).expect("Failed to write output");
+
+    log(&format!(
+        "Compilation Successful ✅ \n\tSource: {}\n\tOutput: {}\n",
+        input_path.display(),
+        output_path.display(),
+    ));
+
+    Ok(())
+}

compiler/src/main.rs

@@ -0,0 +1,19 @@
+use std::path::Path;
+
+fn main() {
+    // read from input file: syntax "c_compiler <src.c> [output.dsa]"
+    let args: Vec<String> = std::env::args().collect();
+    if args.len() < 2 {
+        eprintln!("Usage: c_compiler <src.dsc> [output.dsa]");
+        return;
+    }
+
+    let input_file = &args[1];
+    let output_file = if args.len() > 2 {
+        &args[2]
+    } else {
+        "output.dsa"
+    };
+
+    compiler::compile_file(Path::new(input_file), Path::new(output_file)).unwrap();
+}

compiler/src/model.rs

@@ -0,0 +1,495 @@
+use core::fmt;
+
+#[allow(unused)]
+#[derive(Debug, Clone)]
+pub enum CompilerError {
+    UnexpectedToken(String),
+    UnexpectedEndOfInput,
+    UnexpectedCharacter(char),
+    Undefined(Name),
+    InvalidSyntax(String),
+    Generic(String),
+    UnknownType,
+    TypeMismatch(TypeId, TypeId),
+    Unimplemented(String),
+}
+
+#[derive(Debug, PartialEq, Eq, Clone)]
+pub struct Name {
+    pub name: String,
+    pub namespace: Option<String>,
+}
+impl Name {
+    pub fn new(name: impl Into<String>, namespace: Option<String>) -> Self {
+        Self {
+            name: name.into(),
+            namespace,
+        }
+    }
+}
+
+#[derive(Debug, Clone)]
+pub struct Program {
+    pub declarations: Vec<Declaration>,
+}
+
+#[allow(unused)]
+#[derive(Debug, Clone)]
+pub enum Declaration {
+    Function {
+        name: String,
+        return_type: TypeId,
+        params: Vec<Variable>,
+        body: Block,
+    },
+    Variable {
+        var: Variable,
+        init: Option<ConstExpr>,
+        is_const: bool,
+    },
+    Dependency(Dependency),
+    Struct {
+        name: Name,
+        fields: Vec<Variable>,
+    },
+}
+
+#[derive(Debug, Clone)]
+pub struct Dependency {
+    pub name: String,
+    pub path: String,
+}
+
+#[allow(unused)]
+#[derive(Debug, Clone, PartialEq)]
+pub enum TypeId {
+    U8,
+    U16,
+    U32,
+    I8,
+    I16,
+    I32,
+    Bool,
+    Char,
+    Void,
+    Ptr(Box<TypeId>),
+    Ref(Box<TypeId>),
+    Tuple(Vec<TypeId>),
+    Array {
+        r#type: Box<TypeId>,
+        size: usize,
+    },
+    UnknownCustom {
+        name: Name,
+        generics: Vec<TypeId>,
+    },
+    Struct {
+        name: Name,
+        fields: Vec<TypeId>,
+        generics: Vec<TypeId>,
+    },
+}
+
+impl TypeId {
+    pub fn size(&self) -> usize {
+        match self {
+            Self::U8 => 1,
+            Self::U16 => 2,
+            Self::U32 => 4,
+            Self::I8 => 1,
+            Self::I16 => 2,
+            Self::I32 => 4,
+            Self::Bool => 1,
+            Self::Char => 1,
+            Self::Void => 0,
+            Self::Ptr(t) => t.size(),
+            Self::Ref(t) => t.size(),
+            Self::Tuple(types) => types.iter().map(|t| t.size()).sum(),
+            Self::Array { r#type, size } => r#type.size() * size,
+            Self::UnknownCustom { .. } => 1, /* TODO: calculate type size during */
+            // semantic analysis
+            Self::Struct { fields, .. } => fields.iter().map(|t| t.size()).sum(),
+        }
+    }
+}
+
+impl fmt::Display for TypeId {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            Self::U8 => write!(f, "u8"),
+            Self::U16 => write!(f, "u16"),
+            Self::U32 => write!(f, "u32"),
+            Self::I8 => write!(f, "i8"),
+            Self::I16 => write!(f, "i16"),
+            Self::I32 => write!(f, "i32"),
+            Self::Bool => write!(f, "bool"),
+            Self::Char => write!(f, "char"),
+            Self::Void => write!(f, "void"),
+            Self::Ptr(t) => write!(f, "*{}", t),
+            Self::Ref(t) => write!(f, "&{}", t),
+            Self::Tuple(elems) => write!(
+                f,
+                "({})",
+                elems
+                    .iter()
+                    .map(|t| t.to_string())
+                    .collect::<Vec<String>>()
+                    .join(", ")
+            ),
+            Self::Array { r#type, size } => write!(f, "[{}; {}]", r#type, size),
+            Self::UnknownCustom { name, generics } => {
+                write!(
+                    f,
+                    "{}<{}>",
+                    name,
+                    generics
+                        .iter()
+                        .map(|t| t.to_string())
+                        .collect::<Vec<String>>()
+                        .join(", ")
+                )
+            }
+            Self::Struct {
+                name,
+                fields,
+                generics,
+            } => write!(
+                f,
+                "struct<{}> {} {{{}}}",
+                generics
+                    .iter()
+                    .map(|t| t.to_string())
+                    .collect::<Vec<String>>()
+                    .join(", "),
+                name,
+                fields
+                    .iter()
+                    .map(|t| t.to_string())
+                    .collect::<Vec<String>>()
+                    .join(", ")
+            ),
+        }
+    }
+}
+
+pub type Block = Vec<Statement>;
+
+#[allow(unused)]
+#[derive(Debug, Clone, PartialEq)]
+pub struct Variable {
+    pub name: String,
+    pub type_id: TypeId,
+}
+
+#[allow(unused)]
+#[derive(Debug, Clone)]
+pub enum Statement {
+    Block(Block),
+    Declaration {
+        var: Variable,
+        value: Option<Expression>,
+    },
+    Assign {
+        varname: String,
+        operator: AssignmentOperator,
+        value: Expression,
+    },
+    PtrWrite {
+        ptr: Expression,
+        value: Expression,
+    },
+    Expression {
+        expr: Expression,
+    },
+    If {
+        condition: Expression,
+        then_stmt: Block,
+        else_stmt: Block,
+    },
+    While {
+        condition: Expression,
+        body: Vec<Statement>,
+    },
+    Loop(Block),
+    Defer(Call),
+    Break,
+    Continue,
+    Return(Option<Expression>),
+}
+
+#[derive(Debug, Clone)]
+pub enum ConstExpr {
+    Number(i32),
+    String(String),
+}
+
+impl fmt::Display for ConstExpr {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        match self {
+            ConstExpr::Number(n) => write!(f, "{}", n),
+            ConstExpr::String(s) => write!(f, "\"{}\"", s),
+        }
+    }
+}
+
+#[allow(unused)]
+#[derive(Debug, Clone)]
+pub enum Expression {
+    Empty,
+    Binary {
+        op: BinaryOperator,
+        left: Box<Expression>,
+        right: Box<Expression>,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    Unary {
+        op: UnaryOperator,
+        operand: Box<Expression>,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    UnaryPostfix {
+        op: UnaryOperator,
+        operand: Box<Expression>,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    Variable {
+        name: Name,
+        expr_type: Option<TypeId>,
+    },
+    TypeCast {
+        expr: Box<Expression>,
+        target_type: TypeId,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    IndexAccess {
+        expr: Box<Expression>,
+        index: Box<Expression>,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    MemberAccess {
+        expr: Box<Expression>,
+        field_name: Name,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    Call {
+        func: Call,
+
+        // Post-Semantic Analysis
+        type_id: Option<TypeId>,
+    },
+    Number(Number),
+    StringLiteral(String),
+    CharLiteral(char),
+    ArrayLiteral {
+        elements: Vec<Expression>,
+        type_id: Option<TypeId>,
+    },
+    StructLiteral {
+        name: Name,
+        fields: Vec<(Name, Expression)>,
+        type_id: Option<TypeId>,
+    },
+}
+
+#[derive(Debug, Clone)]
+pub enum Number {
+    Signed(i32, Option<TypeId>),
+    Unsigned(u32, Option<TypeId>),
+}
+
+#[derive(Debug, Clone)]
+pub struct Call {
+    pub name: Name,
+    pub args: Vec<Expression>,
+}
+
+impl Expression {
+    pub fn is_pure(&self) -> bool {
+        match self {
+            Expression::Number { .. } => true,
+            Expression::StringLiteral(_) => true,
+            Expression::CharLiteral(_) => true,
+            Expression::Call { .. } => false,
+            Expression::Binary { left, right, .. } => left.is_pure() && right.is_pure(),
+            Expression::Unary { operand, .. } => operand.is_pure(),
+            Expression::UnaryPostfix { operand, .. } => operand.is_pure(),
+            Expression::Empty => true,
+            Expression::Variable { .. } => true,
+            Expression::TypeCast { expr, .. } => expr.is_pure(),
+            Expression::IndexAccess { expr, index, .. } => {
+                expr.is_pure() && index.is_pure()
+            }
+            Expression::MemberAccess { expr, .. } => expr.is_pure(),
+            Expression::ArrayLiteral { elements, .. } => {
+                elements.iter().all(|element| element.is_pure())
+            }
+            Expression::StructLiteral { fields, .. } => {
+                fields.iter().all(|(_, expr)| expr.is_pure())
+            }
+        }
+    }
+
+    pub fn type_id(&self) -> Result<TypeId, CompilerError> {
+        match self {
+            Expression::Number(
+                Number::Signed(_, type_id) | Number::Unsigned(_, type_id),
+            ) => type_id.clone().ok_or(CompilerError::UnknownType),
+            Expression::StringLiteral(_) => Ok(TypeId::Ptr(Box::new(TypeId::Char))),
+            Expression::CharLiteral(_) => Ok(TypeId::Char),
+            Expression::Call { type_id, .. } => {
+                type_id.clone().ok_or(CompilerError::UnknownType)
+            }
+            Expression::Binary { type_id, .. } => {
+                type_id.clone().ok_or(CompilerError::UnknownType)
+            }
+            Expression::Unary { type_id, .. } => {
+                type_id.clone().ok_or(CompilerError::UnknownType)
+            }
+            Expression::UnaryPostfix { type_id, .. } => {
+                type_id.clone().ok_or(CompilerError::UnknownType)
+            }
+            Expression::Empty => Ok(TypeId::Void),
+            Expression::Variable { expr_type, .. } => {
+                expr_type.clone().ok_or(CompilerError::UnknownType)
+            }
+            Expression::TypeCast { type_id, .. } => {
+                type_id.clone().ok_or(CompilerError::UnknownType)
+            }
+            Expression::IndexAccess { expr, .. } => expr.type_id(),
+            Expression::MemberAccess { expr, .. } => expr.type_id(),
+            Expression::ArrayLiteral { elements, .. } => {
+                let element_type = elements
+                    .first()
+                    .map_or(TypeId::Void, |e| e.type_id().unwrap_or(TypeId::Void));
+                Ok(TypeId::Array {
+                    r#type: Box::new(element_type),
+                    size: elements.len(),
+                })
+            }
+            Expression::StructLiteral { name, fields, .. } => {
+                let fields = fields
+                    .iter()
+                    .map(|(_, expr)| expr.type_id())
+                    .collect::<Result<Vec<_>, _>>()?;
+                Ok(TypeId::Struct {
+                    name: name.clone(),
+                    fields,
+                    generics: Vec::new(),
+                })
+            }
+        }
+    }
+}
+
+#[derive(Debug, Clone, PartialEq)]
+pub enum AssignmentOperator {
+    Assign,
+    AddAssign,
+    SubAssign,
+    MulAssign,
+    DivAssign,
+    ModAssign,
+    AndAssign,
+    OrAssign,
+    XorAssign,
+    LeftShiftAssign,
+    RightShiftAssign,
+}
+
+#[allow(unused)]
+#[derive(Debug, Clone, PartialEq)]
+pub enum BinaryOperator {
+    // arithmetic
+    Add,
+    Sub,
+    Mul,
+    Div,
+    Mod,
+
+    // comparison
+    Equal,
+    NotEqual,
+    LessThan,
+    GreaterThan,
+    LessOrEqual,
+    GreaterOrEqual,
+
+    // bitwise
+    BitwiseAnd,
+    BitwiseOr,
+    BitwiseXor,
+
+    // logical
+    LogicalAnd,
+    LogicalOr,
+
+    // shift
+    LeftShift,
+    RightShift,
+}
+
+impl fmt::Display for BinaryOperator {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        match self {
+            Self::Add => write!(f, "+"),
+            Self::Sub => write!(f, "-"),
+            Self::Mul => write!(f, "*"),
+            Self::Div => write!(f, "/"),
+            Self::Mod => write!(f, "%"),
+            Self::Equal => write!(f, "=="),
+            Self::NotEqual => write!(f, "!="),
+            Self::LessThan => write!(f, "<"),
+            Self::GreaterThan => write!(f, ">"),
+            Self::LessOrEqual => write!(f, "<="),
+            Self::GreaterOrEqual => write!(f, ">="),
+            Self::BitwiseAnd => write!(f, "&"),
+            Self::BitwiseOr => write!(f, "|"),
+            Self::BitwiseXor => write!(f, "^"),
+            Self::LogicalAnd => write!(f, "&&"),
+            Self::LogicalOr => write!(f, "||"),
+            Self::LeftShift => write!(f, "<<"),
+            Self::RightShift => write!(f, ">>"),
+        }
+    }
+}
+
+#[derive(Debug, Clone, PartialEq)]
+pub enum UnaryOperator {
+    Plus,
+    Minus,
+    AddressOf,
+    Dereference,
+    BitwiseNot,
+    LogicalNot,
+    Increment,
+    Decrement,
+    SizeOf,
+}
+
+impl fmt::Display for UnaryOperator {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        match self {
+            Self::Increment => write!(f, "++"),
+            Self::Decrement => write!(f, "--"),
+            Self::Plus => write!(f, "+"),
+            Self::Minus => write!(f, "-"),
+            Self::Dereference => write!(f, "*"),
+            Self::AddressOf => write!(f, "&"),
+            Self::BitwiseNot => write!(f, "~"),
+            Self::LogicalNot => write!(f, "!"),
+            Self::SizeOf => write!(f, "sizeof"),
+        }
+    }
+}

compiler/src/specialised/brainf.rs

@@ -0,0 +1,135 @@
+#[must_use]
+pub fn build(src: &str) -> String {
+    parse(src).join("\n")
+}
+
+#[must_use]
+#[expect(clippy::too_many_lines)]
+pub fn parse(src: &str) -> Vec<String> {
+    let stack = "0x10000";
+    let acc = "acc";
+    let rga = "rga";
+
+    let bpr = "bpr";
+    let spr = "spr";
+    let mut instrs = Vec::<String>::new();
+
+    // Define symbols
+    let print_start = "print";
+
+    let tokens = lex(src);
+
+    let mut idstack = Vec::<u32>::new();
+
+    // set up a stack
+    instrs.push(format!("\tlwi {}, {}", stack, bpr));
+    instrs.push(format!("\tmov {}, {}", bpr, spr));
+    // set up the data pointer
+    instrs.push(format!("{}: \t lwi 0x30000, {}", "main", rga));
+
+    for (id, tok) in tokens.iter().enumerate() {
+        match tok {
+            BfToken::Inc => {
+                instrs.push(format!("\tinc {}", acc));
+            }
+            BfToken::Dec => {
+                instrs.push(format!("\tdec {}", acc));
+            }
+            BfToken::IncPtr => {
+                instrs.push(format!("\tstw {}, {}, 0", acc, rga));
+                instrs.push(format!("\taddi {}, 4, {}", rga, rga));
+                instrs.push(format!("\tlwd {}, {}, 0", rga, acc));
+            }
+            BfToken::DecPtr => {
+                instrs.push(format!("\tstw {}, {}, 0", acc, rga));
+                instrs.push(format!("\tsubi {}, 4, {}", rga, rga));
+                instrs.push(format!("\tlwd {}, {}, 0", rga, acc));
+            }
+            BfToken::Out => {
+                instrs.push(format!("\tpush {}", acc));
+                instrs.push(format!("\tcall {}", print_start));
+                instrs.push(format!("\tpop zero"));
+            }
+            BfToken::In => {
+                instrs.push(format!("\tlwd 0x40000, {}, 0", acc));
+            }
+            BfToken::Forward => {
+                let loop_start = format!("loop_start_{}", id);
+                let loop_end = format!("loop_end_{}", id);
+                idstack.push(id as u32);
+                instrs.push(format!("\tcmp {}, zero", acc));
+                instrs.push(format!("\tjeq {}, zero", loop_end));
+                instrs.push(format!("{}: \tnop", loop_start));
+            }
+            BfToken::Back => {
+                if let Some(start_id) = idstack.pop() {
+                    let loop_start = format!("loop_start_{}", start_id);
+                    let loop_end = format!("loop_end_{}", start_id);
+                    instrs.push(format!("\tcmp {}, zero", acc));
+                    instrs.push(format!("\tjne {}, zero", loop_start));
+                    instrs.push(format!("{}: \tnop", loop_end));
+                } else {
+                    eprintln!("Warning: Unmatched ']' at position {}", id);
+                }
+            }
+        }
+    }
+
+    instrs.push("\thlt".to_string());
+
+    insert_lib(&mut instrs);
+
+    instrs
+}
+
+fn insert_lib(instrs: &mut Vec<String>) {
+    let bpr = "bpr";
+    let spr = "spr";
+    let rg0 = "rg0";
+    let rg1 = "rg1";
+
+    let print_start = "print";
+    let current = "current";
+    instrs.push(format!("\tdw {}, 0x20000", current));
+    instrs.push(format!("{}: \tpush {}", print_start, bpr));
+    instrs.push(format!("\tmov {}, {}", spr, bpr));
+    instrs.push(format!("\tlwd {}, {}, 8", bpr, rg0));
+    instrs.push(format!("\tlwd {}, {}, 0", current, rg1));
+    instrs.push(format!("\tstb {}, {}, 0", rg0, rg1));
+    instrs.push(format!("\taddi {}, 1, {}", rg1, rg1));
+    instrs.push(format!("\tstw {}, {}, 0", rg1, current));
+    instrs.push(format!("\tmov {}, {}", bpr, spr));
+    instrs.push(format!("\tpop {}", bpr));
+    instrs.push("\treturn".to_string());
+}
+
+enum BfToken {
+    Inc,
+    Dec,
+    IncPtr,
+    DecPtr,
+    Out,
+    In,
+    Forward,
+    Back,
+}
+
+fn lex(src: &str) -> Vec<BfToken> {
+    src.chars()
+        .filter_map(|c| match c {
+            '+' => Some(BfToken::Inc),
+            '-' => Some(BfToken::Dec),
+            '>' => Some(BfToken::IncPtr),
+            '<' => Some(BfToken::DecPtr),
+            '.' => Some(BfToken::Out),
+            ',' => Some(BfToken::In),
+            '[' => Some(BfToken::Forward),
+            ']' => Some(BfToken::Back),
+            _ => None,
+        })
+        .collect()
+}
+
+fn _create_symbol(id: u32) -> String {
+    format!("label_{}", id)
+}

compiler/src/specialised/mod.rs

@@ -0,0 +1,13 @@
+use crate::model::CompilerError;
+
+pub mod brainf;
+
+pub fn build_specialised(ext: &str, data: &str) -> Option<Result<String, CompilerError>> {
+    match ext {
+        "bf" => {
+            let res = brainf::build(data);
+            Some(Ok(res))
+        }
+        _ => None,
+    }
+}

docs/DSA_ISA_Specification.md

@@ -0,0 +1,429 @@
+# DSA Instruction Set Architecture Specification
+
+## Overview
+
+The Damn Simple Architecture (DSA) is a 32-bit RISC-style architecture designed for simplicity and educational purposes. This document provides the complete instruction set architecture specification, including all hardware instructions, registers, and encoding formats.
+
+## Data Types and Sizes
+
+| Type | Size | Alignment |
+|------|------|-----------|
+| Byte | 8 bits | 1-byte aligned |
+| Halfword | 16 bits | 2-byte aligned |
+| Word | 32 bits | 4-byte aligned |
+
+**Note on Endianness:**
+- Instructions and numeric data in memory: Little-endian
+- Data defined via `db/dh/dw` directives: Big-endian (assembler-specific)
+
+## Registers
+
+DSA provides 32 programmer-accessible registers plus several internal system registers.
+
+### Programmer-Accessible Registers
+
+| Hex | Register | Type | Description |
+|-----|----------|------|-------------|
+| 0x00-0x0F | **rg0-rgf** | General Purpose | 16 general-purpose registers for variables and temporary values |
+| 0x10 | **acc** | Special | Accumulator for calculations and temporary storage<br/>⚠️ Used as scratch by pseudo-instructions - volatile |
+| 0x11 | **spr** | Special | Stack pointer - points to top of stack |
+| 0x12 | **bpr** | Special | Base pointer - used for stack frame management |
+| 0x13 | **ret** | Special | Return address register - used for function returns |
+| 0x14 | **idr** | Privileged | Interrupt descriptor table address<br/>Read/write triggers protection fault in user mode |
+| 0x15 | **mmr** | Privileged | Hardware memory map table address<br/>Read/write triggers protection fault in user mode |
+| 0x16 | **zero** | Read-only | Constant zero value<br/>Reads always return 0, writes are discarded |
+| 0x17 | **noreg** | Placeholder | Indicates unused register field<br/>Read/write triggers illegal instruction fault<br/>Can also be referenced as **null** |
+| 0x18-0x1F | - | Reserved | Reserved for future use |
+
+**System Registers (indices 0x18-0x1C):**
+These exist in the encoding space but are internal to the CPU implementation:
+
+| Hex | Register | Description |
+|-----|----------|-------------|
+| 0x18 | **mar** | Memory Address Register (CPU internal) |
+| 0x19 | **mdr** | Memory Data Register (CPU internal) |
+| 0x1A | **sts** | Status Register (CPU internal) |
+| 0x1B | **cir** | Current Instruction Register (CPU internal) |
+| 0x1C | **pcx** | Program Counter (read-only, special access) |
+
+**Note on PCX (Program Counter):**
+- PCX can be read in certain contexts (e.g., stored during CALL)
+- Writing to PCX triggers a protection fault
+- PCX is automatically updated by jump and branch instructions
+
+### Status Register (STS) Layout
+
+The status register is a 32-bit register with the following flag bits:
+
+| Bit | Name | Description | Boot Value |
+|-----|------|-------------|------------|
+| 0 | **Equal** | Set if last comparison result was equal | 0 |
+| 1 | **GreaterThan** | Set if last comparison result was greater than | 0 |
+| 2 | **GreaterThanOrEqual** | Set if last comparison was greater than or equal | 0 |
+| 3 | **LessThan** | Set if last comparison result was less than | 0 |
+| 4 | **LessThanOrEqual** | Set if last comparison was less than or equal | 0 |
+| 5 | **Zero** | Set if last arithmetic/logic operation result was zero | 0 |
+| 6-31 | - | Reserved | 0 |
+
+## Instruction Encoding Formats
+
+DSA uses three instruction encoding formats:
+
+### R-Type (Register) Instructions
+
+Used for operations with register operands only, including shifts.
+
+```
+ 31-26  |  25-21  |  20-16  |  15-11  |  10-6  |  5-0
+--------+---------+---------+---------+--------+-------
+ Opcode | SrcReg1 | SrcReg2 | DestReg | ShiftAmt | Unused
+```
+
+- **Opcode** (6 bits): Instruction operation code
+- **SrcReg1** (5 bits): First source register
+- **SrcReg2** (5 bits): Second source register  
+- **DestReg** (5 bits): Destination register
+- **ShiftAmt** (5 bits): Shift amount (for shift instructions only, must be 0 otherwise)
+- **Unused** (6 bits): Must be 0
+
+**Important Rules:**
+- ShiftAmt must be 0 for non-shift instructions (else illegal instruction fault)
+- Unused register fields must be set to `noreg` (0x17) if not used
+- Using registers in unexpected positions may cause illegal instruction fault
+
+### I-Type (Immediate) Instructions
+
+Used for operations with a 16-bit immediate value.
+
+```
+ 31-26  |  25-21  |  20-16  |    15-0
+--------+---------+---------+-------------
+ Opcode | SrcReg  | DestReg | 16-bit Immediate
+```
+
+- **Opcode** (6 bits): Instruction operation code
+- **SrcReg** (5 bits): Source register (base for memory ops)
+- **DestReg** (5 bits): Destination register (or offset register for jumps)
+- **Immediate** (16 bits): Signed 16-bit immediate value or offset
+
+**Usage:**
+- Arithmetic: Immediate is a signed value
+- Memory access: Immediate is a signed byte offset from base address
+- Branches: Immediate is a signed offset added to base register
+- Literal loads: Immediate is unsigned 16-bit value
+
+### J-Type (Jump) Instructions
+
+Used for absolute jumps with large address ranges.
+
+```
+ 31-26  |         25-0
+--------+----------------------
+ Opcode |   26-bit Address
+```
+
+- **Opcode** (6 bits): Jump instruction code
+- **Address** (26 bits): Partial address for jump
+
+**Address Calculation:**
+1. Left-shift the 26-bit address by 2 (word alignment)
+2. OR with upper 4 bits of current PCX
+3. Result is final 32-bit jump address
+
+**Jump Range:** 256MB region around current PC (±128MB)
+
+**Note:** J-type instructions are defined but currently unused. Use I-type JMP with register addressing for all jumps.
+
+## Hardware Instructions
+
+### Data Movement
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x00 | **NOP** | R | - | No operation - does nothing |
+| 0x01 | **MOV** | R | SrcReg, DestReg | Copy value from SrcReg to DestReg |
+| 0x02 | **MOVS** | R | SrcReg, DestReg | Copy with sign extension to fill 32 bits |
+
+**MOV/MOVS Details:**
+- MOV performs direct copy (all 32 bits)
+- MOVS sign-extends the value (useful after byte/halfword loads)
+- Both instructions set the Zero flag if result is zero
+
+### Memory Access - Load Instructions
+
+All loads require proper alignment or trigger an alignment fault.
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x03 | **LDB** | I | BaseReg, DestReg, Offset | Load byte (8-bit), zero-extend to 32 bits |
+| 0x04 | **LDBS** | I | BaseReg, DestReg, Offset | Load byte (8-bit), sign-extend to 32 bits |
+| 0x05 | **LDH** | I | BaseReg, DestReg, Offset | Load halfword (16-bit), zero-extend to 32 bits |
+| 0x06 | **LDHS** | I | BaseReg, DestReg, Offset | Load halfword (16-bit), sign-extend to 32 bits |
+| 0x07 | **LDW** | I | BaseReg, DestReg, Offset | Load word (32-bit) |
+
+**Load Operation:**
+- Effective address = BaseReg + SignExtend(Offset)
+- Offset is a signed 16-bit value
+- Alignment requirements:
+  - LDB/LDBS: No alignment required (byte-aligned)
+  - LDH/LDHS: Must be 2-byte aligned
+  - LDW: Must be 4-byte aligned
+
+**Encoding Note:** 
+In machine code, the order is: BaseReg (SrcReg field), DestReg field, Offset (Immediate field)
+
+### Memory Access - Store Instructions
+
+All stores require proper alignment or trigger an alignment fault.
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x08 | **STB** | I | SrcReg, BaseReg, Offset | Store byte (8-bit) to memory |
+| 0x09 | **STH** | I | SrcReg, BaseReg, Offset | Store halfword (16-bit) to memory |
+| 0x0A | **STW** | I | SrcReg, BaseReg, Offset | Store word (32-bit) to memory |
+
+**Store Operation:**
+- Effective address = BaseReg + SignExtend(Offset)
+- Offset is a signed 16-bit value
+- Only the relevant bits are stored (8, 16, or 32)
+- Alignment requirements:
+  - STB: No alignment required (byte-aligned)
+  - STH: Must be 2-byte aligned
+  - STW: Must be 4-byte aligned
+
+**Encoding Note:**
+In machine code: SrcReg (SrcReg field), BaseReg (DestReg field), Offset (Immediate field)
+
+### Immediate Load Instructions
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x0B | **LLI** | I | Value, DestReg | Load 16-bit value into lower 16 bits<br/>⚠️ **CLEARS upper 16 bits!** |
+| 0x0C | **LUI** | I | Value, DestReg | Load 16-bit value into upper 16 bits<br/>Lower 16 bits unchanged |
+
+**Usage for 32-bit Values:**
+```
+LLI 0x1234, rg0    ; rg0 = 0x00001234
+LUI 0xABCD, rg0    ; rg0 = 0xABCD1234
+```
+
+**⚠️ CRITICAL:** Always execute LLI before LUI, as LLI clears the upper 16 bits!
+
+**Note on LUI:** The assembler may shift the immediate value right by 16 bits when encoding, so specify the upper 16 bits directly (e.g., `LUI 0xABCD, rg0` not `LUI 0xABCD0000, rg0`).
+
+**Encoding Note:**
+In machine code: Value (Immediate field), DestReg (SrcReg field for LLI, SrcReg field for LUI)
+
+### Jump and Branch Instructions
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x0D | **JMP** | I | Offset, BaseReg | Unconditional jump to (BaseReg + Offset) |
+| 0x0E | **JEQ** | I | Offset, BaseReg | Jump if Equal flag set |
+| 0x0F | **JNE** | I | Offset, BaseReg | Jump if Equal flag NOT set |
+| 0x10 | **JGT** | I | Offset, BaseReg | Jump if GreaterThan flag set |
+| 0x11 | **JGE** | I | Offset, BaseReg | Jump if GreaterThan OR Equal flag set |
+| 0x12 | **JLT** | I | Offset, BaseReg | Jump if LessThan flag set |
+| 0x13 | **JLE** | I | Offset, BaseReg | Jump if LessThan OR Equal flag set |
+
+**Jump Calculation:**
+- Target address = BaseReg + SignExtend(Offset)
+- If BaseReg = zero, this becomes absolute addressing with Offset
+- If BaseReg = ret, this becomes return-style addressing
+- Conditional jumps check flags in STS register
+
+**Common Patterns:**
+```
+JMP label, zero     ; Absolute jump to label address
+JMP 0, ret          ; Jump to address in ret register
+JMP 4, ret          ; Jump to (ret + 4)
+```
+
+**Encoding Note:**
+In machine code: Offset (Immediate field), BaseReg (SrcReg field) (DestReg unused, set to noreg)
+
+### Comparison
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x14 | **CMP** | R | Reg1, Reg2 | Compare Reg1 with Reg2, set flags in STS |
+
+**Flag Setting:**
+- Equal: Set if Reg1 == Reg2
+- GreaterThan: Set if Reg1 > Reg2 (signed)
+- GreaterThanOrEqual: Set if Reg1 >= Reg2 (signed)
+- LessThan: Set if Reg1 < Reg2 (signed)
+- LessThanOrEqual: Set if Reg1 <= Reg2 (signed)
+- Zero: Set if (Reg1 - Reg2) == 0 (same as Equal)
+
+**Encoding Note:**
+DestReg and ShiftAmt fields unused (set to noreg and 0)
+
+### Arithmetic Instructions
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x15 | **INC** | R | Reg | Increment register by 1 |
+| 0x16 | **DEC** | R | Reg | Decrement register by 1 |
+| 0x19 | **ADD** | R | Src1, Src2, Dest | Dest = Src1 + Src2 |
+| 0x1A | **SUB** | R | Src1, Src2, Dest | Dest = Src1 - Src2 |
+| 0x25 | **IADD** | I | Src, Literal, Dest | Dest = Src + SignExtend(Literal) |
+| 0x26 | **ISUB** | I | Src, Literal, Dest | Dest = Src - SignExtend(Literal) |
+
+**Flag Effects:**
+- Zero flag set if result is zero
+- Other flags undefined after arithmetic (use CMP for comparisons)
+
+**Encoding Notes:**
+- INC/DEC: Reg in SrcReg1 field, DestReg set to noreg
+- IADD/ISUB: Immediate is signed 16-bit value, all three operands required
+
+### Bitwise Logical Operations
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x1B | **AND** | R | Src1, Src2, Dest | Dest = Src1 & Src2 (bitwise AND) |
+| 0x1C | **OR** | R | Src1, Src2, Dest | Dest = Src1 \| Src2 (bitwise OR) |
+| 0x1D | **NOT** | R | Src, Dest | Dest = ~Src (bitwise NOT) |
+| 0x1E | **XOR** | R | Src1, Src2, Dest | Dest = Src1 ^ Src2 (bitwise XOR) |
+| 0x1F | **NAND** | R | Src1, Src2, Dest | Dest = ~(Src1 & Src2) (bitwise NAND) |
+| 0x20 | **NOR** | R | Src1, Src2, Dest | Dest = ~(Src1 \| Src2) (bitwise NOR) |
+| 0x21 | **XNOR** | R | Src1, Src2, Dest | Dest = ~(Src1 ^ Src2) (bitwise XNOR) |
+
+**Flag Effects:**
+- Zero flag set if result is zero
+- Other flags undefined
+
+**Encoding Note:**
+NOT uses only Src (SrcReg1) and Dest (DestReg); SrcReg2 unused (set to noreg)
+
+### Shift Operations
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x17 | **SHL** | R | Reg, ShiftAmount | Shift Reg left by ShiftAmount bits<br/>Zero-fill from right |
+| 0x18 | **SHR** | R | Reg, ShiftAmount | Shift Reg right by ShiftAmount bits<br/>Zero-fill from left (logical shift) |
+
+**Shift Amount:**
+- **Literal shifts**: ShiftAmount is a 5-bit literal (0-31) in assembly
+  - Stored in ShiftAmt field of instruction
+  - SrcReg2 set to noreg
+- **Register shifts**: ShiftAmount is a register containing shift value
+  - Register specified in SrcReg2 field
+  - ShiftAmt field must be 0
+  - Only low 5 bits of register value used
+
+**Note:** Current assembler implementation may only support literal shifts. Check assembler documentation.
+
+**Flag Effects:**
+- Zero flag set if result is zero
+
+**Encoding Notes:**
+- Reg in both SrcReg1 and DestReg fields (shifted in place)
+- For literal shifts: ShiftAmt field contains shift count, SrcReg2 = noreg
+- For register shifts: SrcReg2 contains register, ShiftAmt must be 0
+
+### System and Control Instructions
+
+| Hex | Mnemonic | Type | Operands | Description |
+|-----|----------|------|----------|-------------|
+| 0x22 | **INT** | I | InterruptCode | Trigger interrupt with 8-bit code<br/>Saves return address to ret register<br/>Sets bpr to kernel stack |
+| 0x23 | **IRT** | R | - | Return from interrupt<br/>Restores execution context |
+| 0x24 | **HLT** | R | - | Halt processor execution<br/>Stops fetch-decode-execute cycle |
+
+**INT Behavior:**
+1. Save current PCX to ret register
+2. Switch bpr to kernel stack address
+3. Look up interrupt handler address in interrupt descriptor table (idr)
+4. Jump to handler at interrupt vector
+
+**IRT Behavior:**
+1. Restore previous execution context
+2. Return to address in ret register
+3. Restore user stack pointer
+
+**Encoding Notes:**
+- INT: InterruptCode in low 8 bits of Immediate field
+- IRT/HLT: All register fields set to noreg, ShiftAmt to 0
+
+### Meta Instructions (Assembler/Linker)
+
+These instructions are used by the assembler and linker but may not represent real CPU operations.
+
+| Hex | Mnemonic | Description |
+|-----|----------|-------------|
+| 0x27 | **SEGMENT** | Segment marker (implementation-specific) |
+| 0x3E | **DATA** | Raw data embedding |
+
+**Note:** The SEGMENT instruction opcode may vary between implementations (0x27 in assembler, 0x3F in some contexts). Consult your specific toolchain documentation.
+
+## Instruction Summary Table
+
+| Opcode | Mnemonic | Type | Category |
+|--------|----------|------|----------|
+| 0x00 | NOP | R | Control |
+| 0x01 | MOV | R | Data Movement |
+| 0x02 | MOVS | R | Data Movement |
+| 0x03 | LDB | I | Memory Load |
+| 0x04 | LDBS | I | Memory Load |
+| 0x05 | LDH | I | Memory Load |
+| 0x06 | LDHS | I | Memory Load |
+| 0x07 | LDW | I | Memory Load |
+| 0x08 | STB | I | Memory Store |
+| 0x09 | STH | I | Memory Store |
+| 0x0A | STW | I | Memory Store |
+| 0x0B | LLI | I | Immediate Load |
+| 0x0C | LUI | I | Immediate Load |
+| 0x0D | JMP | I | Jump |
+| 0x0E | JEQ | I | Branch |
+| 0x0F | JNE | I | Branch |
+| 0x10 | JGT | I | Branch |
+| 0x11 | JGE | I | Branch |
+| 0x12 | JLT | I | Branch |
+| 0x13 | JLE | I | Branch |
+| 0x14 | CMP | R | Comparison |
+| 0x15 | INC | R | Arithmetic |
+| 0x16 | DEC | R | Arithmetic |
+| 0x17 | SHL | R | Shift |
+| 0x18 | SHR | R | Shift |
+| 0x19 | ADD | R | Arithmetic |
+| 0x1A | SUB | R | Arithmetic |
+| 0x1B | AND | R | Logical |
+| 0x1C | OR | R | Logical |
+| 0x1D | NOT | R | Logical |
+| 0x1E | XOR | R | Logical |
+| 0x1F | NAND | R | Logical |
+| 0x20 | NOR | R | Logical |
+| 0x21 | XNOR | R | Logical |
+| 0x22 | INT | I | System |
+| 0x23 | IRT | R | System |
+| 0x24 | HLT | R | System |
+| 0x25 | IADD | I | Arithmetic |
+| 0x26 | ISUB | I | Arithmetic |
+| 0x27 | SEGMENT | - | Meta |
+| 0x3E | DATA | - | Meta |
+
+## Exception Conditions
+
+The following conditions trigger exceptions:
+
+| Exception | Trigger Condition |
+|-----------|------------------|
+| **Illegal Instruction** | - Invalid opcode<br/>- noreg used as source/destination<br/>- ShiftAmt non-zero for non-shift instruction<br/>- Register field violations |
+| **Protection Fault** | - Write to pcx register<br/>- Read/write idr or mmr in user mode<br/>- Read from noreg<br/>- Write to zero register (discarded, no fault) |
+| **Alignment Fault** | - LDH/LDHS/STH with odd address<br/>- LDW/STW with address not divisible by 4 |
+| **Memory Access Violation** | - Access to unmapped or protected memory<br/>- Stack overflow/underflow |
+
+## Calling Convention
+
+See the DSA Assembly Language Reference for the complete calling convention and ABI specification.
+
+## Notes on Design
+
+1. **Word Size:** All addresses and general computation is 32-bit
+2. **Endianness:** Little-endian for instructions and runtime data; assembler data directives may use big-endian
+3. **Stack Growth:** Stack grows **downward** (toward lower addresses) - PUSH decrements SPR
+4. **Alignment:** Natural alignment required for halfword and word accesses
+5. **Sign Extension:** All immediate values are sign-extended unless noted
+6. **Zero Register:** Provides constant zero, writes are legal but discarded
+7. **Reserved Encodings:** Opcodes 0x27-0x3D and 0x3F reserved or implementation-specific

docs/DSA_Project_Roadmap.md

@@ -0,0 +1,875 @@
+# DSA Project Roadmap & Task Breakdown
+
+> **Damn Simple Architecture** — Full ecosystem development plan including emulator, assembler, compiler, debugger, and tooling infrastructure.
+
+---
+
+## Table of Contents
+
+1. [Phase 1: Foundation & Core Infrastructure](#phase-1-foundation--core-infrastructure)
+   - [1.1 Binary Format & Linking System](#11-binary-format--linking-system)
+   - [1.2 Assembler Rewrite](#12-assembler-rewrite)
+   - [1.3 Documentation Updates](#13-documentation-updates)
+2. [Phase 2: Compiler Development](#phase-2-compiler-development)
+   - [2.1 Language Design & Implementation](#21-language-design--implementation)
+   - [2.2 Standard Library](#22-standard-library)
+3. [Phase 3: Build System & Package Management](#phase-3-build-system--package-management)
+   - [3.1 Build System](#31-build-system)
+   - [3.2 Package Management System](#32-package-management-system)
+4. [Phase 4: Debugger & Development Tools](#phase-4-debugger--development-tools)
+   - [4.1 Debug Symbol System](#41-debug-symbol-system)
+   - [4.2 Debugger Implementation](#42-debugger-implementation)
+   - [4.3 Enhanced Editor Integration](#43-enhanced-editor-integration)
+5. [Phase 5: Integration & Polish](#phase-5-integration--polish)
+6. [Phase 6: Future Enhancements (NTH)](#phase-6-future-enhancements-nth)
+7. [Summary Timeline](#summary-timeline)
+8. [Critical Path](#critical-path)
+9. [Recommended Work Order](#recommended-work-order)
+
+---
+
+## Phase 1: Foundation & Core Infrastructure
+
+**Estimated Duration: 3–4 weeks**
+
+---
+
+### 1.1 Binary Format & Linking System
+
+> **Priority: CRITICAL** — Everything depends on this.  
+> **Total Estimate: 1.5 weeks**
+
+---
+
+#### 1.1.1 Design New Binary Format Specification
+
+**Estimate: 2 days**  
+**Dependencies:** None  
+**Deliverable:** `docs/binary-format-spec.md`
+
+- [ ] Research existing object file formats (ELF, COFF, Mach-O) for inspiration
+- [ ] Design `.dsb` object file format specification
+  - [ ] Symbol table structure
+  - [ ] Relocation table format
+  - [ ] Section definitions (code, data, rodata, bss)
+  - [ ] Debug information structure
+  - [ ] Metadata headers
+- [ ] Design `.dse` executable format specification
+  - [ ] Entry point definition
+  - [ ] Memory layout requirements
+  - [ ] Linking metadata
+- [ ] Document format specifications in markdown
+- [ ] Create format version strategy for future compatibility
+
+---
+
+#### 1.1.2 Implement DSB Object File Writer
+
+**Estimate: 3 days**  
+**Dependencies:** 1.1.1  
+**Deliverable:** `dsa-binary-format` crate v0.1.0
+
+- [ ] Create new crate: `dsa-binary-format`
+- [ ] Implement object file structures
+  - [ ] Header structure
+  - [ ] Symbol table builder
+  - [ ] Section manager
+  - [ ] Relocation entry creator
+- [ ] Write serialization logic
+- [ ] Add validation and error handling
+- [ ] Write unit tests for each structure
+- [ ] Integration tests for complete object files
+
+---
+
+#### 1.1.3 Build Linker Program
+
+**Estimate: 4 days**  
+**Dependencies:** 1.1.2  
+**Deliverable:** `dsa-link` executable
+
+- [ ] Create new crate: `dsa-linker`
+- [ ] Implement symbol resolution
+  - [ ] Global symbol table
+  - [ ] Symbol conflict detection
+  - [ ] Weak symbol handling
+- [ ] Implement relocation processing
+  - [ ] Address calculation
+  - [ ] Patch generation
+  - [ ] Cross-section references
+- [ ] Build executable generator
+  - [ ] Combine sections
+  - [ ] Generate final memory layout
+  - [ ] Write `.dse` output
+- [ ] Add linker script support (basic)
+- [ ] Comprehensive error messages
+- [ ] Test suite with complex linking scenarios
+
+---
+
+### 1.2 Assembler Rewrite
+
+> **Priority: HIGH** — Required for all compiled code.  
+> **Total Estimate: 1.5 weeks**
+
+---
+
+#### 1.2.1 Assembler Architecture Design
+
+**Estimate: 1 day**  
+**Dependencies:** 1.1.1  
+**Deliverable:** `docs/assembler-architecture.md`
+
+- [ ] Design multi-pass architecture
+  - [ ] Pass 1: Symbol collection
+  - [ ] Pass 2: Macro expansion
+  - [ ] Pass 3: Code generation
+  - [ ] Pass 4: Relocation generation
+- [ ] Plan error handling strategy
+- [ ] Design threading model for parallel file processing
+- [ ] Define module/import resolution system
+- [ ] Plan integration points with DSC compiler
+
+---
+
+#### 1.2.2 Implement Core Assembler
+
+**Estimate: 5 days**  
+**Dependencies:** 1.1.2, 1.2.1  
+**Deliverable:** `dsa-asm` executable v2.0.0
+
+- [ ] Create new crate: `dsa-assembler-ng` (next-gen)
+- [ ] Implement lexer with better error recovery
+- [ ] Build parser with detailed error messages
+  - [ ] Instruction parsing
+  - [ ] Directive handling
+  - [ ] Macro system
+  - [ ] Include resolution
+- [ ] Symbol table management
+- [ ] Code generator outputting to DSB format
+- [ ] Multi-threading for file parsing
+- [ ] Comprehensive test suite
+- [ ] Error message testing
+
+---
+
+#### 1.2.3 Import System & DSC Integration
+
+**Estimate: 2 days**  
+**Dependencies:** 1.2.2, 2.1.2  
+**Deliverable:** Working import system
+
+- [ ] Design import protocol between DSC and assembler
+- [ ] Implement symbol table merging
+- [ ] Handle pre-compiled object imports
+- [ ] Test DSC → Assembly → Object pipeline
+- [ ] Document integration process
+
+---
+
+### 1.3 Documentation Updates
+
+> **Priority: MEDIUM** — Can be done alongside development.  
+> **Total Estimate: 3 days (distributed)**
+
+---
+
+#### 1.3.1 Update Assembly Documentation
+
+**Estimate: 1 day**  
+**Dependencies:** 1.2.2  
+**Deliverable:** Updated `docs/dsa-assembly-reference.md`
+
+- [ ] Review all instruction documentation
+- [ ] Document new pseudo-instructions
+- [ ] Update calling convention docs
+- [ ] Add examples for new features
+- [ ] Document assembler directives
+- [ ] Macro system documentation
+
+---
+
+#### 1.3.2 Architecture Documentation
+
+**Estimate: 1 day**  
+**Dependencies:** None (can start anytime)  
+**Deliverable:** `docs/dsa-architecture.md`
+
+- [ ] Document ISA specification
+- [ ] Memory model documentation
+- [ ] Interrupt handling
+- [ ] Hardware peripheral specs
+- [ ] Timing/performance characteristics
+
+---
+
+#### 1.3.3 Build Tools Documentation
+
+**Estimate: 1 day**  
+**Dependencies:** 1.2.2, 1.1.3, 3.1.2  
+**Deliverable:** `docs/build-tools-guide.md`
+
+- [ ] Assembler usage guide
+- [ ] Linker usage guide
+- [ ] Build system guide
+- [ ] Tutorial: Building a simple program
+- [ ] Tutorial: Multi-file projects
+
+---
+
+## Phase 2: Compiler Development
+
+**Estimated Duration: 3–4 weeks**
+
+---
+
+### 2.1 Language Design & Implementation
+
+> **Priority: HIGH** — Core functionality.  
+> **Total Estimate: 2.5 weeks**
+
+---
+
+#### 2.1.1 Language Syntax Design
+
+**Estimate: 2 days**  
+**Dependencies:** None  
+**Deliverable:** `docs/language-spec.md`
+
+- [x] Define syntax goals (simplicity, systems programming)
+- [ ] Design type system
+  - [x] Primitive types
+  - [x] Pointers/references
+  - [ ] Structs
+  - [ ] Arrays
+  - [x] Function types
+- [x] Control flow syntax
+- [x] Function declaration syntax
+- [x] Module/import system
+- [x] Operator precedence
+- [ ] Write EBNF grammar
+- [x] Create example programs
+
+---
+
+#### 2.1.2 Lexer & Parser Implementation
+
+**Estimate: 4 days**  
+**Dependencies:** 2.1.1  
+**Deliverable:** Parser in `dsc-compiler` crate
+
+- [x] Adapt existing C lexer to new syntax
+- [ ] Implement new parser for designed syntax
+  - [ ] Array syntax
+  - [ ] Struct syntax
+  - [x] Pointer syntax
+  - [x] Namespaced call syntax
+- [x] AST node definitions
+- [ ] Error recovery mechanisms
+- [ ] Comprehensive parser tests
+- [ ] Syntax error message quality testing
+- [x] Implement C frontend by moving lexer/parser from `c_compiler` to the new `compiler` project structure
+- [ ] Evaluate possible memory management strategies (e.g., keep all variables on the stack vs spill only when calling functions)
+
+---
+
+#### 2.1.3 Code Generation Improvements
+
+**Estimate: 5 days**  
+**Dependencies:** 2.1.2, 1.2.2  
+**Deliverable:** Working code generator
+
+- [x] Review and fix existing codegen issues
+- [ ] Implement missing language features
+  - [ ] Structs
+  - [ ] Arrays
+  - [x] Pointers/memory operations
+  - [ ] For loops
+  - [ ] Switch statements
+  - [ ] Break/continue
+- [ ] Optimize register allocation further
+- [x] Implement proper function calling conventions
+- [ ] Add constant folding optimization
+- [x] Dead code elimination
+- [ ] Test each feature thoroughly
+
+---
+
+#### 2.1.4 Type Checking & Semantic Analysis
+
+**Estimate: 3 days**  
+**Dependencies:** 2.1.2  
+**Deliverable:** Type checker integrated in compiler
+
+- [ ] Implement type checker
+- [ ] Symbol table for scoping
+- [ ] Type inference where applicable
+- [ ] Const checking
+- [ ] Definite assignment analysis
+- [ ] Comprehensive semantic error messages
+- [ ] Test suite for type errors
+
+---
+
+### 2.2 Standard Library
+
+> **Priority: MEDIUM** — Needed for useful programs.  
+> **Total Estimate: 1 week**
+
+---
+
+#### 2.2.1 Core Runtime Library (in Assembly)
+
+**Estimate: 3 days**  
+**Dependencies:** 1.2.2  
+**Deliverable:** `lib/runtime/` directory
+
+- [ ] Memory allocation (malloc/free)
+- [ ] String operations
+- [ ] Math functions
+  - [x] Multiply
+  - [ ] Divide (fix as very slow and broken)
+- [ ] I/O functions (improved print, read)
+  - [x] Print number
+  - [x] Print hex value
+  - [x] Print word
+  - [x] Print byte
+  - [x] Print from string ptr
+  - [x] Print whitespace and newline
+  - [x] Reset display
+  - [x] Reset cursor
+- [ ] System call interface
+- [ ] Tests for each function
+
+---
+
+#### 2.2.2 Standard Library (in DSC)
+
+**Estimate: 2 days**  
+**Dependencies:** 2.1.3, 2.2.1  
+**Deliverable:** `lib/std/` directory
+
+- [ ] String module
+- [ ] Collections (array utilities, maybe simple list)
+- [ ] File I/O module
+- [ ] Math utilities
+- [ ] Tests and examples
+
+---
+
+## Phase 3: Build System & Package Management
+
+**Estimated Duration: 2–3 weeks**
+
+---
+
+### 3.1 Build System
+
+> **Priority: HIGH** — Required for complex projects.  
+> **Total Estimate: 1.5 weeks**
+
+---
+
+#### 3.1.1 Build System Design
+
+**Estimate: 1 day**  
+**Dependencies:** None  
+**Deliverable:** `docs/build-system-design.md`
+
+- [x] Define project structure conventions
+- [ ] Design build manifest format (`dsa-project.toml` or similar)
+- [ ] Dependency resolution strategy
+- [ ] Build cache design
+- [ ] Incremental build strategy
+- [ ] Multi-target support
+
+---
+
+#### 3.1.2 Build Tool Implementation
+
+**Estimate: 5 days**  
+**Dependencies:** 3.1.1, 1.2.2, 1.1.3, 2.1.3  
+**Deliverable:** `dsa-build` executable
+
+- [x] Create crate: `dsa-build`
+- [ ] Manifest parser
+- [ ] Dependency graph builder
+- [ ] Task orchestrator
+  - [x] Compilation tasks
+  - [x] Assembly tasks
+  - [ ] Linking tasks
+- [ ] Build cache implementation
+- [ ] Parallel build support
+- [ ] Clean, rebuild commands
+- [ ] Watch mode for development
+- [ ] Comprehensive tests
+
+---
+
+#### 3.1.3 Project Management Commands
+
+**Estimate: 2 days**  
+**Dependencies:** 3.1.2  
+**Deliverable:** Enhanced `dsa-build` with project management
+
+- [x] `dsa new <project>` — Create new project
+- [x] `dsa init` — Initialize in existing directory
+- [ ] `dsa add <dependency>` — Add dependency
+- [ ] Binary vs library project types
+- [x] Template system for project scaffolding
+- [ ] Documentation for each command
+
+---
+
+### 3.2 Package Management System
+
+> **Priority: MEDIUM** — Enables code sharing.  
+> **Total Estimate: 1.5 weeks**
+
+---
+
+#### 3.2.1 Package Registry Design
+
+**Estimate: 2 days**  
+**Dependencies:** 3.1.1  
+**Deliverable:** `docs/package-registry-design.md`
+
+- [ ] Decide: Git monorepo vs custom hosting
+- [ ] Design package naming conventions
+- [ ] Version resolution strategy (semver)
+- [ ] Package manifest format
+- [ ] Security considerations
+- [ ] Package storage format (source/binary/both)
+- [ ] API design for registry server
+
+---
+
+#### 3.2.2 Local Package Manager Tool
+
+**Estimate: 4 days**  
+**Dependencies:** 3.2.1, 3.1.2  
+**Deliverable:** `dsa-pkg` tool integrated with `dsa-build`
+
+- [ ] Create crate: `dsa-pkg`
+- [ ] Package index synchronization
+- [ ] Dependency resolver
+- [ ] Package download/cache system
+- [ ] Integration with build system
+- [ ] Commands:
+  - [ ] `dsa install <package>`
+  - [ ] `dsa publish`
+  - [ ] `dsa search <query>`
+  - [ ] `dsa update`
+- [ ] Lock file generation
+- [ ] Test with mock registry
+
+---
+
+#### 3.2.3 Package Registry Implementation
+
+**Estimate: 3 days**  
+**Dependencies:** 3.2.1  
+**Deliverable:** Package registry (URL or repo)
+
+- [ ] If **Git monorepo** approach:
+  - [ ] Set up repository structure
+  - [ ] CI/CD for validation
+  - [ ] Submission process
+  - [ ] Package browser website
+- [ ] If **custom hosting**:
+  - [ ] Simple web server (Rust + Axum/Actix)
+  - [ ] Package upload API
+  - [ ] Package search API
+  - [ ] Basic web UI
+  - [ ] Database for metadata
+- [ ] Documentation for publishing
+
+---
+
+## Phase 4: Debugger & Development Tools
+
+**Estimated Duration: 3–4 weeks**
+
+---
+
+### 4.1 Debug Symbol System
+
+> **Priority: HIGH** — Foundation for debugging.  
+> **Total Estimate: 1 week**
+
+---
+
+#### 4.1.1 Debug Symbol Format Design
+
+**Estimate: 1 day**  
+**Dependencies:** 1.1.1  
+**Deliverable:** `docs/debug-symbol-format.md`
+
+- [ ] Design symbol table format
+  - [ ] Function addresses → names
+  - [ ] Line number → address mapping
+  - [ ] Variable location information
+  - [ ] Type information
+- [ ] Define symbol table file format
+- [ ] Plan for embedding in DSE/DSB files
+
+---
+
+#### 4.1.2 Symbol Generation in Tools
+
+**Estimate: 3 days**  
+**Dependencies:** 4.1.1, 1.2.2, 2.1.3  
+**Deliverable:** Debug symbols in build output
+
+- [ ] Modify assembler to emit debug symbols
+- [ ] Modify compiler to emit debug symbols
+  - [ ] Source file/line tracking
+  - [ ] Variable scope tracking
+- [ ] Linker merges debug symbols
+- [ ] Test symbol generation pipeline
+
+---
+
+#### 4.1.3 Symbol Table Loader in Emulator
+
+### Pre-Debugger Editor Integration Tasks
+
+- **Integrate compiler into editor**  
+  - Add a build command that invokes the full compiler pipeline (lexer → parser → codegen).  
+  - Show compilation output and errors in the console panel.  
+
+- **DSC language support**  
+  - Enable syntax highlighting and auto‑completion for DSC files within the editor.  
+  - Provide a dedicated “Build DSC” command that uses the integrated compiler.  
+
+- **Editor diagnostics**  
+  - Wire compiler error messages to the editor’s gutter so users can click to jump to source lines.  
+
+
+**Estimate: 2 days**  
+**Dependencies:** 4.1.2  
+**Deliverable:** Symbol loading in emulator crate
+
+- [ ] Implement symbol table parser
+- [ ] Build address → symbol lookup (HashMap)
+- [ ] Build symbol → address lookup
+- [ ] Memory efficient storage
+- [ ] Tests for symbol resolution
+
+---
+
+### 4.2 Debugger Implementation
+
+> **Priority: HIGH** — Major productivity boost.  
+> **Total Estimate: 2 weeks**
+
+---
+
+#### 4.2.1 Core Debugger Features
+
+**Estimate: 5 days**  
+**Dependencies:** 4.1.3  
+**Deliverable:** Debugger backend
+
+- [ ] Execution control
+  - [ ] Step instruction
+  - [ ] Step over function calls
+  - [ ] Continue to breakpoint
+  - [ ] Run to cursor
+- [ ] Breakpoint system
+  - [ ] Address breakpoints
+  - [ ] Conditional breakpoints
+  - [ ] Watchpoints (memory access)
+- [ ] Register inspection
+- [ ] Memory inspection
+- [ ] Stack trace generation
+- [ ] Test debugger commands
+
+---
+
+#### 4.2.2 Disassembler with Symbol Resolution
+
+**Estimate: 3 days**  
+**Dependencies:** 4.1.3  
+**Deliverable:** Enhanced disassembler
+
+- [ ] Instruction decoder
+- [ ] Format with labels instead of addresses
+- [ ] Show function names at call sites
+- [ ] Inline comments with variable names
+- [ ] Color coding for instruction types
+- [ ] Tests for disassembly output
+
+---
+
+#### 4.2.3 Pseudo-Instruction Decompiler
+
+> ⚠️ **COMPLEX TASK** — Separate pass to decompile assembly into readable pseudo-instructions.
+
+**Estimate: 4 days**  
+**Dependencies:** 4.2.2  
+**Deliverable:** Pseudo-instruction view mode
+
+- [ ] Pattern recognition for common sequences
+  - [ ] Function prologue/epilogue
+  - [ ] Multiplication using shifts/adds
+  - [ ] Division
+  - [ ] Conditional moves
+- [ ] Control flow reconstruction
+  - [ ] If/else detection
+  - [ ] Loop detection
+  - [ ] Switch statement detection
+- [ ] Expression reconstruction
+- [ ] Format as higher-level pseudo-code
+- [ ] Extensive pattern testing
+
+---
+
+#### 4.2.4 Execution History Tracking
+
+**Estimate: 2 days**  
+**Dependencies:** 4.2.1  
+**Deliverable:** Execution trace feature
+
+- [ ] Circular buffer for instruction history
+- [ ] Register state snapshots over time
+- [ ] Configurable history depth
+- [ ] Efficient memory usage
+- [ ] Playback/reverse debugging (basic)
+- [ ] Export trace to file
+
+---
+
+### 4.3 Enhanced Editor Integration
+
+> **Priority: MEDIUM** — UX improvement.  
+> **Total Estimate: 1 week**
+
+---
+
+#### 4.3.1 Tiling Window System
+
+**Estimate: 2 days**  
+**Dependencies:** None (UI work)  
+**Deliverable:** Panel system in emulator
+
+- [ ] Research Rust tiling libraries (`egui_tiles`, or custom)
+- [ ] Design panel layout system
+  - [ ] Code editor panel
+  - [ ] Disassembly panel
+  - [ ] Register panel
+  - [ ] Memory panel
+  - [ ] Console panel
+- [ ] Implement drag-and-drop panel management
+- [ ] Save/load layouts
+
+---
+
+#### 4.3.2 Assembly Editor Improvements
+
+**Estimate: 2 days**  
+**Dependencies:** 4.3.1  
+**Deliverable:** Enhanced assembly editor
+
+- [ ] Syntax highlighting for DSA assembly
+- [ ] Auto-completion for instructions
+- [ ] Label/symbol auto-completion
+- [ ] Error highlighting
+- [ ] Inline documentation tooltips
+- [ ] Jump-to-definition for labels
+
+---
+
+#### 4.3.3 High-Level Language Editor
+
+**Estimate: 2 days**  
+**Dependencies:** 4.3.1, 2.1.4  
+**Deliverable:** DSC language editor
+
+- [ ] Syntax highlighting for DSC language
+- [ ] Basic auto-completion
+- [ ] Bracket matching
+- [ ] Error highlighting from compiler
+- [ ] Go-to-definition (using debug symbols)
+- [ ] Inline type hints
+
+---
+
+#### 4.3.4 Integrate Build Tools and Compiler into Editor
+
+Estimate: 1 day
+Dependencies: 4.3.1, 3.1.2, 2.1.2
+Deliverable: Integrated build experience with compiler support
+
+- [ ] Build button/command in UI that invokes the full compiler pipeline
+- [ ] Show build output and compilation errors in console panel
+- [ ] Error navigation (click to jump to source)
+- [ ] Hot reload on successful build
+- [ ] Build status indicator
+- [ ] Hook DSC language support into editor for syntax highlighting and auto‑completion
+- [ ] Provide dedicated DSC build command that uses the new compiler integration
+
+---
+
+## Phase 5: Integration & Polish
+
+**Estimated Duration: 1–2 weeks**
+
+---
+
+### 5.1 Tool Integration
+
+> **Priority: HIGH** — Everything works together.  
+> **Total Estimate: 1 week**
+
+---
+
+#### 5.1.1 Unified Toolchain
+
+**Estimate: 3 days**  
+**Dependencies:** All previous phases  
+**Deliverable:** `dsa` unified command-line tool
+
+- [ ] Create meta-crate: `dsa-tools`
+- [ ] Unified CLI with subcommands
+  - [ ] `dsa build`
+  - [ ] `dsa run`
+  - [ ] `dsa debug`
+  - [ ] `dsa test`
+  - [ ] `dsa pkg`
+- [ ] Shared configuration system
+- [ ] Tool interop testing
+- [ ] Documentation for workflow
+
+---
+
+#### 5.1.2 Emulator Integration
+
+**Estimate: 2 days**  
+**Dependencies:** 5.1.1, 4.3.4  
+**Deliverable:** Fully integrated development environment
+
+- [ ] Add build tools as emulator dependencies
+- [ ] In-editor build triggered from emulator
+- [ ] Debugger uses build output directly
+- [ ] Source-level debugging with line mapping
+- [ ] Test full edit → build → debug cycle
+
+---
+
+#### 5.1.3 Documentation & Tutorials
+
+**Estimate: 2 days**  
+**Dependencies:** 5.1.2  
+**Deliverable:** Complete documentation suite
+
+- [ ] Getting started guide
+- [ ] Full tutorial: Building a simple game
+- [ ] Debugger usage guide
+- [ ] Best practices document
+- [ ] Troubleshooting guide
+
+---
+
+## Phase 6: Future Enhancements (NTH)
+
+> **Priority: LOW** — Nice to have, long-term goal.  
+> **Estimated Duration: 4+ weeks**
+
+---
+
+### 6.1 Command-Line Emulator
+
+> ⚠️ **COMPLEX LONG-TERM GOAL** — Requires significant design and UX consideration.
+
+---
+
+#### 6.1.1 Design Phase
+
+**Estimate: 1 week**  
+**Dependencies:** None  
+**Deliverable:** `docs/cli-emulator-design.md`
+
+- [ ] UX research for terminal-based debuggers
+- [ ] Design TUI layout (using `ratatui` or similar)
+- [ ] Command syntax design
+- [ ] Scripting support design
+- [ ] Accessibility considerations
+
+---
+
+#### 6.1.2 Implementation
+
+**Estimate: 3+ weeks**  
+**Dependencies:** 6.1.1, Phase 4 complete  
+**Deliverable:** `dsa-emu-cli` executable
+
+- [ ] TUI framework setup
+- [ ] Core emulator integration
+- [ ] Command parser
+- [ ] Panel rendering (code, registers, memory, etc.)
+- [ ] Keyboard shortcuts
+- [ ] Mouse support
+- [ ] Configuration system
+- [ ] Extensive usability testing
+
+---
+
+## Summary Timeline
+
+| Phase                         | Duration  | Key Dependencies    |
+| ----------------------------- | --------- | ------------------- |
+| Phase 1: Foundation           | 3–4 weeks | None                |
+| Phase 2: Compiler             | 3–4 weeks | Phase 1 complete    |
+| Phase 3: Build System         | 2–3 weeks | Phases 1–2 complete |
+| Phase 4: Debugger             | 3–4 weeks | Phases 1–3 complete |
+| Phase 5: Integration          | 1–2 weeks | Phases 1–4 complete |
+| Phase 6: CLI Emulator _(NTH)_ | 4+ weeks  | Phase 4 complete    |
+
+**Total Estimated Time: 12–17 weeks (3–4 months) for Phases 1–5**
+
+---
+
+## Critical Path
+
+The following tasks are on the critical path and will block other work if delayed:
+
+```
+1.1.1 Binary format design
+  └── 1.1.2 Object file writer
+        └── 1.1.3 Linker
+              └── 1.2.2 Assembler rewrite
+                    └── 2.1.3 Compiler codegen
+                          └── 3.1.2 Build system
+                                └── 4.1.2 Debug symbols
+                                      └── 4.2.1 Debugger
+```
+
+---
+
+## Recommended Work Order
+
+| Weeks | Focus                                 | Tasks                                             |
+| ----- | ------------------------------------- | ------------------------------------------------- |
+| 1–2   | Binary Format & Linker                | 1.1.1 → 1.1.2 → 1.1.3                             |
+| 3–4   | Assembler Rewrite                     | 1.2.1 → 1.2.2                                     |
+| 5–6   | Compiler Syntax & Parser              | 2.1.1 → 2.1.2 _(start 1.3 docs in parallel)_      |
+| 7–9   | Compiler Codegen & Types              | 2.1.3 → 2.1.4 _(start 2.2.1 runtime in parallel)_ |
+| 10–11 | Build System                          | 3.1.1 → 3.1.2 → 3.1.3                             |
+| 12–13 | Package Management _(if desired now)_ | 3.2.1 → 3.2.2 → 3.2.3                             |
+| 14–15 | Debug Symbols                         | 4.1.1 → 4.1.2 → 4.1.3                             |
+| 16–18 | Core Debugger                         | 4.2.1 → 4.2.2 → 4.2.4                             |
+| 19–20 | Editor Enhancements                   | 4.3.1 → 4.3.2 → 4.3.3 → 4.3.4                     |
+| 21–22 | Integration & Polish                  | 5.1.1 → 5.1.2 → 5.1.3                             |
+
+---
+
+## Notes
+
+- Time estimates assume ~6–8 productive hours per day.
+- Add **20–30% buffer** for unexpected issues.
+- Testing time is included in each estimate.
+- Documentation is distributed throughout rather than batched at the end.
+- Package management (3.2) can be deferred if time-constrained.
+- Pseudo-instruction decompiler (4.2.3) can be a stretch goal.
+- CLI emulator (Phase 6) is explicitly a "nice to have" and should not block other work.

docs/IMPLEMENTATION_DISCREPANCIES.md

@@ -0,0 +1,638 @@
+# DSA Implementation vs Documentation Discrepancies
+
+## Critical Discrepancies
+
+### 1. **Stack Growth Direction** ❌ CRITICAL
+
+**Documentation states:** Stack grows upward (toward higher addresses)
+
+**Implementation shows (expand.rs:44-51):**
+```rust
+fn expand_push(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    // ...
+    nodes.extend(vec![
+        node!(label, Opcode::SubI, spr, 4, spr),  // spr = spr - 4
+        node!(None, Opcode::Stw, reg, spr, 0),
+    ]);
+```
+
+**Implementation shows (expand.rs:130-137):**
+```rust
+fn expand_pop(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    // ...
+    nodes.extend(vec![
+        node!(label, Opcode::Ldw, spr, reg, 0),
+        node!(None, Opcode::AddI, spr, 4, spr),  // spr = spr + 4
+    ]);
+```
+
+**Reality:** Stack grows **DOWNWARD** (toward lower addresses)
+- PUSH: Decrements SPR by 4, then stores
+- POP: Loads, then increments SPR by 4
+
+**Impact:** All documentation examples and calling convention diagrams are backwards!
+
+---
+
+### 2. **CALL Pseudo-instruction Expansion** ❌ CRITICAL
+
+**Documentation states (DSA_Assembly_Reference.md):**
+```asm
+; call print::print expands to:
+lwi print::print, ret       ; Load function address into ret
+jmp 0, ret                  ; Jump to function (saves return in pcx)
+```
+
+**Implementation shows (expand.rs:109-123):**
+```rust
+fn expand_call(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    nodes.extend(vec![
+        node!(label, Opcode::SubI, spr, 4, spr),    // Decrement stack pointer
+        node!(None, Opcode::Stw, pcx, spr, 0),       // Store PCX (return addr) on stack
+        node!(None, Opcode::Jmp, addr, zero),        // Jump to function
+    ]);
+```
+
+**Reality:** CALL expansion is:
+1. Decrement SPR by 4
+2. Store PCX (return address) to stack
+3. Jump to function address
+
+**Impact:** Return address is stored on the STACK, not in RET register!
+
+---
+
+### 3. **RETURN Pseudo-instruction Expansion** ❌ CRITICAL
+
+**Documentation states:**
+```asm
+; return expands to:
+jmp 0, ret                  ; Jump to address in ret register
+```
+
+**Implementation shows (expand.rs:125-135):**
+```rust
+fn expand_return(current: &Node, nodes: &mut Vec<Node>) {
+    nodes.extend(vec![
+        node!(label, Opcode::Ldw, spr, ret, 0),     // Load return addr from stack
+        node!(None, Opcode::AddI, spr, 4, spr),     // Increment stack pointer
+        node!(None, Opcode::Jmp, 4, ret),           // Jump to (ret + 4)
+    ]);
+}
+```
+
+**Reality:** RETURN expansion is:
+1. Load return address from stack into RET register
+2. Increment SPR by 4
+3. Jump to (RET + 4)
+
+**Why +4?** The stored PCX points to the instruction AFTER the call's jump, so we need to add 4 to skip past the stored PCX instruction itself... or this might be a bug in the implementation.
+
+**Impact:** Return mechanism is completely different from documentation!
+
+---
+
+### 4. **Calling Convention - Stack Frame Layout** ❌ CRITICAL
+
+**Documentation states:**
+```
+Higher Addresses
+├─────────────┤
+│   Arg N     │  ← spr + (8 + 4*(N-1))
+│   ...       │
+│   Arg 2     │  ← spr + 16
+│   Arg 1     │  ← spr + 12
+│   Arg 0     │  ← spr + 8
+├─────────────┤
+│   Ret Addr  │  ← spr + 4
+├─────────────┤
+│   Old BPR   │  ← spr + 0
+├─────────────┤  ← bpr, spr
+│   Locals    │
+Lower Addresses
+```
+
+**Reality based on implementation:**
+Since stack grows DOWN:
+```
+Lower Addresses
+├─────────────┤  ← Current SPR/BPR
+│   Old BPR   │  ← spr + 0 (immediately above SPR)
+├─────────────┤
+│   Ret Addr  │  ← spr + 4 (pushed by CALL)
+├─────────────┤
+│   Arg 0     │  ← spr + 8
+│   Arg 1     │  ← spr + 12
+│   Arg 2     │  ← spr + 16
+│   ...       │
+│   Arg N     │  ← spr + (8 + 4*(N-1))
+├─────────────┤
+Higher Addresses
+```
+
+**The diagram needs to be flipped!** The offsets are correct, but the direction is wrong.
+
+---
+
+### 5. **Label-Based Load/Store Scratch Register** ⚠️ IMPORTANT
+
+**Documentation states:** Uses `rgf` as scratch register
+
+**Implementation confirms (expand.rs:138-153):**
+```rust
+fn expand_ldx(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    // For ldb label, reg:
+    nodes.extend(vec![
+        node!(current.label(), Opcode::Lli, name, reg),
+        node!(None, Opcode::Lui, name, reg),
+        node!(None, opcode, reg, reg, offset),
+    ]);
+```
+
+**Wait! This is WRONG in the implementation!**
+
+The load expansion uses the DESTINATION register as scratch:
+```asm
+ldb buffer, rg2  expands to:
+    lli buffer, rg2    ; Uses rg2 as destination
+    lui buffer, rg2    ; Uses rg2 as destination
+    ldb rg2, rg2, 0    ; Uses rg2 as base
+```
+
+**Documentation says it should use rgf:**
+```asm
+ldb buffer, rg2 expands to:
+    lli buffer, rgf    ; Uses rgf as scratch
+    lui buffer, rgf    ; Uses rgf as scratch
+    ldb rgf, rg2, 0    ; Load from rgf into rg2
+```
+
+**For stores (expand.rs:155-176):**
+```rust
+fn expand_stx(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    // For stb reg, label:
+    let temp = Token::Register(Register::Acc);  // Uses ACC, not RGF!
+    
+    nodes.extend(vec![
+        node!(current.label(), Opcode::Lli, dest, temp),
+        node!(None, Opcode::Lui, dest, temp),
+        node!(None, opcode, base, temp, offset),
+    ]);
+```
+
+**Reality:**
+- Load pseudo-instructions use the DESTINATION register as scratch
+- Store pseudo-instructions use the ACC register as scratch, NOT rgf
+
+**Impact:** Documentation is incorrect about which registers are used!
+
+---
+
+### 6. **LWI Pseudo-instruction** ✅ CORRECT
+
+**Documentation and implementation agree:**
+```rust
+fn expand_lwi(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    nodes.extend(vec![
+        node!(current.label(), Opcode::Lli, val, reg),
+        node!(None, Opcode::Lui, val, reg),
+    ]);
+```
+
+This matches the documented expansion.
+
+---
+
+### 7. **PUSHA/POPA Pseudo-instructions** 📝 UNDOCUMENTED
+
+**These exist in implementation but are NOT in documentation!**
+
+**expand.rs:53-76:**
+```rust
+fn expand_pusha(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    let count = expect_token!(arg0, Immediate)?;
+    let spr = Token::Register(Register::Spr);
+    let registers: Vec<Register> = Register::general();
+
+    nodes.push(node!(label, Opcode::SubI, spr, Token::Immediate(count * 4), spr));
+    
+    nodes.extend((0..count).rev().map(|i| {
+        node!(None, Opcode::Stw,
+            Token::Register(registers[i as usize]),
+            spr,
+            Token::Immediate(i * 4)
+        )
+    }));
+```
+
+**expand.rs:78-101:**
+```rust
+fn expand_popa(current: &Node, nodes: &mut Vec<Node>) -> Result<(), AssembleError> {
+    let count = expect_token!(arg0, Immediate)?;
+    
+    nodes.extend((0..count).rev().map(|i| {
+        node!(
+            { if i == 0 { label.clone() } else { None } },
+            Opcode::Ldw,
+            spr,
+            Token::Register(registers[i as usize]),
+            Token::Immediate(i * 4)
+        )
+    }));
+    
+    nodes.push(node!(None, Opcode::AddI, spr, Token::Immediate(count * 4), spr));
+```
+
+**What they do:**
+- `pusha N` - Push first N general-purpose registers (rg0-rgN) to stack
+- `popa N` - Pop first N general-purpose registers from stack
+
+**Missing from documentation entirely!**
+
+---
+
+### 8. **Register Index Encoding** ⚠️ IMPORTANT
+
+**Documentation states:** System registers like MAR, MDR, STS, CIR, PCX are "internal" and not accessible
+
+**Implementation shows (instructions.rs:148-153):**
+```rust
+0x18 => Self::Mar,
+0x19 => Self::Mdr,
+0x1A => Self::Sts,
+0x1B => Self::Cir,
+0x1C => Self::Pcx,
+```
+
+**Reality:** These registers ARE encoded in the instruction format at indices 0x18-0x1C!
+
+**However, instructions.rs:186 shows:**
+```rust
+"null" => Ok(Self::NoReg),  // Can parse "null" as NoReg
+```
+
+**Documentation never mentions "null" as an alternative name for noreg!**
+
+---
+
+### 9. **LUI Immediate Value Handling** ⚠️ IMPORTANT
+
+**Documentation states:**
+```
+lui immediate, dest_reg     ; Load immediate into upper 16 bits
+```
+
+**Implementation shows (codegen.rs:248-254):**
+```rust
+fn build_load_immediate_instruction(...) -> Result<Instruction, AssembleError> {
+    // ...
+    match opcode {
+        Opcode::Lli => {
+            let instruction_args = args!(I, immediate: value as u16, r1: dest);
+            Ok(Instruction::LoadLowerImmediate(instruction_args))
+        }
+        Opcode::Lui => {
+            let upper_value = value >> 16;  // Shifts right by 16!
+            let instruction_args = args!(I, immediate: upper_value as u16, r1: dest);
+            Ok(Instruction::LoadUpperImmediate(instruction_args))
+        }
+```
+
+**Reality:** When assembling `lui immediate, reg`, the assembler:
+1. Takes the immediate value
+2. Shifts it RIGHT by 16 bits
+3. Stores the result in the instruction
+
+**This means:**
+```asm
+lli 0x1234, rg0    ; Stores 0x1234 in lower 16 bits
+lui 0xABCD0000, rg0  ; Right-shifts to 0xABCD, stores in upper 16 bits
+```
+
+**Or more likely, the assembler expects:**
+```asm
+lli 0x1234, rg0    ; Stores 0x1234 in lower 16 bits
+lui 0xABCD, rg0    ; Stores 0xABCD in upper 16 bits (no shift needed)
+```
+
+**Documentation needs clarification on what immediate value format LUI expects!**
+
+---
+
+### 10. **Data Definition Encoding** ⚠️ IMPORTANT
+
+**Implementation (expand.rs:217-267):**
+```rust
+fn process_dx_data(args: Vec<Token>, size: usize) -> Result<Vec<u32>, AssembleError> {
+    for token in args {
+        match token {
+            Token::StringLit(mut s) => {
+                s.push('\0');  // Automatically adds null terminator!
+                for ch in s.chars() {
+                    let mut char_buf = [0u8; 4];
+                    let char_bytes = ch.encode_utf8(&mut char_buf);
+                    buffer.extend_from_slice(char_bytes.as_bytes());
+                }
+            }
+            Token::Immediate(value) => {
+                buffer.extend_from_slice(&value.to_be_bytes());  // BIG ENDIAN!
+            }
+```
+
+**Key findings:**
+1. String literals automatically get null terminator appended
+2. Numeric values are stored in **BIG ENDIAN** format (to_be_bytes)
+3. Documentation says "little-endian byte order" globally
+
+**Contradiction:** Data definition uses BIG ENDIAN, but doc says LITTLE ENDIAN!
+
+---
+
+### 11. **Segment Instruction** 📝 UNDOCUMENTED
+
+**Implementation has a SEGMENT instruction (0x27/0x3F):**
+```rust
+Segment(u32) = 0x3F,
+```
+
+**This is completely undocumented!**
+
+From model.rs:
+```rust
+Self::Segment => write!(f, "[SEGMENT]"),
+```
+
+From codegen.rs:
+```rust
+Opcode::Segment => build_segment_instruction(&args),
+```
+
+**Purpose unclear, needs documentation!**
+
+---
+
+### 12. **Data Instruction** 📝 UNDOCUMENTED
+
+**Implementation has a DATA instruction (0x3E):**
+```rust
+Data(u32) = 0x3E,
+```
+
+**This appears to be a meta-instruction for embedding raw data, but it's undocumented in the assembly reference!**
+
+---
+
+### 13. **INC/DEC Instruction Encoding** ⚠️ MINOR
+
+**Implementation (codegen.rs:293-299):**
+```rust
+fn build_inc_dec_instruction(opcode: Opcode, args: &[Token]) -> Result<Instruction, AssembleError> {
+    let reg = expect_token!(reg_token, Register)?;
+    match opcode {
+        Opcode::Inc => Ok(Instruction::Increment(args!(R, sr1: reg))),
+        Opcode::Dec => Ok(Instruction::Decrement(args!(R, sr1: reg))),
+```
+
+**Reality:** INC/DEC only set SR1 field, not DR field.
+
+**But args.rs shows:**
+```rust
+impl RTypeArgs {
+    pub fn new(...) -> Self {
+        let sr1 = sr1.unwrap_or_default();  // Defaults to NoReg
+        let dr = dr.unwrap_or_default();     // Defaults to NoReg
+```
+
+**So the DR field gets set to NoReg, which is correct per documentation.**
+
+**However, the Display impl (instructions.rs:449) shows:**
+```rust
+Self::Increment(a) | Self::Decrement(a) => write!(f, " {}", a.sr1),
+```
+
+**This is correct - only shows SR1 in disassembly.**
+
+---
+
+### 14. **Shift Instruction Operand Order** ⚠️ MINOR
+
+**Implementation (codegen.rs:301-312):**
+```rust
+fn build_shift_instruction(opcode: Opcode, args: &[Token]) -> Result<Instruction, AssembleError> {
+    let reg = expect_token!(reg_token, Register)?;
+    let amount = expect_token!(amount_token, Immediate)? as u8;
+    
+    match opcode {
+        Opcode::Shl => Ok(Instruction::ShiftLeft(args!(R, sr1: reg, shamt: amount))),
+```
+
+**This only handles LITERAL shift amounts, not REGISTER shift amounts!**
+
+**Documentation states both are supported:**
+```asm
+shl rg0, 2      ; Literal shift
+shl rg0, rg1    ; Register shift
+```
+
+**The current codegen only handles the literal case!**
+
+**This is a BUG in the implementation - register shifts aren't properly assembled!**
+
+---
+
+### 15. **Jump Instruction Operand Order** ⚠️ CONFUSION
+
+**Documentation shows assembly syntax:**
+```asm
+jmp addr [, offset_reg]
+```
+
+**But implementation (codegen.rs:256-270):**
+```rust
+fn build_jump_instruction(opcode: Opcode, args: &[Token]) -> Result<Instruction, AssembleError> {
+    let address = expect_token!(address_token, Immediate)?;
+    let offset = expect_token!(offset_token, Register)?;
+    let instruction_args = args!(I, immediate: address as u16, r1: offset);
+```
+
+**This expects:**
+1. First arg: immediate (address)
+2. Second arg: register (offset)
+
+**So assembly syntax should be:**
+```asm
+jmp immediate, offset_register
+```
+
+**Example:**
+```asm
+jmp 0x1000, zero   ; Jump to 0x1000
+jmp 4, ret         ; Jump to (ret + 4)
+```
+
+**Documentation syntax is correct, but parameter names are confusing!**
+
+The "address" is actually an OFFSET, and the register is the BASE!
+
+**Better naming:**
+```asm
+jmp offset, base_register
+; Target = base_register + offset
+```
+
+---
+
+### 16. **NOT Instruction Operand Count** ✅ MINOR ISSUE
+
+**Documentation shows:**
+```asm
+not src, dest       ; Two operands
+```
+
+**Implementation (instructions.rs:428-429):**
+```rust
+Self::Compare(args) | Self::Not(args) => {
+    write!(f, " {}, {}", args.sr1, args.sr2)
+}
+```
+
+**This displays BOTH sr1 and sr2 for NOT!**
+
+**But codegen.rs:354-362:**
+```rust
+fn build_not_instruction(args: &[Token]) -> Result<Instruction, AssembleError> {
+    let reg = expect_token!(reg_token, Register)?;
+    let dest = expect_token!(dest_token, Register)?;
+    Ok(Instruction::Not(args!(R, sr1: reg, dr: dest)))
+```
+
+**Sets sr1 and dr, NOT sr1 and sr2!**
+
+**The Display impl is WRONG - should show sr1 and dr:**
+```rust
+Self::Not(args) => write!(f, " {}, {}", args.sr1, args.dr)
+```
+
+**This is a display bug in the implementation!**
+
+---
+
+### 17. **Register File Indexing** ✅ CORRECT
+
+**Documentation and implementation both agree:**
+- 0x00-0x0F: rg0-rgf (general purpose)
+- 0x10: acc
+- 0x11: spr
+- 0x12: bpr
+- 0x13: ret
+- 0x14: idr
+- 0x15: mmr
+- 0x16: zero
+- 0x17: noreg
+
+**This matches perfectly.**
+
+---
+
+### 18. **Immediate Arithmetic Destination** ⚠️ MINOR
+
+**Implementation (codegen.rs:314-330):**
+```rust
+fn build_arithmetic_immediate_instruction(...) -> Result<Instruction, AssembleError> {
+    let reg = expect_token!(reg_token, Register)?;
+    let immediate = expect_token!(immediate_token, Immediate)? as u16;
+    let dest = expect_token!(dest_token, Register)?;
+    let instruction_args = args!(I, immediate: immediate, r1: reg, r2: dest);
+```
+
+**This REQUIRES three arguments:**
+1. Source register
+2. Immediate value
+3. Destination register
+
+**But documentation says destination is optional:**
+```
+iadd src_reg, imm [, dest_reg]   ; dest optional
+```
+
+**Reality:** The assembler REQUIRES the destination register!
+
+**If you want in-place operation:**
+```asm
+iadd rg0, 10, rg0    ; Required to specify rg0 twice
+```
+
+**Not:**
+```asm
+iadd rg0, 10         ; This won't work!
+```
+
+**Documentation is misleading - destination is NOT optional!**
+
+---
+
+### 19. **Memory Instruction Offsets** ✅ CORRECT
+
+**Implementation correctly handles signed 16-bit offsets:**
+```rust
+let offset = expect_token!(offset_token, Immediate)? as u16;
+```
+
+**These are stored as u16 but interpreted as signed i16 at runtime.**
+
+**Documentation is correct about this.**
+
+---
+
+### 20. **Instruction Opcode Values** ✅ VERIFIED
+
+Comparing model.rs opcodes with instructions.rs:
+
+| Instruction | model.rs | instructions.rs | Match |
+|-------------|----------|-----------------|-------|
+| Nop | 0x00 | 0x0 | ✅ |
+| Mov | 0x01 | 0x1 | ✅ |
+| MovSigned | 0x02 | 0x2 | ✅ |
+| LoadByte | 0x03 | 0x3 | ✅ |
+| ... | ... | ... | ✅ |
+| AddImmediate | 0x25 | 0x25 | ✅ |
+| SubImmediate | 0x26 | 0x26 | ✅ |
+| Segment | 0x27 | 0x3F | ❌ MISMATCH! |
+
+**CRITICAL:** Segment instruction has opcode **0x27** in model.rs but **0x3F** in instructions.rs!
+
+---
+
+## Summary of Critical Issues
+
+### Must Fix in Documentation:
+
+1. ✅ **Stack grows DOWNWARD** - flip all diagrams
+2. ✅ **CALL expansion** - uses stack, not ret register directly
+3. ✅ **RETURN expansion** - loads from stack, jumps to ret+4
+4. ✅ **Stack frame layout** - flip diagram vertically
+5. ✅ **Load pseudo scratch register** - uses DEST reg, not rgf
+6. ✅ **Store pseudo scratch register** - uses ACC, not rgf
+7. ✅ **Add PUSHA/POPA documentation**
+8. ✅ **Add SEGMENT instruction documentation**
+9. ✅ **Add DATA instruction documentation**
+10. ✅ **Clarify LUI immediate value handling**
+11. ✅ **Fix endianness** - data definition uses BIG endian
+12. ✅ **IADD/ISUB destination NOT optional**
+13. ✅ **Add "null" as alias for noreg**
+14. ✅ **Fix Segment opcode** - 0x27 or 0x3F?
+
+### Potential Implementation Bugs:
+
+1. ⚠️ **Shift instruction** - doesn't handle register shifts
+2. ⚠️ **NOT display** - shows sr2 instead of dr
+3. ⚠️ **RETURN +4 offset** - why is this needed?
+4. ⚠️ **Segment opcode mismatch** - 0x27 vs 0x3F
+
+### Minor Documentation Improvements:
+
+1. Add explicit examples of stack growth direction
+2. Show complete memory layout diagrams
+3. Document which registers are volatile/preserved
+4. Add troubleshooting section for common mistakes
+5. Clarify jump instruction parameter semantics

docs/ISA REVISION notes.md

@@ -0,0 +1,4 @@
+
+- we definitely need to be able to use registers for shift operations.
+- we need logical boolean operations in addition to the bitwise ones.
+- better conditionals.

docs/inconsistencies.md

@@ -0,0 +1,149 @@
+# DSA Documentation Inconsistencies Analysis
+
+## 1. Register Descriptions
+
+### Issue: System Registers vs Assembly-Accessible Registers
+- `registers.md` lists MAR, STS, CIR, MDR as "System" registers
+- These are NOT mentioned in `dsa_assembly_reference.md` or `instruction_set.md`
+- **Resolution**: System registers are internal CPU registers not directly accessible in assembly. They should be documented separately from programmer-accessible registers.
+
+### Issue: Register Naming Inconsistencies
+- `registers.md` uses `RG0-RGF` (uppercase)
+- `dsa_assembly_reference.md` uses `rg0-rgf` (lowercase)
+- **Resolution**: Assembly syntax should be lowercase (standard convention)
+
+### Issue: NOREG Register
+- `registers.md`: "Loads/using as dest register must cause an illegal instruction trap"
+- `dsa_assembly_reference.md`: "on-read/write: illegal instruction fault"
+- **Resolution**: Consistent terminology needed - use "illegal instruction fault"
+
+## 2. Instruction Operand Order Inconsistencies
+
+### Issue: Load Instructions
+- `instruction_set.md`: `LDB BaseReg, Offset, DestReg`
+- `dsa_assembly_reference.md`: `LDB base_reg, dest_reg [, offset]`
+- **Resolution**: Assembly reference shows standard syntax (base, dest, offset optional), instruction set shows encoding order
+
+### Issue: Store Instructions
+- `instruction_set.md`: `STB SrcReg, BaseReg, Offset`
+- `dsa_assembly_reference.md`: `STB src_reg, base_reg [, offset]`
+- **Resolution**: Consistent - offset is optional
+
+### Issue: Immediate Load Instructions
+- `instruction_set.md`: `LLI DstReg, Value` (destination first)
+- `dsa_assembly_reference.md`: `LLI imm, dest_reg` (immediate first)
+- **Resolution**: Assembly reference shows gas-style syntax (source, dest), instruction set shows encoding order
+
+### Issue: Jump Instructions
+- `instruction_set.md`: `JMP DestReg, Offset | Address`
+- `dsa_assembly_reference.md`: `JMP addr [, offset_reg]` or `JMP imm, offset_reg`
+- **Resolution**: Different perspectives - instruction set shows encoding, assembly shows usage
+
+## 3. Instruction Behavior Differences
+
+### Issue: IADD/ISUB Operands
+- `instruction_set.md`: `IADD Src1, Literal, Dest` (3 operands)
+- `dsa_assembly_reference.md`: `IADD src_reg, imm [, dest_reg]` (dest optional)
+- **Resolution**: Assembly allows dest to default to src_reg
+
+### Issue: SHL/SHR Operands
+- `instruction_set.md`: `SHL Reg, Literal | ValReg`
+- `dsa_assembly_reference.md`: `SHL reg, shift_amount`
+- **Resolution**: Both literal and register shifts supported
+
+## 4. Pseudo-Instruction Inconsistencies
+
+### Issue: PUSH/POP Expansion
+- `pseudoinstructions.md`: 
+  - PUSH = `INC SPR` then `STW register, SPR`
+  - POP = `LDW SPR, register` then `DEC SPR`
+- Standard stack conventions suggest PUSH should decrement (grow down)
+- **Resolution**: Clarify stack growth direction
+
+### Issue: LDB/LDH/LDW Pseudo vs Hardware
+- `pseudoinstructions.md` lists LDB, LDH, LDW as pseudo-instructions with label addressing
+- `instruction_set.md` lists them as hardware instructions
+- **Resolution**: Both exist - hardware instructions use registers, pseudo-instructions add label support
+
+### Issue: LWI Naming
+- `dsa_assembly_reference.md`: LWI = Load Word Immediate (load address)
+- Could be confused with "Load Word Immediate" (load literal value)
+- **Resolution**: LWI specifically means "Load Word address Into register"
+
+## 5. Calling Convention Details
+
+### Issue: Argument Offsets
+- Calling convention says "first 3 args at offsets 8, 12, 16"
+- This assumes 32-bit words (4 bytes each)
+- Offset 8 is position of first argument (after return address at offset 4, and old BPR at offset 0)
+- **Resolution**: Clarify that SPR+0 = old BPR, SPR+4 = return address, SPR+8 = first arg
+
+### Issue: Return Value Location
+- Says "Store return value (if any) to `spr+8`"
+- This overwrites the first argument
+- **Resolution**: This is intentional - return value replaces first argument position after cleanup
+
+## 6. Missing Information
+
+### From instruction_set.md not in assembly reference:
+- Instruction encoding details (R-type, I-type, J-type)
+- Hex opcodes for each instruction
+- Alignment requirements for memory operations
+- Sign extension behavior details
+
+### From assembly reference not in instruction_set:
+- Complete pseudo-instruction expansions showing what they compile to
+- Library examples (multiply, print)
+- Detailed calling convention walkthrough
+- Module system (INCLUDE directive)
+
+### From registers.md not elsewhere:
+- STS (Status Register) bit layout
+- Boot values for status flags
+- System registers (MAR, STS, CIR, MDR)
+
+## 7. Terminology Inconsistencies
+
+- "halfword" vs "half-word" vs "16-bit value"
+- "word" assumed to be 32-bit (should be explicit)
+- "register" vs "reg" in syntax
+- "immediate" vs "literal" vs "constant"
+
+## 8. Critical Missing Details
+
+### CALL and RETURN Pseudo-instructions
+- Assembly reference shows them but doesn't show their expansion
+- Need to document what they expand to
+
+### Label Addressing Mode
+- Shows expansions for loads/stores with labels
+- Uses RGF as scratch register - should this be documented as reserved for this purpose?
+
+### Stack Direction
+- Not explicitly stated whether stack grows up or down
+- PUSH uses INC SPR (suggests growing up) - unusual!
+
+## Recommendations
+
+1. **Separate Documentation into Logical Layers**:
+   - ISA Specification (hardware-level, for CPU implementers)
+   - Assembly Language Reference (for programmers)
+   - ABI/Calling Convention (for compiler/linker writers)
+
+2. **Standardize Terminology**:
+   - Use consistent casing (lowercase for assembly mnemonics)
+   - Define terms clearly (word = 32-bit, halfword = 16-bit, byte = 8-bit)
+   - Distinguish "literal" (immediate value in code) from "address" (memory location)
+
+3. **Document Stack Convention Clearly**:
+   - Explicitly state stack grows upward (unusual but valid)
+   - Show memory layout diagrams
+
+4. **Show Complete Pseudo-instruction Expansions**:
+   - CALL, RETURN need full expansion documentation
+   - Document which register(s) are used as temporaries
+
+5. **Clarify Register Usage Conventions**:
+   - ACC: used by pseudo-instructions, volatile
+   - RGF: used by label addressing, volatile
+   - RG0-RGE: general purpose, callee may use per calling convention

docs/todo.md

@@ -0,0 +1,26 @@
+# General TODO's
+
+# Bugfixes
+- [x] [EASY] Investigate logical and operator not compiling - either a lexer or parser issue.
+  - **note**: this was a parser issue.
+
+# Missing features
+- [x] [MEDIUM] Get shift operations working correctly.
+- [ ] [MEDIUM] proper prefix/postfix inc/dec implementation. slightly more complex as we need to check for a variable and modify it in place
+- [ ] [EASY] Add multiply and divide operations to code generation
+  - **note**: very easy to do but our division algorithm is hopelessly slow so not worth doing for now.
+
+# Performance Improvements
+- [ ] [MEDIUM] implement a proper div/mod library that's not slow af.
+- [ ] [HARD] Immediate operations for values that support it (up to +/- u16::max for addi and subi respectively) 
+  - this requires significant complexity in code generation as we need to traverse down the tree when we come across these operations to prevent additional register allocations.
+
+# Compiler optimisations
+
+# Codegen improvements
+- [ ] [MEDIUM / time consuming] Add scoping to code generation
+- [ ] [MEDIUM / time consuming] Rewrite entire codegen to imrpove code quality and make the code more readable.
+  - [ ] type-safe instruction builder
+    - [ ] Instruction & Register enums
+    - [ ] Instruction builder helper fns eg `fn add(left: &Register, right: &Register, dest: &Register) -> Instruction`
+    - [ ] Instruction Block types.

dsa_editor/src/syntax/dsc.rs

@@ -0,0 +1,25 @@
+use super::Syntax;
+use std::collections::BTreeSet;
+
+impl Syntax {
+    pub fn dsc() -> Self {
+        Syntax {
+            language: "Damn Simple Code",
+            case_sensitive: false,
+            comment: "//",
+            comment_multiline: ["/*", "*/"],
+            hyperlinks: BTreeSet::from(["http"]),
+            keywords: BTreeSet::from([
+                "include", "fn", "let", "const", "static", "if", "else", "while", "for",
+                "break", "continue", "loop", "return",
+            ]),
+            types: BTreeSet::from([
+                "u32", "u16", "u8", "i32", "i16", "i8", "str", "char", "bool", "void",
+            ]),
+            special: BTreeSet::from([
+                ",", ";", ".", ":", "=", "+", "-", "*", "/", "%", "&", "|", "^", "~",
+                "!", "?", "<", ">", "<<", ">>", "==", "!=", "<=", ">=", "&&", "||",
+            ]),
+        }
+    }
+}

dsa_editor/src/syntax/mod.rs

@@ -1,5 +1,6 @@
 #![allow(dead_code)]
 pub mod dsa;
+pub mod dsc;
 
 use std::collections::BTreeSet;
 use std::hash::{Hash, Hasher};

dsx-build/Cargo.toml

@@ -0,0 +1,10 @@
+[package]
+name = "dsx-build"
+version.workspace = true
+edition.workspace = true
+authors.workspace = true
+
+[dependencies]
+compiler = { path = "../compiler" }
+assembler = { path = "../assembler" }
+chrono = "0.4.43"

dsx-build/src/main.rs

@@ -0,0 +1,200 @@
+use std::process::{Command, Stdio};
+use std::{
+    env, fs,
+    path::{Path, PathBuf},
+};
+
+use crate::templates::{Dsa, Dsc, Template};
+
+mod templates;
+
+/// Run a command and exit on failure.
+fn run(cmd: &mut Command) {
+    let status = cmd.status().expect("failed to execute command");
+    if !status.success() {
+        std::process::exit(1);
+    }
+}
+
+fn main() {
+    // Very small CLI – only three sub‑commands.
+    let args: Vec<String> = env::args().collect();
+    if args.len() < 2 {
+        eprintln!("Usage: dsx-build <new|build|package> [options]");
+        std::process::exit(1);
+    }
+    match args[1].as_str() {
+        "new" => cmd_new(&args[2..]),
+        "build" => cmd_build(),
+        "package" => todo!("Package manager stub – not implemented yet."),
+        _ => {
+            eprintln!("Unknown command: {}", args[1]);
+            std::process::exit(1);
+        }
+    }
+}
+
+// ----------  new project ----------------------------------------------------
+fn cmd_new(args: &[String]) {
+    let mut lang = "dsa";
+    for i in 0..args.len() {
+        if args[i] == "--lang" && i + 1 < args.len() {
+            lang = &args[i + 1];
+        }
+    }
+
+    let lib = args.contains(&"--lib".to_string());
+
+    // Determine project root: a subdirectory named after the supplied --name argument.
+    let mut name_opt = None;
+    for i in 0..args.len() {
+        if args[i] == "--name" && i + 1 < args.len() {
+            name_opt = Some(&args[i + 1]);
+            break;
+        }
+    }
+
+    let project_name = match name_opt {
+        Some(name) => name.to_string(),
+        None => {
+            eprintln!("Error: --name argument required");
+            std::process::exit(1);
+        }
+    };
+
+    let cwd = env::current_dir().unwrap();
+    let src_path = cwd.join(&project_name).join("src");
+    fs::create_dir_all(&src_path).expect("Failed to create project directory");
+
+    match lang {
+        "dsa" => {
+            // Minimal DSA binary template.
+            let path = src_path.join(format!("main.dsa"));
+
+            let template = Dsa::create(&project_name, lib);
+
+            fs::write(path, template).expect("Unable to write DSA file");
+        }
+        "dsc" => {
+            let path = src_path.join(format!("main.dsc"));
+
+            let template = Dsc::create(&project_name, lib);
+
+            fs::write(path, template).expect("Unable to write DSC file");
+        }
+        _ => {
+            eprintln!("Unsupported language: {}", lang);
+            std::process::exit(1);
+        }
+    }
+
+    fs::create_dir_all(src_path.join("lib")).expect("Failed to create lib directory");
+    fs::write(
+        src_path.join("lib/print.dsa"),
+        templates::create_print_lib(),
+    )
+    .expect("Failed to create print.dsa");
+    fs::write(
+        src_path.join("lib/maths.dsa"),
+        templates::create_maths_lib(),
+    )
+    .expect("Failed to create maths.dsa");
+
+    println!(
+        "Created new {} project in {}.",
+        lang,
+        src_path.parent().unwrap().display()
+    );
+}
+
+// ----------  build ----------------------------------------------------------
+fn cmd_build() {
+    let cwd = env::current_dir().unwrap();
+
+    // Detect .dsc or .dsa files in current directory.
+    let mut has_dsc = false;
+    let mut has_dsa = false;
+    for entry in fs::read_dir(&cwd.join("src")).expect("unable to read dir") {
+        if let Ok(entry) = entry {
+            let path = entry.path();
+            if path.extension().and_then(|s| s.to_str()) == Some("dsc") {
+                has_dsc = true;
+            } else if path.extension().and_then(|s| s.to_str()) == Some("dsa") {
+                has_dsa = true;
+            }
+        }
+    }
+
+    if !has_dsc && !has_dsa {
+        eprintln!("No .dsc or .dsa source found in src directory.");
+        std::process::exit(1);
+    }
+
+    // Assemble main.dsa to a dsb binary.
+    println!("Assembling Project to a DSB binary...");
+    let build_dir = cwd.join("build");
+    fs::create_dir_all(&build_dir).expect("Failed to create build directory");
+
+    // Copy everything from `cwd/src` to the build directory.
+    fn copy_recursively(src: &Path, dst: &Path) {
+        if src.is_file() {
+            fs::create_dir_all(dst.parent().unwrap())
+                .expect("Failed to create parent directory");
+            fs::copy(src, dst).expect("Failed to copy file");
+        } else if src.is_dir() {
+            for entry in fs::read_dir(src).expect("Unable to read source dir") {
+                let entry = entry.expect("Failed to read entry");
+                let child_src = entry.path();
+                let child_dst = dst.join(entry.file_name());
+                copy_recursively(&child_src, &child_dst);
+            }
+        }
+    }
+
+    let src_dir = cwd.join("src");
+    if src_dir.exists() {
+        copy_recursively(&src_dir, &build_dir);
+    }
+
+    // Change current working directory to the build directory.
+    env::set_current_dir(&build_dir).expect("Failed to change to build directory");
+
+    if has_dsc {
+        println!("Compiling DSC to DSA...");
+        fn compile_recursive(path: &Path) {
+            if path.is_dir() {
+                for entry in fs::read_dir(path).expect("unable to read dir") {
+                    let entry = entry.expect("failed to read entry");
+                    compile_recursive(&entry.path());
+                }
+            } else if path.extension().and_then(|s| s.to_str()) == Some("dsc") {
+                let input_path = path;
+                let output_path = path.with_extension("dsa");
+                compiler::compile_file(&input_path, &output_path).unwrap_or_else(|e| {
+                    eprintln!("Failed to compile {:?}: {}", input_path, e);
+                    std::process::exit(1);
+                });
+            }
+        }
+        compile_recursive(&build_dir);
+    }
+
+    // Replace .dsc with .dsa only in include statements, recursively for each file.
+    let mut sed_cmd = Command::new("bash");
+    sed_cmd.args(&[
+        "-c",
+        &format!(
+            "find \"{}\" -type f -name '*.dsa' -exec sed -i '/^include/ s/\\.dsc/.dsa/g' {{}} +",
+            build_dir.display()
+        ),
+    ]);
+    run(&mut sed_cmd);
+
+    fs::create_dir_all(&cwd.join("artifacts")).expect("Failed to create build directory");
+    assembler::assemble_file("./main.dsa", "../artifacts/out.dsb").unwrap_or_else(|e| {
+        eprintln!("Failed to assemble {:?}: {}", "./main.dsa", e);
+        std::process::exit(1);
+    });
+
+    println!("Build finished. Binary at {}/main.dsb", build_dir.display());
+}

dsx-build/src/templates.rs

@@ -0,0 +1,589 @@
+pub trait Template {
+    fn lib(project: &str) -> String;
+    fn bin(project: &str) -> String;
+
+    fn create(project: &str, lib: bool) -> String {
+        if lib {
+            Self::lib(project)
+        } else {
+            Self::bin(project)
+        }
+    }
+}
+
+pub struct Dsa;
+pub struct Dsc;
+
+impl Template for Dsa {
+    fn lib(project: &str) -> String {
+        format!(
+            r#"//
+lib.dsa
+// usage:
+//
+// include {project} "<relative path>"
+//
+// usage for {project}_main:
+// push (arg1)
+// push (arg0)
+// call {project}::{project}_main
+// pop (arg0)
+// pop (arg1)
+
+// Example data declarations
+// dw example_data: 0x0000
+
+// Main function template
+{project}_main:
+    // the correct way to start a function as defined by the calling convention
+    push bpr
+    mov spr, bpr
+
+    // explanation of how to access args
+    ldw bpr, rg0, 8  // arg 0
+    ldw bpr, rg0, 12 // arg 1
+
+    // your code goes here
+    // Example: load example_data into rg1
+    // ldw example_data, rg1
+
+    // the correct way to end a function as defined by the calling convention
+    mov bpr, spr
+    pop bpr
+    return
+"#,
+        )
+    }
+
+    fn bin(project: &str) -> String {
+        format!(
+            r#"
+// GENERATED BY DSX-BUILD
+// Generated at: {timestamp}
+// Project name: {project}
+
+// Imports
+include print: "./lib/print.dsa"
+
+// Globals & Reserved Memory
+dw stack: 0x10000
+db message: "Process Exited with code:"
+
+// Entry Point
+_init:
+    ldw stack, bpr
+    mov bpr, spr
+    push zero
+    call main
+    call print::print_newline
+    lwi message, rg0
+    push rg0
+    call print::print
+    pop zero
+    call print::print_hex_word
+    pop zero
+    hlt
+
+main:
+    push bpr
+    mov spr, bpr
+
+    // Your code goes here
+
+    // Return zero
+    stw zero, bpr, 8
+
+    mov bpr, spr
+    pop bpr
+    return"#,
+            timestamp = chrono::Utc::now().format("%Y-%m-%d %H:%M:%S").to_string()
+        )
+    }
+}
+
+impl Template for Dsc {
+    fn lib(project: &str) -> String {
+        format!(
+            r#"
+// GENERATED BY DSX-BUILD
+// Generated at: {timestamp}
+// Project name: {project}
+
+// Imports
+include print: "./lib/print.dsa";
+
+// Main Function
+fn {project}_main() -> u32 {{
+    return 0;
+}}"#,
+            timestamp = chrono::Utc::now().format("%Y-%m-%d %H:%M:%S").to_string()
+        )
+    }
+
+    fn bin(project: &str) -> String {
+        format!(
+            r#"
+// GENERATED BY DSX-BUILD
+// Generated at: {timestamp}
+// Project name: {project}
+
+// Imports
+include print: "./lib/print.dsa";
+
+// Main Function
+fn main() -> u32 {{
+    return 0;
+}}"#,
+            timestamp = chrono::Utc::now().format("%Y-%m-%d %H:%M:%S").to_string()
+        )
+    }
+}
+
+pub fn create_print_lib() -> String {
+    format!(
+        r#"
+// lib:
+// print.dsa
+
+// usage:
+//
+// include print "<relative path>""
+//
+// usage for print:
+//      push (register containing address of string)
+//      push pcx
+//      jmp print::print
+//
+// usage for reset:
+//      push pcx
+//      jmp print::reset
+//
+// usage for clear:
+//      push pcx
+//      jmp print::clear
+//
+// usage for print_byte:
+//      push (register containing byte)
+//      push pcx
+//      jmp print::print_byte
+//
+// usage for print_word:
+//      push (register containing word)
+//      push pcx
+//      jmp print::print_word
+//
+// usage for print_num:
+//      push (register containing number to print in decimal)
+//      push pcx
+//      jmp print::print_num
+//
+
+include maths "./maths.dsa"
+
+dw display: 0x20000
+dw current: 0x20000
+
+// ------------------------------------------
+// prints the string at addr(arg[0]) to the screen. (no trailing whitespace unless explicitly provided)
+print:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8
+    ldw current, rg1
+
+_print_loop:
+    ldb rg0, acc
+    cmp acc, zero
+    jeq _end
+    stb acc, rg1
+
+    addi rg0, 1
+    addi rg1, 1
+
+    jmp _print_loop
+
+// ------------------------------------------
+println:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8
+    ldw current, rg1
+
+_println_loop:
+    ldb rg0, acc
+    cmp acc, zero
+    jeq _println_end
+    stb acc, rg1
+
+    addi rg0, 1
+    addi rg1, 1
+
+    jmp _println_loop
+
+_println_end:
+    call print_newline
+    jmp _end
+
+// ------------------------------------------
+// prints the value of arg[0] to the screen.
+print_word:
+    // initialise
+    push bpr
+    mov spr, bpr
+
+    // load byte into acc
+    ldw bpr, rg0, 8
+    ldw current, rg1
+
+    addi rg1, 3
+
+    stb rg0, rg1
+    subi rg1, 1
+    shr rg0, 8
+    stb rg0, rg1
+    subi rg1, 1
+    shr rg0, 8
+    stb rg0, rg1
+    subi rg1, 1
+    shr rg0, 8
+    stb rg0, rg1
+
+    addi rg1, 4
+    jmp _end
+
+// ------------------------------------------
+// prints the last byte of arg[0] to the screen.
+print_byte:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8
+    ldw current, rg1
+
+    stb rg0, rg1
+    addi rg1, 1
+    jmp _end
+
+// ------------------------------------------
+// prints the value of arg[0] to the screen in hex.
+print_hex_word:
+    push bpr
+    mov spr, bpr
+
+    ldw current, rg1
+
+    ldb bpr, rg0, 8
+    push rg0
+    call _print_hex_byte
+    addi spr, 4
+
+    ldb bpr, rg0, 9
+    push rg0
+    call _print_hex_byte
+    addi spr, 4
+
+    ldb bpr, rg0, 10
+    push rg0
+    call _print_hex_byte
+    addi spr, 4
+
+    ldb bpr, rg0, 11
+    push rg0
+    call _print_hex_byte
+    addi spr, 4
+
+    jmp _end
+
+// ------------------------------------------
+// prints the last byte of arg[0] to the screen in hex.
+print_hex_byte:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8
+    ldw current, rg1
+
+    call _print_hex_byte
+    jmp _end
+
+// function body
+_print_hex_byte:
+    // mask to get lower nibble
+    lli 0xF, rg2
+    // save rg0 state
+    push rg0
+
+    shr rg0, 4
+    and rg0, rg2, rg0
+    call _print_hex_nibble
+    pop rg0
+
+    and rg0, rg2, rg0
+    call _print_hex_nibble
+    return
+
+// print a hex digit
+_print_hex_nibble:
+    lli 10, rg3
+    cmp rg0, rg3
+    jlt _print_hex_nibble_number
+    addi rg0, 0x37, rg0
+    stb rg0, rg1
+    addi rg1, 1
+    return
+
+// helper function.
+_print_hex_nibble_number:
+    addi rg0, 0x30, rg0
+    stb rg0, rg1
+    addi rg1, 1
+    return
+
+// ------------------------------------------
+// print whitespace
+print_whitespace:
+    push bpr
+    mov spr, bpr
+
+    ldw current, rg1
+    lli 0x20, rg0
+    stb rg0, rg1
+    addi rg1, 1
+    jmp _end
+
+// ------------------------------------------
+// print newline
+print_newline:
+    push bpr
+    mov spr, bpr
+
+    // load variables into registers
+    ldw display, rg0
+    ldw current, rg1
+
+    // get the offset from the display base
+    sub rg1, rg0, rg0
+
+    lwi 80, rg2
+    pusha 3
+    push rg0
+    push rg2
+    call maths::divmod
+    pop zero  // result
+    pop rg3 // remainder
+    popa 3
+
+    sub rg1, rg3, rg2
+    addi rg2, 80, rg1
+
+    // _end saves the display state
+    jmp _end
+
+// ------------------------------------------
+// prints arg[0] as a decimal number to the screen.
+print_num:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8      // load number to print
+    lli 0, rg5           // rg5 = digit counter
+
+    // check if number is zero
+    cmp rg0, zero
+    jne _print_num_extract_digits
+
+    // special case: print '0' for zero
+    lli 0x30, rg6
+    push rg6             // push digit to stack buffer
+    lli 1, rg5           // we have 1 digit
+    jmp _print_num_output
+
+_print_num_extract_digits:
+    // divide by 10 repeatedly to get digits
+    cmp rg0, zero
+    jeq _print_num_output
+
+    // call divmod(rg0, 10)
+    push rg0             // dividend
+    lli 10, rg1
+    push rg1             // divisor (10)
+    call maths::divmod
+    pop rg0              // quotient (continue dividing this)
+    pop rg1              // remainder (the digit)
+
+    // convert digit to ASCII and push to stack buffer
+    addi rg1, 0x30, rg6  // convert to ASCII
+    push rg6             // push digit to stack
+    inc rg5              // increment digit counter
+
+    jmp _print_num_extract_digits
+
+_print_num_output:
+    // now print digits (pop them off in reverse order)
+    ldw current, rg1     // get display pointer
+
+_print_num_output_loop:
+    // check if we've printed all digits
+    cmp rg5, zero
+    jeq _print_num_done
+
+    // pop digit and print it
+    pop rg6
+    stb rg6, rg1
+    addi rg1, 1
+    dec rg5
+
+    jmp _print_num_output_loop
+
+_print_num_done:
+    jmp _end
+
+// ------------------------------------------
+// resets the cursor position on the screen to 0x20000. (0,0)
+reset:
+    push bpr
+    mov spr, bpr
+    ldw display, rg1
+    jmp _end
+
+// ------------------------------------------
+// clears the screen
+clear:
+    push bpr
+    mov spr, bpr
+    // display size = 2000 bytes / 500 words
+    lli 500 rg0
+    ldw display, rg1
+
+_clear_loop:
+    dec rg0
+    stw zero, rg1
+    addi rg1, 4
+    cmp rg0, zero
+    jgt _clear_loop
+    jmp _end
+
+// ------------------------------------------
+// return
+_end:
+    stw rg1, current
+
+    mov bpr, spr
+    pop bpr
+    return
+"#
+    )
+}
+
+pub fn create_maths_lib() -> String {
+    format!(
+        r#"
+// multiply.dsa
+// usage:
+//
+// include multiply "<relative path>"
+//
+// usage for multiply:
+// push (arg1)
+// push (arg0)
+// call multiply::multiply
+// pop (arg0)
+// pop (arg1)
+
+multiply:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8  // load op 2
+    ldw bpr, rg1, 12 // load op 1
+    lwi 0, rg2       // initialise rg2 to zero
+
+_multiply_loop:
+    add rg2, rg0, rg2
+    dec rg1
+
+    cmp rg1, zero
+    jgt _multiply_loop
+
+_multiply_end:
+    stw rg2, bpr, 8
+
+    mov bpr, spr
+    pop bpr
+    return
+
+divmod:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg1, 8  // load op 2
+    ldw bpr, rg0, 12 // load op 1
+
+    lli 0, rg3
+
+_divmod_loop:
+    cmp rg0, rg1
+    jlt _divmod_end
+
+    sub rg0, rg1, rg0
+    inc rg3
+
+    jmp _divmod_loop
+
+_divmod_end:
+    // store div in first arg
+    // store mod in second arg
+    stw rg3, bpr, 8
+    stw rg0, bpr, 12
+
+    mov bpr, spr
+    pop bpr
+    return
+
+// multiply.dsa - improved version
+// Multiplies two 32-bit numbers using shift-and-add
+//
+// Usage:
+// push operand2  (multiplier)
+// push operand1  (multiplicand)
+// call multiply::multiply
+// pop result
+// pop zero       (discard second argument)
+
+new_multiply:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8     // rg0 = multiplicand
+    ldw bpr, rg1, 12    // rg1 = multiplier
+
+    lli 0, rg2          // rg2 = result (accumulator)
+    lli 32, rg3         // rg3 = bit counter
+
+mult_loop:
+    // Check if lowest bit of multiplier is 1
+    lli 1, acc
+    and rg1, acc, acc   // acc = rg1 & 1
+    cmp acc, zero
+    jeq skip_add        // if (rg1 & 1) == 0, skip addition
+
+    // Add multiplicand to result
+    add rg2, rg0, rg2
+
+skip_add:
+    shl rg0, 1          // shift multiplicand left
+    shr rg1, 1          // shift multiplier right
+
+    dec rg3
+    cmp rg3, zero
+    jgt mult_loop
+
+    stw rg2, bpr, 8     // store result
+    mov bpr, spr
+    pop bpr
+    return
+"#
+    )
+}

emulator/Cargo.toml

@@ -16,6 +16,7 @@ required-features = ["config"]
 [dependencies]
 common = { path = "../common" }
 assembler = { path = "../assembler" }
+compiler = { path = "../compiler" }
 dsa_editor = { path = "../dsa_editor" }
 egui = "0.31.1"
 dirs = "6.0.0"

emulator/src/emulator/system/memory.rs

@@ -83,7 +83,7 @@ impl MemoryUnit for MainStore {
 
     #[inline]
     fn read_word(&mut self, addr: u32) -> Result<u32, ProcessorError> {
-        if addr % 4 != 0 {
+        if !addr.is_multiple_of(4) {
             return Err(ProcessorError::BadMemoryAccess(addr));
         }
 
@@ -116,7 +116,7 @@ impl MemoryUnit for MainStore {
 
     #[inline]
     fn write_word(&mut self, addr: u32, value: u32) -> Result<(), ProcessorError> {
-        if addr % 4 != 0 {
+        if !addr.is_multiple_of(4) {
             return Err(ProcessorError::BadMemoryAccess(addr));
         }
 

emulator/src/emulator/system/model.rs

@@ -286,11 +286,10 @@ impl RegFile {
             Register::Sts => &mut self.sts,
             Register::Cir => &mut self.cir,
             Register::Pcx => &mut self.pcx,
-            _ => return Err(ProcessorError::InvalidRegister(Register::NoReg as u8)),
+            _ => return Err(ProcessorError::InvalidRegister(Register::Null as u8)),
         })
     }
 
-    #[must_use]
     pub const fn get(&self, reg: Register) -> Result<u32, ProcessorError> {
         Ok(match reg {
             Register::Rg0 => self.rg0,
@@ -321,7 +320,7 @@ impl RegFile {
             Register::Cir => self.cir,
             Register::Pcx => self.pcx,
             Register::Zero => 0,
-            _ => return Err(ProcessorError::InvalidRegister(Register::NoReg as u8)),
+            _ => return Err(ProcessorError::InvalidRegister(Register::Null as u8)),
         })
     }
 }

emulator/src/emulator/system/processor/mod.rs

@@ -21,10 +21,6 @@ pub struct Processor {
     pub cache: Cache,
 }
 
-fn log(message: &str) {
-    println!("\x1b[32mINFO:\x1b[0m {message}");
-}
-
 impl Processor {
     #[must_use]
     pub fn new(memory: Box<dyn MemoryUnit>, io_devices: Vec<Arc<dyn IODevice>>) -> Self {
@@ -366,17 +362,13 @@ impl Executable for Instruction {
             // Left shifts the value in Reg by the given amount (either a register, or a
             // literal value)
             Self::ShiftLeft(a) => {
-                let reg = cpu.get(a.sr1)?;
-                let val = a.shamt;
-                *cpu.reg(a.sr1)? = shl(reg, val);
+                *cpu.reg(a.dr)? = shl(cpu.get(a.sr1)?, a.shamt + cpu.get(a.sr2)? as u8);
             }
 
             // Right shifts the value in Reg by the given amount (either a register, or a
             // literal value).
             Self::ShiftRight(a) => {
-                let regval = cpu.get(a.sr1)?;
-                let val = a.shamt;
-                *cpu.reg(a.sr1)? = shr(regval, val);
+                *cpu.reg(a.dr)? = shr(cpu.get(a.sr1)?, a.shamt + cpu.get(a.sr2)? as u8);
             }
 
             // Adds the value of Src2 to Src1 and writes the result to a.dr

emulator/src/emulator/system/processor/tests.rs

@@ -464,7 +464,7 @@ fn test_shift_left_with_shamt() {
     let shl_instr = Instruction::ShiftLeft(RTypeArgs::new(
         Some(Register::Rg1),
         Some(Register::Zero),
-        None,
+        Some(Register::Rg1),
         Some(2),
     ));
 
@@ -485,7 +485,7 @@ fn test_shift_right_with_shamt() {
     let shr_instr = Instruction::ShiftRight(RTypeArgs::new(
         Some(Register::Rg1),
         Some(Register::Zero),
-        None,
+        Some(Register::Rg1),
         Some(2),
     ));
 

emulator/src/emulator/ui/editor.rs

@@ -117,10 +117,7 @@ impl Editor {
                 .file_name()
                 .unwrap_or_else(|| OsStr::new("Unnamed!"))
                 .to_str()
-                .map_or_else(
-                    || unreachable!("File name should be valid UTF-8."),
-                    |ext| ext,
-                );
+                .unwrap_or_else(|| unreachable!("File name should be valid UTF-8."));
         }
         "Unnamed!"
     }
@@ -129,12 +126,9 @@ impl Editor {
         if let Some(path) = &self.path {
             return path
                 .extension()
-                .map_or_else(|| OsStr::new("Unknown!"), |ext| ext)
+                .unwrap_or_else(|| OsStr::new("Unknown!"))
                 .to_str()
-                .map_or_else(
-                    || unreachable!("File name should be valid UTF-8."),
-                    |ext| ext,
-                );
+                .unwrap_or_else(|| unreachable!("File name should be valid UTF-8."));
         }
         "Unknown!"
     }
@@ -393,25 +387,20 @@ impl Editor {
     fn render_editor(&mut self, _state: &mut State, ui: &mut Ui, _ctx: &Context) {
         let available_width = ui.available_width();
         let syntax = match self.extension() {
-            "dsa" => Some(Syntax::new("dsa")),
-            _ => None,
+            "dsa" => Syntax::dsa(),
+            "dsc" => Syntax::dsc(),
+            _ => Syntax::default(),
         };
 
-        let ed = CodeEditor::default()
+        let mut editor = CodeEditor::default()
             .id_source("editor")
             .with_fontsize(12.0)
             .with_rows(0)
             .with_theme(ColorTheme::default())
-            .with_syntax(Syntax::dsa())
+            .with_syntax(syntax)
             .with_numlines(true)
             .desired_width(available_width - 500.0);
 
-        let mut editor = ed.clone();
-
-        if let Some(syntax) = syntax {
-            editor = ed.with_syntax(syntax);
-        }
-
         editor.show(ui, &mut self.text);
     }
 
@@ -451,6 +440,29 @@ impl Editor {
                         .flat_map(|i| i.encode().to_be_bytes().to_vec())
                         .collect();
                 }
+                Some("dsc") => {
+                    let output_path = Path::new(path).with_extension("dsa");
+                    if let Err(e) = compiler::compile_file(path, &output_path) {
+                        self.error = Some(format!("Compiler error: {e}"));
+                    }
+
+                    let mut compiler = CompilerEngine::new();
+                    compiler.start_compilation(&output_path);
+
+                    // Or block until done
+                    let instructions = match compiler.wait_for_result() {
+                        Ok(instructions) => instructions,
+                        Err(e) => {
+                            self.error = Some(format!("Assembler error: {e}"));
+                            return;
+                        }
+                    };
+
+                    self.output = instructions
+                        .iter()
+                        .flat_map(|i| i.encode().to_be_bytes().to_vec())
+                        .collect();
+                }
                 Some("dsb") => {
                     if let Ok(bytes) = fs::read(path) {
                         self.output = bytes;

emulator/src/emulator/ui/loader.rs

@@ -79,10 +79,7 @@ impl Loader {
                 .file_name()
                 .unwrap_or_else(|| OsStr::new("Unnamed!"))
                 .to_str()
-                .map_or_else(
-                    || unreachable!("File name should be valid UTF-8."),
-                    |ext| ext,
-                );
+                .unwrap_or_else(|| unreachable!("File name should be valid UTF-8."));
         }
         "Unnamed!"
     }

emulator/src/emulator/ui/stack_inspector.rs

@@ -51,7 +51,6 @@ impl Component for StackInspector {
                             ui.label("Address");
                             ui.label("Value");
                             ui.end_row();
-
                             for (i, value) in
                                 state.stack_view.chunks(4).take(32).enumerate()
                             {
@@ -59,9 +58,9 @@ impl Component for StackInspector {
                                     "Could not read 4 byte instruction or data! Something is wrong.",
                                 ));
                                 ui.label(format!(
-                                    "{} [{}]",
-                                    i,
-                                    state.reg_file.get(Register::Spr).expect("SPR should never be invalid") - i as u32 * 4
+                                    "+{} [{}]",
+                                    i*4,
+                                    state.reg_file.get(Register::Spr).expect("SPR should never be invalid") + i as u32 * 4
                                 ));
                                 ui.label(format!("0x{value:08X} ({value})"));
                                 ui.end_row();

resources/dsa/example.dsc

@@ -0,0 +1,30 @@
+include print: "./lib/io/print.dsa";
+include arena: "./lib/memory/arena_alloc.dsa";
+
+fn main() -> u32 {
+
+    let x: u32 = 0;
+    let y: u32 = &x;
+
+    let alloc: u32 = arena::new(512);
+    let ptr1: u32 = arena::alloc(alloc, 32);
+    let ptr2: u32 = arena::alloc(alloc, 32);
+
+    print::print_hex_word(alloc);
+    print::print_newline();
+    print::print_hex_word(ptr1);
+    print::print_newline();
+    print::print_hex_word(ptr2);
+    print::print_newline();
+    print::print_num(*ptr2);
+    print::print_newline();
+    *ptr2 = 42;
+
+    print::print_hex_word(ptr2);
+    print::print_newline();
+    print::print_num(*ptr2);
+    print::print_newline();
+    print::println("end");
+
+    return 0;
+}

resources/dsa/lib/error/handlers.dsa

@@ -11,7 +11,7 @@ setup_idt:
 
 	mov bpr, spr
 	pop bpr
-	irt
+	return
 
 setup_hard_fault_handler:
     push bpr
@@ -22,7 +22,7 @@ setup_hard_fault_handler:
 
     mov bpr, spr
     pop bpr
-    irt
+    return
 
 dw hard_fault_err: "FATAL: Illegal Instruction or Memory Access!"
 handle_hard_fault:
@@ -32,3 +32,13 @@ handle_hard_fault:
 	call print::print
 	pop zero
 	hlt
+
+trigger:
+    push bpr
+    mov spr, bpr
+
+    int 0x01
+
+    mov bpr, spr
+    pop bpr
+    return

resources/dsa/lib/io/print.dsa

@@ -28,6 +28,11 @@
 //      push pcx
 //      jmp print::print_word
 //
+// usage for print_num:
+//      push (register containing number to print in decimal)
+//      push pcx
+//      jmp print::print_num
+//
 
 include maths "../maths/core.dsa"
 
@@ -35,7 +40,7 @@ dw display: 0x20000
 dw current: 0x20000
 
 // ------------------------------------------
-// prints the string at addr(arg[0]) to the screen.
+// prints the string at addr(arg[0]) to the screen. (no trailing whitespace unless explicitly provided)
 print:
     push bpr
     mov spr, bpr
@@ -45,13 +50,36 @@ print:
 
 _print_loop:
 	ldb rg0, acc
+	cmp acc, zero
+	jeq _end
 	stb acc, rg1
 
 	addi rg0, 1
 	addi rg1, 1
 
+	jmp _print_loop
+
+// ------------------------------------------
+println:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8
+    ldw current, rg1
+
+_println_loop:
+	ldb rg0, acc
 	cmp acc, zero
-	jne _print_loop
+	jeq _println_end
+	stb acc, rg1
+
+	addi rg0, 1
+	addi rg1, 1
+
+	jmp _println_loop
+
+_println_end:
+    call print_newline
     jmp _end
 
 // ------------------------------------------
@@ -209,6 +237,65 @@ print_newline:
     // _end saves the display state
     jmp _end
 
+// ------------------------------------------
+// prints arg[0] as a decimal number to the screen.
+print_num:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8      // load number to print
+    lli 0, rg5           // rg5 = digit counter
+
+    // check if number is zero
+    cmp rg0, zero
+    jne _print_num_extract_digits
+
+    // special case: print '0' for zero
+    lli 0x30, rg6
+    push rg6             // push digit to stack buffer
+    lli 1, rg5           // we have 1 digit
+    jmp _print_num_output
+
+_print_num_extract_digits:
+    // divide by 10 repeatedly to get digits
+    cmp rg0, zero
+    jeq _print_num_output
+
+    // call divmod(rg0, 10)
+    push rg0             // dividend
+    lli 10, rg1
+    push rg1             // divisor (10)
+    call maths::divmod
+    pop rg0              // quotient (continue dividing this)
+    pop rg1              // remainder (the digit)
+
+    // convert digit to ASCII and push to stack buffer
+    addi rg1, 0x30, rg6  // convert to ASCII
+    push rg6             // push digit to stack
+    inc rg5              // increment digit counter
+
+    jmp _print_num_extract_digits
+
+_print_num_output:
+    // now print digits (pop them off in reverse order)
+    ldw current, rg1     // get display pointer
+
+_print_num_output_loop:
+    // check if we've printed all digits
+    cmp rg5, zero
+    jeq _print_num_done
+
+    // pop digit and print it
+    pop rg6
+    stb rg6, rg1
+    addi rg1, 1
+    dec rg5
+
+    jmp _print_num_output_loop
+
+_print_num_done:
+    jmp _end
+
 // ------------------------------------------
 // resets the cursor position on the screen to 0x20000. (0,0)
 reset:

resources/dsa/lib/maths/core.dsa

@@ -16,6 +16,7 @@ multiply:
 
 	ldw bpr, rg0, 8  // load op 2
 	ldw bpr, rg1, 12 // load op 1
+	lwi 0, rg2       // initialise rg2 to zero
 
 _multiply_loop:
 	add rg2, rg0, rg2
@@ -58,3 +59,46 @@ _divmod_end:
 	mov bpr, spr
 	pop bpr
 	return
+
+// multiply.dsa - improved version
+// Multiplies two 32-bit numbers using shift-and-add
+//
+// Usage:
+// push operand2  (multiplier)
+// push operand1  (multiplicand)
+// call multiply::multiply
+// pop result
+// pop zero       (discard second argument)
+
+new_multiply:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8     // rg0 = multiplicand
+    ldw bpr, rg1, 12    // rg1 = multiplier
+
+    lli 0, rg2          // rg2 = result (accumulator)
+    lli 32, rg3         // rg3 = bit counter
+
+mult_loop:
+    // Check if lowest bit of multiplier is 1
+    lli 1, acc
+    and rg1, acc, acc   // acc = rg1 & 1
+    cmp acc, zero
+    jeq skip_add        // if (rg1 & 1) == 0, skip addition
+
+    // Add multiplicand to result
+    add rg2, rg0, rg2
+
+skip_add:
+    shl rg0, 1          // shift multiplicand left
+    shr rg1, 1          // shift multiplier right
+
+    dec rg3
+    cmp rg3, zero
+    jgt mult_loop
+
+    stw rg2, bpr, 8     // store result
+    mov bpr, spr
+    pop bpr
+    return

resources/dsa/lib/maths/fib.dsa

@@ -5,27 +5,20 @@ fib_n:
 	mov spr, bpr
 
 	ldw bpr, rg0, 8  // load arg
-    mov rg1, rg2
-    lwi 1, rg1
+    lwi 0, rg1
+    lwi 1, rg2
 
-start:
+_start:
     add rg1, rg2, rg3
 
-    pusha 4
-    push rg1
-    call print::print_hex_byte
-    call print::print_newline
-    pop zero
-    popa 4
-
     mov rg2, rg1
     mov rg3, rg2
 
     dec rg0
     cmp rg0, zero
-    jgt start
+    jgt _start
 
-    stw rg1, bpr, 8
+    stw rg3, bpr, 8
     mov bpr, spr
     pop bpr
     return

resources/dsa/lib/memory/arena_alloc.dsa

@@ -0,0 +1,100 @@
+dw heap_start: 196608
+dw heap_end: 262144
+dw heap_current: 196608
+
+new:
+	push bpr
+	mov spr, bpr
+
+	ldw bpr, rg0, 8
+	lli 12, rg1
+	add rg0, rg1, rg2
+	ldw heap_current, rg1
+	add rg1, rg2, rg3
+	ldw heap_end, rg4
+	cmp rg3, rg4
+	lli 0, rg5
+	jle _cmp_end_2
+	lli 1, rg5
+_cmp_end_2:
+	cmp rg5, zero
+	jeq _else_4
+_then_3:
+	lli 0, rg4
+	stw rg4, bpr, 8
+	jmp _ret
+	jmp _end_5
+_else_4:
+	nop
+_end_5:
+	lli 12, rg4
+	add rg1, rg4, rg5
+	add rg1, rg2, rg4
+	stw rg5, rg1
+	lli 4, rg6
+	add rg1, rg6, rg7
+	stw rg5, rg7
+	lli 8, rg6
+	add rg1, rg6, rg7
+	stw rg4, rg7
+	stw rg3, heap_current
+	stw rg1, bpr, 8
+	jmp _ret
+
+alloc:
+	push bpr
+	mov spr, bpr
+
+	ldw bpr, rg0, 8
+	ldw bpr, rg1, 12
+	lli 4, rg2
+	add rg0, rg2, rg3
+	ldw rg3, rg2
+	lli 8, rg3
+	add rg0, rg3, rg4
+	ldw rg4, rg3
+	add rg2, rg1, rg4
+	cmp rg4, rg3
+	lli 0, rg5
+	jle _cmp_end_6
+	lli 1, rg5
+_cmp_end_6:
+	cmp rg5, zero
+	jeq _else_8
+_then_7:
+	lli 0, rg5
+	stw rg5, bpr, 8
+	jmp _ret
+	jmp _end_9
+_else_8:
+	nop
+_end_9:
+	lli 4, rg5
+	add rg0, rg5, rg6
+	stw rg4, rg6
+	stw rg2, bpr, 8
+	jmp _ret
+
+destroy:
+	push bpr
+	mov spr, bpr
+
+	ldw bpr, rg0, 8
+	lli 0, rg1
+	stw rg1, bpr, 8
+	jmp _ret
+
+reset_all:
+	push bpr
+	mov spr, bpr
+
+	ldw heap_start, rg0
+	stw rg0, heap_current
+	lli 0, rg0
+	stw rg0, bpr, 8
+	jmp _ret
+
+_ret:
+    mov bpr, spr
+    pop bpr
+    return

resources/dsa/lib/memory/arena_alloc.dsc

@@ -0,0 +1,77 @@
+// Arena Allocator
+// Supports multiple arenas that can be destroyed independently
+// Much more practical than a simple bump allocator
+
+// Global heap management
+static heap_start: u32 = 0x30000;
+static heap_end: u32 = 0x40000;
+static heap_current: u32 = 0x30000;
+
+// Arena structure (stored at the start of each arena):
+// [0-3]: start_address (u32)
+// [4-7]: current_position (u32)
+// [8-11]: end_address (u32)
+// Total header size: 12 bytes
+
+// Create a new arena with given size
+// Returns pointer to arena handle (or 0 if failed)
+fn arena_create(size: u32) -> u32 {
+    let total_size: u32 = size + 12;
+    let arena_ptr: u32 = heap_current;
+    let new_current: u32 = arena_ptr + total_size;
+
+    // Check if we have space
+    if new_current > heap_end {
+        return 0;
+    }
+
+    // Calculate arena data region
+    let data_start: u32 = arena_ptr + 12;
+    let data_end: u32 = arena_ptr + total_size;
+
+    // Initialize arena header
+    // Note: In real implementation, you'd use pointer writes here
+    // For now, using placeholder comments:
+    *arena_ptr = data_start;        // start_address
+    *(arena_ptr + 4) = data_start;  // current_position
+    *(arena_ptr + 8) = data_end;    // end_address
+
+    heap_current = new_current;
+
+    return arena_ptr;
+}
+
+// Allocate from an arena
+// Returns pointer to allocated memory (or 0 if failed)
+fn arena_alloc(arena: u32, size: u32) -> u32 {
+    // Read current position from arena
+    let current: u32 = *(arena + 4);
+    let end: u32 = *(arena + 8);
+
+    let new_current: u32 = current + size;
+
+    // Check if arena has space
+    if new_current > end {
+        return 0;
+    }
+
+    // Update current position in arena
+    *(arena + 4) = new_current;
+
+    return current;
+}
+
+// Destroy an arena (in bump allocator, this is a no-op)
+// In a real allocator, you'd mark the memory as free
+fn arena_destroy(arena: u32) {
+    // In a true allocator, mark memory as reusable
+    // For bump allocator, we can't reclaim memory
+    // unless we destroy ALL arenas and reset
+    return 0;
+}
+
+// Reset entire heap (destroys ALL arenas)
+fn reset_all() {
+    heap_current = heap_start;
+    return 0;
+}

resources/dsa/lib/memory/block_alloc.dsa

@@ -0,0 +1,216 @@
+// block_alloc.dsa
+// Fixed-size block allocator
+//
+// Memory layout:
+// [base + 0]: free list head pointer (pointer to first free block, or 0 if none)
+// [base + 4]: block size
+// [base + 8]: total blocks
+// [base + 12]: base address of block pool
+// [base + 16+]: block pool (each block starts with a 4-byte next pointer)
+//
+// Usage:
+// include block_alloc "./lib/memory/block_alloc.dsa"
+//
+// For init:
+//   push num_blocks (e.g., 32)
+//   push block_size (e.g., 64 bytes)
+//   call block_alloc::init
+//   pop block_size
+//   pop num_blocks
+//   ; result in spr+8 (allocator handle)
+//
+// For alloc:
+//   push allocator_handle
+//   call block_alloc::alloc
+//   pop allocator_handle
+//   ; result in spr+8 (pointer to block, or 0 if out of memory)
+//
+// For free:
+//   push block_pointer
+//   push allocator_handle
+//   call block_alloc::free
+//   pop allocator_handle
+//   pop block_pointer
+
+dw heap_start: 0x30000  // Start of our heap area
+
+// Initialize the allocator
+// Args: block_size, num_blocks
+// Returns: allocator handle (pointer to metadata)
+init:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8   // block_size
+    ldw bpr, rg1, 12  // num_blocks
+
+    // Allocate metadata (16 bytes) + pool space
+    ldw heap_start, rg2  // base address for this allocator
+    mov rg2, rg3         // save base in rg3
+
+    // Calculate total size needed: 16 + (block_size * num_blocks)
+    // We'll use a simple multiplication by repeated addition
+    mov rg0, rg4         // block_size to rg4
+    mov rg1, rg5         // num_blocks to rg5
+    lli 0, acc           // accumulator for multiplication
+
+_multiply_loop:
+    cmp rg5, zero
+    jeq _multiply_done
+    add acc, rg4, acc
+    dec rg5
+    jmp _multiply_loop
+
+_multiply_done:
+    // acc now contains block_size * num_blocks
+    addi acc, 16         // add metadata size
+
+    // Update heap_start for next allocation
+    add rg2, acc, acc
+    stw acc, heap_start
+
+    // Now set up metadata at rg3 (base)
+    // [base + 0]: free list head (will point to first block)
+    // [base + 4]: block_size
+    // [base + 8]: total blocks
+    // [base + 12]: pool base address
+
+    addi rg3, 16, rg6     // rg6 = pool base
+
+    stw rg6, rg3          // store pool base as free list head initially
+    stw rg0, rg3, 4       // store block_size
+    stw rg1, rg3, 8       // store total blocks
+    stw rg6, rg3, 12      // store pool base address
+
+    // Now initialize the free list
+    // Each block's first 4 bytes point to the next block
+    // rg6 = current block pointer
+    // rg0 = block_size
+    // rg1 = num_blocks (counter)
+
+    dec rg1              // we'll count down from num_blocks-1
+
+_init_loop:
+    cmp rg1, zero
+    jeq _init_loop_done
+
+    // Calculate next block address: current + block_size
+    add rg6, rg0, rg7    // rg7 = next block address
+
+    // Store next pointer at current block
+    stw rg7, rg6
+
+    // Move to next block
+    mov rg7, rg6
+    dec rg1
+    jmp _init_loop
+
+_init_loop_done:
+    // Last block points to null (0)
+    lli 0, acc
+    stw acc, rg6
+
+    // Return allocator handle (base address - 16 to get back to metadata start)
+    stw rg3, bpr, 8
+
+    mov bpr, spr
+    pop bpr
+    return
+
+
+// Allocate a block
+// Args: allocator_handle
+// Returns: pointer to block (or 0 if out of memory)
+alloc:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8      // allocator handle (metadata base)
+
+    // Load free list head
+    ldw rg0, rg1         // rg1 = free list head
+
+    // Check if free list is empty
+    cmp rg1, zero
+    jeq _alloc_fail
+
+    // Free list is not empty, pop the first block
+    // Load the next pointer from the block we're allocating
+    ldw rg1, rg2         // rg2 = next free block
+
+    // Update free list head to point to next block
+    stw rg2, rg0
+
+    // Return the allocated block (rg1)
+    stw rg1, bpr, 8
+    jmp _alloc_done
+
+_alloc_fail:
+    // No free blocks, return 0
+    lli 0, acc
+    stw acc, bpr, 8
+
+_alloc_done:
+    mov bpr, spr
+    pop bpr
+    return
+
+
+// Free a block
+// Args: allocator_handle, block_pointer
+// Returns: nothing (but could return error code if block is invalid)
+free:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8      // allocator handle
+    ldw bpr, rg6, 12     // pointer to the block pointer to free
+    ldw rg6, rg1         // rg1 = block pointer to free
+
+    // Load current free list head
+    ldw rg0, rg2         // rg2 = current head
+
+    // Set the freed block's next pointer to current head
+    stw rg2, rg1
+
+    // Update free list head to point to freed block
+    stw rg1, rg0
+
+    // Update the freed block's previous pointer to NULL
+    lli 0, rg1
+    stw rg1, rg6
+
+    mov bpr, spr
+    pop bpr
+    return
+
+
+// Debug function: get stats
+// Args: allocator_handle
+// Returns: nothing (but could populate a stats structure)
+get_stats:
+    push bpr
+    mov spr, bpr
+
+    ldw bpr, rg0, 8      // allocator handle
+
+    // Count free blocks by traversing the free list
+    ldw rg0, rg1         // rg1 = free list head
+    lli 0, rg2           // rg2 = counter
+
+count_loop:
+    cmp rg1, zero
+    jeq count_done
+
+    inc rg2
+    ldw rg1, rg1         // move to next block
+    jmp count_loop
+
+count_done:
+    // rg2 now contains number of free blocks
+    // Could store this somewhere or return it
+    stw rg2, bpr, 8
+
+    mov bpr, spr
+    pop bpr
+    return

resources/dsa/main.dsc

@@ -0,0 +1,37 @@
+include print: "./lib/io/print.dsa";
+include alloc: "./lib/memory/block_alloc.dsa";
+
+fn main() -> u32 {
+	let allocator: u32 = alloc::init(64, 32);
+	
+	print::print_hex_word(allocator);
+
+	print::print_newline();
+
+	let ptr: u32 = alloc::alloc(allocator);
+	print::print_hex_word(ptr);
+	*ptr = 200;
+
+	print::print_newline();
+
+	let p2: u32 = alloc::alloc(allocator);
+	print::print_hex_word(p2);
+	print::print_newline();
+	print::print_num(*ptr);
+
+	alloc::free(allocator, &ptr);
+	let ptr3: u32 = alloc::alloc(allocator);
+
+	print::print_newline();
+
+	print::print_hex_word(ptr3);
+
+	print::print_newline();
+	print::print_hex_word(ptr);
+
+	if ptr == 0 {
+		print::print("successful free of ptr");
+	}
+
+	return 0;
+}

resources/dsa/test.dsc

@@ -0,0 +1,20 @@
+include print: "./lib/io/print.dsa";
+
+fn main() -> u32 {
+	let x: u32 = 30;
+	print::print_num(x);
+
+	200 + 5;
+
+	let p: Point = Point {
+		x: 10,
+		y: 20,
+		test: [2, 3, 4]
+	};
+}
+
+struct Point {
+	x: u32,
+	y: u32,
+	test: [u32; 3],
+}

resources/dsa/testprint.dsa

@@ -1,4 +1,4 @@
-include print "./lib/io/print.dsa"
+include print: "./lib/io/print.dsa"
 
 dw idt: 0xFFFF0000
 dw stack: 0x10000

resources/examples/example.c

@@ -0,0 +1,11 @@
+int factorial(int n) {
+  if (n <= 1) {
+    return 1;
+  }
+  return n * factorial(n - 1);
+}
+
+int main() {
+  int res = factorial(3);
+  return res;
+}

src/main.rs

@@ -1,53 +0,0 @@
-use std::{
-    sync::{Arc, Mutex},
-    thread,
-};
-
-use dsa_rs::emulator::{
-    system::{emulator::run_emulator, memory::MainStore, processor::Processor},
-    ui::{
-        control_unit::ControlPanel, interface::EmulatorUI, memory_inspector::MemoryInspector,
-        stack_inspector::StackInspector,
-    },
-};
-
-fn main() -> Result<(), eframe::Error> {
-    // Initialize Channels
-    let (cmd_sender, cmd_receiver) = std::sync::mpsc::channel();
-    let (state_sender, state_receiver) = std::sync::mpsc::channel();
-
-    let mainstore = MainStore::new();
-    let processor = Processor::new(Box::new(mainstore), vec![]);
-
-    thread::spawn(move || {
-        run_emulator(&cmd_receiver, &state_sender, processor);
-    });
-
-    // Create UI
-    let mut ui = EmulatorUI::new(cmd_sender.clone(), state_receiver);
-
-    // Create UI modules
-    let control_unit = ControlPanel::new(cmd_sender.clone());
-    ui.add_component(Box::new(control_unit));
-
-    let mem_inspector = MemoryInspector::new(cmd_sender.clone());
-    ui.add_component(Box::new(mem_inspector));
-
-    let stack_inspector = StackInspector::new();
-    ui.add_component(Box::new(stack_inspector));
-
-    // Run UI
-    let options = eframe::NativeOptions {
-        viewport: egui::ViewportBuilder::default().with_inner_size([800.0, 600.0]),
-        ..Default::default()
-    };
-
-    eframe::run_native(
-        "DSA Simulator (Damn Simple Architecture 🔥)",
-        options,
-        Box::new(move |cc| {
-            cc.egui_ctx.set_visuals(egui::Visuals::default());
-            Ok(Box::new(ui))
-        }),
-    )
-}