WIP: core/ndstrides: implement np.newaxis and ... + IRRT_DEBUG_ASSERT

`cargo test -F test` would compile `nac3core/irrt/irrt_test.cpp`
targetted to the host machine (it gets to use `std`) and run the
test executable.

.clang-format

@@ -1,32 +1,3 @@
-BasedOnStyle: LLVM
-
-Language: Cpp
-Standard: Cpp11
-
-AccessModifierOffset: -1
-AlignEscapedNewlines: Left
-AlwaysBreakAfterReturnType: None
-AlwaysBreakTemplateDeclarations: Yes
-AllowAllParametersOfDeclarationOnNextLine: false
-AllowShortFunctionsOnASingleLine: Inline
-BinPackParameters: false
-BreakBeforeBinaryOperators: NonAssignment
-BreakBeforeTernaryOperators: true
-BreakConstructorInitializers: AfterColon
-BreakInheritanceList: AfterColon
-ColumnLimit: 120
-ConstructorInitializerAllOnOneLineOrOnePerLine: true
-ContinuationIndentWidth: 4
-DerivePointerAlignment: false
-IndentCaseLabels: true
-IndentPPDirectives: None
+BasedOnStyle: Google
 IndentWidth: 4
-MaxEmptyLinesToKeep: 1
-PointerAlignment: Left
-ReflowComments: true
-SortIncludes: false
-SortUsingDeclarations: true
-SpaceAfterTemplateKeyword: false
-SpacesBeforeTrailingComments: 2
-TabWidth: 4
-UseTab: Never
+ReflowComments: false

.pre-commit-config.yaml

@@ -1,24 +1,24 @@
 # See https://pre-commit.com for more information
 # See https://pre-commit.com/hooks.html for more hooks
 
-default_stages: [pre-commit]
+default_stages: [commit]
 
 repos:
   - repo: local
     hooks:
       - id: nac3-cargo-fmt
         name: nac3 cargo format
-        entry: nix
+        entry: cargo
         language: system
         types: [file, rust]
         pass_filenames: false
         description: Runs cargo fmt on the codebase.
-        args: [develop, -c, cargo, fmt, --all]
+        args: [fmt]
       - id: nac3-cargo-clippy
         name: nac3 cargo clippy
-        entry: nix
+        entry: cargo
         language: system
         types: [file, rust]
         pass_filenames: false
         description: Runs cargo clippy on the codebase.
-        args: [develop, -c, cargo, clippy, --tests]
+        args: [clippy, --tests]

Cargo.toml

@@ -4,7 +4,6 @@ members = [
   "nac3ast",
   "nac3parser",
   "nac3core",
-  "nac3core/nac3core_derive",
   "nac3standalone",
   "nac3artiq",
   "runkernel",

flake.lock

@@ -2,11 +2,11 @@
   "nodes": {
     "nixpkgs": {
       "locked": {
-        "lastModified": 1733940404,
-        "narHash": "sha256-Pj39hSoUA86ZePPF/UXiYHHM7hMIkios8TYG29kQT4g=",
+        "lastModified": 1721924956,
+        "narHash": "sha256-Sb1jlyRO+N8jBXEX9Pg9Z1Qb8Bw9QyOgLDNMEpmjZ2M=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "5d67ea6b4b63378b9c13be21e2ec9d1afc921713",
+        "rev": "5ad6a14c6bf098e98800b091668718c336effc95",
         "type": "github"
       },
       "original": {

flake.nix

@@ -14,6 +14,7 @@
           ''
           mkdir -p $out/bin
           ln -s ${pkgs.llvmPackages_14.clang-unwrapped}/bin/clang $out/bin/clang-irrt
+          ln -s ${pkgs.llvmPackages_14.clang}/bin/clang $out/bin/clang-irrt-test
           ln -s ${pkgs.llvmPackages_14.llvm.out}/bin/llvm-as $out/bin/llvm-as-irrt
           '';
         demo-linalg-stub = pkgs.rustPlatform.buildRustPackage {
@@ -40,6 +41,7 @@
             cargoLock = {
               lockFile = ./Cargo.lock;
             };
+            cargoTestFlags = [ "--features" "test" ];
             passthru.cargoLock = cargoLock;
             nativeBuildInputs = [ pkgs.python3 (pkgs.wrapClangMulti pkgs.llvmPackages_14.clang) llvm-tools-irrt pkgs.llvmPackages_14.llvm.out llvm-nac3 ];
             buildInputs = [ pkgs.python3 llvm-nac3 ];
@@ -107,18 +109,18 @@
             (pkgs.fetchFromGitHub {
               owner = "m-labs";
               repo = "sipyco";
-              rev = "094a6cd63ffa980ef63698920170e50dc9ba77fd";
-              sha256 = "sha256-PPnAyDedUQ7Og/Cby9x5OT9wMkNGTP8GS53V6N/dk4w=";
+              rev = "939f84f9b5eef7efbf7423c735d1834783b6140e";
+              sha256 = "sha256-15Nun4EY35j+6SPZkjzZtyH/ncxLS60KuGJjFh5kSTc=";
             })
             (pkgs.fetchFromGitHub {
               owner = "m-labs";
               repo = "artiq";
-              rev = "28c9de3e251daa89a8c9fd79d5ab64a3ec03bac6";
-              sha256 = "sha256-vAvpbHc5B+1wtG8zqN7j9dQE1ON+i22v+uqA+tw6Gak=";
+              rev = "923ca3377d42c815f979983134ec549dc39d3ca0";
+              sha256 = "sha256-oJoEeNEeNFSUyh6jXG8Tzp6qHVikeHS0CzfE+mODPgw=";
             })
           ];
           buildInputs = [
-            (python3-mimalloc.withPackages(ps: [ ps.numpy ps.scipy ps.jsonschema ps.lmdb ps.platformdirs nac3artiq-instrumented ]))
+            (python3-mimalloc.withPackages(ps: [ ps.numpy ps.scipy ps.jsonschema ps.lmdb nac3artiq-instrumented ]))
             pkgs.llvmPackages_14.llvm.out
           ];
           phases = [ "buildPhase" "installPhase" ];

nac3artiq/Cargo.toml

@@ -12,10 +12,15 @@ crate-type = ["cdylib"]
 itertools = "0.13"
 pyo3 = { version = "0.21", features = ["extension-module", "gil-refs"] }
 parking_lot = "0.12"
-tempfile = "3.13"
+tempfile = "3.10"
+nac3parser = { path = "../nac3parser" }
 nac3core = { path = "../nac3core" }
 nac3ld = { path = "../nac3ld" }
 
+[dependencies.inkwell]
+version = "0.4"
+default-features = false
+features = ["llvm14-0", "target-x86", "target-arm", "target-riscv", "no-libffi-linking"]
+
 [features]
 init-llvm-profile = []
-no-escape-analysis = ["nac3core/no-escape-analysis"]

nac3artiq/demo/embedding_map.py

@@ -7,6 +7,33 @@ class EmbeddingMap:
         self.function_map = {}
         self.attributes_writeback = []
 
+        # preallocate exception names
+        self.preallocate_runtime_exception_names(["RuntimeError",
+                                          "RTIOUnderflow",
+                                          "RTIOOverflow",
+                                          "RTIODestinationUnreachable",
+                                          "DMAError",
+                                          "I2CError",
+                                          "CacheError",
+                                          "SPIError",
+                                          "0:ZeroDivisionError",
+                                          "0:IndexError",
+                                          "0:ValueError",
+                                          "0:RuntimeError",
+                                          "0:AssertionError",
+                                          "0:KeyError",
+                                          "0:NotImplementedError",
+                                          "0:OverflowError",
+                                          "0:IOError",
+                                          "0:UnwrapNoneError"])
+
+    def preallocate_runtime_exception_names(self, names):
+        for i, name in enumerate(names):
+            if ":" not in name:
+                name = "0:artiq.coredevice.exceptions." + name
+            exn_id = self.store_str(name)
+            assert exn_id == i
+
     def store_function(self, key, fun):
         self.function_map[key] = fun
         return key

nac3artiq/demo/min_artiq.py

@@ -112,15 +112,10 @@ def extern(function):
     register_function(function)
     return function
 
-
-def rpc(arg=None, flags={}):
-    """Decorates a function or method to be executed on the host interpreter."""
-    if arg is None:
-        def inner_decorator(function):
-            return rpc(function, flags)
-        return inner_decorator
-    register_function(arg)
-    return arg
+def rpc(function):
+    """Decorates a function declaration defined by the core device runtime."""
+    register_function(function)
+    return function
 
 def kernel(function_or_method):
     """Decorates a function or method to be executed on the core device."""
@@ -206,7 +201,7 @@ class Core:
         embedding = EmbeddingMap()
 
         if allow_registration:
-            compiler.analyze(registered_functions, registered_classes, set())
+            compiler.analyze(registered_functions, registered_classes)
             allow_registration = False
 
         if hasattr(method, "__self__"):

nac3artiq/demo/str_abi.py

@@ -1,26 +0,0 @@
-from min_artiq import *
-from numpy import ndarray, zeros as np_zeros
-
-
-@nac3
-class StrFail:
-    core: KernelInvariant[Core]
-
-    def __init__(self):
-        self.core = Core()
-
-    @kernel
-    def hello(self, arg: str):
-        pass
-
-    @kernel
-    def consume_ndarray(self, arg: ndarray[str, 1]):
-        pass
-
-    def run(self):
-        self.hello("world")
-        self.consume_ndarray(np_zeros([10], dtype=str))
-
-
-if __name__ == "__main__":
-    StrFail().run()

nac3artiq/src/lib.rs

@@ -1,4 +1,10 @@
-#![deny(future_incompatible, let_underscore, nonstandard_style, clippy::all)]
+#![deny(
+    future_incompatible,
+    let_underscore,
+    nonstandard_style,
+    rust_2024_compatibility,
+    clippy::all
+)]
 #![warn(clippy::pedantic)]
 #![allow(
     unsafe_op_in_unsafe_fn,
@@ -10,65 +16,65 @@
     clippy::wildcard_imports
 )]
 
-use std::{
-    collections::{HashMap, HashSet},
-    fs,
-    io::Write,
-    process::Command,
-    rc::Rc,
-    sync::Arc,
-};
+use std::collections::{HashMap, HashSet};
+use std::fs;
+use std::io::Write;
+use std::process::Command;
+use std::rc::Rc;
+use std::sync::Arc;
 
-use itertools::Itertools;
-use parking_lot::{Mutex, RwLock};
-use pyo3::{
-    create_exception, exceptions,
-    prelude::*,
-    types::{PyBytes, PyDict, PyNone, PySet},
+use inkwell::{
+    context::Context,
+    memory_buffer::MemoryBuffer,
+    module::{Linkage, Module},
+    passes::PassBuilderOptions,
+    support::is_multithreaded,
+    targets::*,
+    OptimizationLevel,
 };
-use tempfile::{self, TempDir};
+use itertools::Itertools;
+use nac3core::codegen::irrt::setup_irrt_exceptions;
+use nac3core::codegen::{gen_func_impl, CodeGenLLVMOptions, CodeGenTargetMachineOptions};
+use nac3core::toplevel::builtins::get_exn_constructor;
+use nac3core::typecheck::typedef::{TypeEnum, Unifier, VarMap};
+use nac3parser::{
+    ast::{ExprKind, Stmt, StmtKind, StrRef},
+    parser::parse_program,
+};
+use pyo3::create_exception;
+use pyo3::prelude::*;
+use pyo3::{exceptions, types::PyBytes, types::PyDict, types::PySet};
+
+use parking_lot::{Mutex, RwLock};
 
 use nac3core::{
-    codegen::{
-        concrete_type::ConcreteTypeStore, gen_func_impl, irrt::load_irrt, CodeGenLLVMOptions,
-        CodeGenTargetMachineOptions, CodeGenTask, CodeGenerator, WithCall, WorkerRegistry,
-    },
-    inkwell::{
-        context::Context,
-        memory_buffer::MemoryBuffer,
-        module::{FlagBehavior, Linkage, Module},
-        passes::PassBuilderOptions,
-        support::is_multithreaded,
-        targets::*,
-        OptimizationLevel,
-    },
-    nac3parser::{
-        ast::{Constant, ExprKind, Located, Stmt, StmtKind, StrRef},
-        parser::parse_program,
-    },
+    codegen::irrt::load_irrt,
+    codegen::{concrete_type::ConcreteTypeStore, CodeGenTask, WithCall, WorkerRegistry},
     symbol_resolver::SymbolResolver,
     toplevel::{
-        builtins::get_exn_constructor,
-        composer::{BuiltinFuncCreator, BuiltinFuncSpec, ComposerConfig, TopLevelComposer},
+        composer::{ComposerConfig, TopLevelComposer},
         DefinitionId, GenCall, TopLevelDef,
     },
-    typecheck::{
-        type_inferencer::PrimitiveStore,
-        typedef::{into_var_map, FunSignature, FuncArg, Type, TypeEnum, Unifier, VarMap},
-    },
+    typecheck::typedef::{FunSignature, FuncArg},
+    typecheck::{type_inferencer::PrimitiveStore, typedef::Type},
 };
+
 use nac3ld::Linker;
 
-use codegen::{
-    attributes_writeback, gen_core_log, gen_rtio_log, rpc_codegen_callback, ArtiqCodeGenerator,
+use tempfile::{self, TempDir};
+
+use crate::codegen::attributes_writeback;
+use crate::{
+    codegen::{rpc_codegen_callback, ArtiqCodeGenerator},
+    symbol_resolver::{DeferredEvaluationStore, InnerResolver, PythonHelper, Resolver},
 };
-use symbol_resolver::{DeferredEvaluationStore, InnerResolver, PythonHelper, Resolver};
-use timeline::TimeFns;
 
 mod codegen;
 mod symbol_resolver;
 mod timeline;
 
+use timeline::TimeFns;
+
 #[derive(PartialEq, Clone, Copy)]
 enum Isa {
     Host,
@@ -122,7 +128,7 @@ struct Nac3 {
     isa: Isa,
     time_fns: &'static (dyn TimeFns + Sync),
     primitive: PrimitiveStore,
-    builtins: Vec<BuiltinFuncSpec>,
+    builtins: Vec<(StrRef, FunSignature, Arc<GenCall>)>,
     pyid_to_def: Arc<RwLock<HashMap<u64, DefinitionId>>>,
     primitive_ids: PrimitivePythonId,
     working_directory: TempDir,
@@ -142,32 +148,14 @@ impl Nac3 {
         module: &PyObject,
         registered_class_ids: &HashSet<u64>,
     ) -> PyResult<()> {
-        let (module_name, source_file, source) =
-            Python::with_gil(|py| -> PyResult<(String, String, String)> {
-                let module: &PyAny = module.extract(py)?;
-                let source_file = module.getattr("__file__");
-                let (source_file, source) = if let Ok(source_file) = source_file {
-                    let source_file = source_file.extract()?;
-                    (
-                        source_file,
-                        fs::read_to_string(source_file).map_err(|e| {
-                            exceptions::PyIOError::new_err(format!(
-                                "failed to read input file: {e}"
-                            ))
-                        })?,
-                    )
-                } else {
-                    // kernels submitted by content have no file
-                    // but still can provide source by StringLoader
-                    let get_src_fn = module
-                        .getattr("__loader__")?
-                        .extract::<PyObject>()?
-                        .getattr(py, "get_source")?;
-                    ("<expcontent>", get_src_fn.call1(py, (PyNone::get(py),))?.extract(py)?)
-                };
-                Ok((module.getattr("__name__")?.extract()?, source_file.to_string(), source))
-            })?;
+        let (module_name, source_file) = Python::with_gil(|py| -> PyResult<(String, String)> {
+            let module: &PyAny = module.extract(py)?;
+            Ok((module.getattr("__name__")?.extract()?, module.getattr("__file__")?.extract()?))
+        })?;
 
+        let source = fs::read_to_string(&source_file).map_err(|e| {
+            exceptions::PyIOError::new_err(format!("failed to read input file: {e}"))
+        })?;
         let parser_result = parse_program(&source, source_file.into())
             .map_err(|e| exceptions::PySyntaxError::new_err(format!("parse error: {e}")))?;
 
@@ -207,8 +195,10 @@ impl Nac3 {
                     body.retain(|stmt| {
                         if let StmtKind::FunctionDef { ref decorator_list, .. } = stmt.node {
                             decorator_list.iter().any(|decorator| {
-                                if let Some(id) = decorator_id_string(decorator) {
-                                    id == "kernel" || id == "portable" || id == "rpc"
+                                if let ExprKind::Name { id, .. } = decorator.node {
+                                    id.to_string() == "kernel"
+                                        || id.to_string() == "portable"
+                                        || id.to_string() == "rpc"
                                 } else {
                                     false
                                 }
@@ -221,8 +211,9 @@ impl Nac3 {
                 }
                 StmtKind::FunctionDef { ref decorator_list, .. } => {
                     decorator_list.iter().any(|decorator| {
-                        if let Some(id) = decorator_id_string(decorator) {
-                            id == "extern" || id == "kernel" || id == "portable" || id == "rpc"
+                        if let ExprKind::Name { id, .. } = decorator.node {
+                            let id = id.to_string();
+                            id == "extern" || id == "portable" || id == "kernel" || id == "rpc"
                         } else {
                             false
                         }
@@ -311,64 +302,6 @@ impl Nac3 {
         None
     }
 
-    /// Returns a [`Vec`] of builtins that needs to be initialized during method compilation time.
-    fn get_lateinit_builtins() -> Vec<Box<BuiltinFuncCreator>> {
-        vec![
-            Box::new(|primitives, unifier| {
-                let arg_ty = unifier.get_fresh_var(Some("T".into()), None);
-
-                (
-                    "core_log".into(),
-                    FunSignature {
-                        args: vec![FuncArg {
-                            name: "arg".into(),
-                            ty: arg_ty.ty,
-                            default_value: None,
-                            is_vararg: false,
-                        }],
-                        ret: primitives.none,
-                        vars: into_var_map([arg_ty]),
-                    },
-                    Arc::new(GenCall::new(Box::new(move |ctx, obj, fun, args, generator| {
-                        gen_core_log(ctx, &obj, fun, &args, generator)?;
-
-                        Ok(None)
-                    }))),
-                )
-            }),
-            Box::new(|primitives, unifier| {
-                let arg_ty = unifier.get_fresh_var(Some("T".into()), None);
-
-                (
-                    "rtio_log".into(),
-                    FunSignature {
-                        args: vec![
-                            FuncArg {
-                                name: "channel".into(),
-                                ty: primitives.str,
-                                default_value: None,
-                                is_vararg: false,
-                            },
-                            FuncArg {
-                                name: "arg".into(),
-                                ty: arg_ty.ty,
-                                default_value: None,
-                                is_vararg: false,
-                            },
-                        ],
-                        ret: primitives.none,
-                        vars: into_var_map([arg_ty]),
-                    },
-                    Arc::new(GenCall::new(Box::new(move |ctx, obj, fun, args, generator| {
-                        gen_rtio_log(ctx, &obj, fun, &args, generator)?;
-
-                        Ok(None)
-                    }))),
-                )
-            }),
-        ]
-    }
-
     fn compile_method<T>(
         &self,
         obj: &PyAny,
@@ -381,7 +314,6 @@ impl Nac3 {
         let size_t = self.isa.get_size_type();
         let (mut composer, mut builtins_def, mut builtins_ty) = TopLevelComposer::new(
             self.builtins.clone(),
-            Self::get_lateinit_builtins(),
             ComposerConfig { kernel_ann: Some("Kernel"), kernel_invariant_ann: "KernelInvariant" },
             size_t,
         );
@@ -458,6 +390,7 @@ impl Nac3 {
                         pyid_to_type: pyid_to_type.clone(),
                         primitive_ids: self.primitive_ids.clone(),
                         global_value_ids: global_value_ids.clone(),
+                        class_names: Mutex::default(),
                         name_to_pyid: name_to_pyid.clone(),
                         module: module.clone(),
                         id_to_pyval: RwLock::default(),
@@ -488,25 +421,9 @@ impl Nac3 {
 
             match &stmt.node {
                 StmtKind::FunctionDef { decorator_list, .. } => {
-                    if decorator_list
-                        .iter()
-                        .any(|decorator| decorator_id_string(decorator) == Some("rpc".to_string()))
-                    {
-                        store_fun
-                            .call1(
-                                py,
-                                (
-                                    def_id.0.into_py(py),
-                                    module.getattr(py, name.to_string().as_str()).unwrap(),
-                                ),
-                            )
-                            .unwrap();
-                        let is_async = decorator_list.iter().any(|decorator| {
-                            decorator_get_flags(decorator)
-                                .iter()
-                                .any(|constant| *constant == Constant::Str("async".into()))
-                        });
-                        rpc_ids.push((None, def_id, is_async));
+                    if decorator_list.iter().any(|decorator| matches!(decorator.node, ExprKind::Name { id, .. } if id == "rpc".into())) {
+                        store_fun.call1(py, (def_id.0.into_py(py), module.getattr(py, name.to_string().as_str()).unwrap())).unwrap();
+                        rpc_ids.push((None, def_id));
                     }
                 }
                 StmtKind::ClassDef { name, body, .. } => {
@@ -514,26 +431,19 @@ impl Nac3 {
                     let class_obj = module.getattr(py, class_name.as_str()).unwrap();
                     for stmt in body {
                         if let StmtKind::FunctionDef { name, decorator_list, .. } = &stmt.node {
-                            if decorator_list.iter().any(|decorator| {
-                                decorator_id_string(decorator) == Some("rpc".to_string())
-                            }) {
-                                let is_async = decorator_list.iter().any(|decorator| {
-                                    decorator_get_flags(decorator)
-                                        .iter()
-                                        .any(|constant| *constant == Constant::Str("async".into()))
-                                });
+                            if decorator_list.iter().any(|decorator| matches!(decorator.node, ExprKind::Name { id, .. } if id == "rpc".into())) {
                                 if name == &"__init__".into() {
                                     return Err(CompileError::new_err(format!(
                                         "compilation failed\n----------\nThe constructor of class {} should not be decorated with rpc decorator (at {})",
                                         class_name, stmt.location
                                     )));
                                 }
-                                rpc_ids.push((Some((class_obj.clone(), *name)), def_id, is_async));
+                                rpc_ids.push((Some((class_obj.clone(), *name)), def_id));
                             }
                         }
                     }
                 }
-                _ => (),
+                _ => ()
             }
 
             let id = *name_to_pyid.get(&name).unwrap();
@@ -572,12 +482,13 @@ impl Nac3 {
             pyid_to_type: pyid_to_type.clone(),
             primitive_ids: self.primitive_ids.clone(),
             global_value_ids: global_value_ids.clone(),
+            class_names: Mutex::default(),
             id_to_pyval: RwLock::default(),
             id_to_primitive: RwLock::default(),
             field_to_val: RwLock::default(),
             name_to_pyid,
             module: module.to_object(py),
-            helper: helper.clone(),
+            helper,
             string_store: self.string_store.clone(),
             exception_ids: self.exception_ids.clone(),
             deferred_eval_store: self.deferred_eval_store.clone(),
@@ -589,8 +500,9 @@ impl Nac3 {
             .unwrap();
 
         // Process IRRT
-        let context = Context::create();
-        let irrt = load_irrt(&context, resolver.as_ref());
+        let context = inkwell::context::Context::create();
+        let irrt = load_irrt(&context);
+        setup_irrt_exceptions(&context, &irrt, resolver.as_ref());
 
         let fun_signature =
             FunSignature { args: vec![], ret: self.primitive.none, vars: VarMap::new() };
@@ -629,12 +541,13 @@ impl Nac3 {
         let top_level = Arc::new(composer.make_top_level_context());
 
         {
+            let rpc_codegen = rpc_codegen_callback();
             let defs = top_level.definitions.read();
-            for (class_data, id, is_async) in &rpc_ids {
+            for (class_data, id) in &rpc_ids {
                 let mut def = defs[id.0].write();
                 match &mut *def {
                     TopLevelDef::Function { codegen_callback, .. } => {
-                        *codegen_callback = Some(rpc_codegen_callback(*is_async));
+                        *codegen_callback = Some(rpc_codegen.clone());
                     }
                     TopLevelDef::Class { methods, .. } => {
                         let (class_def, method_name) = class_data.as_ref().unwrap();
@@ -645,7 +558,7 @@ impl Nac3 {
                             if let TopLevelDef::Function { codegen_callback, .. } =
                                 &mut *defs[id.0].write()
                             {
-                                *codegen_callback = Some(rpc_codegen_callback(*is_async));
+                                *codegen_callback = Some(rpc_codegen.clone());
                                 store_fun
                                     .call1(
                                         py,
@@ -660,11 +573,6 @@ impl Nac3 {
                             }
                         }
                     }
-                    TopLevelDef::Variable { .. } => {
-                        return Err(CompileError::new_err(String::from(
-                            "Unsupported @rpc annotation on global variable",
-                        )))
-                    }
                 }
             }
         }
@@ -685,12 +593,33 @@ impl Nac3 {
         let task = CodeGenTask {
             subst: Vec::default(),
             symbol_name: "__modinit__".to_string(),
+            body: instance.body,
+            signature,
+            resolver: resolver.clone(),
+            store,
+            unifier_index: instance.unifier_id,
+            calls: instance.calls,
+            id: 0,
+        };
+
+        let mut store = ConcreteTypeStore::new();
+        let mut cache = HashMap::new();
+        let signature = store.from_signature(
+            &mut composer.unifier,
+            &self.primitive,
+            &fun_signature,
+            &mut cache,
+        );
+        let signature = store.add_cty(signature);
+        let attributes_writeback_task = CodeGenTask {
+            subst: Vec::default(),
+            symbol_name: "attributes_writeback".to_string(),
             body: Arc::new(Vec::default()),
             signature,
             resolver,
             store,
             unifier_index: instance.unifier_id,
-            calls: instance.calls,
+            calls: Arc::new(HashMap::default()),
             id: 0,
         };
 
@@ -703,7 +632,7 @@ impl Nac3 {
             let buffer = buffer.as_slice().into();
             membuffer.lock().push(buffer);
         })));
-        let size_t = context
+        let size_t = Context::create()
             .ptr_sized_int_type(&self.get_llvm_target_machine().get_target_data(), None)
             .get_bit_width();
         let num_threads = if is_multithreaded() { 4 } else { 1 };
@@ -714,27 +643,19 @@ impl Nac3 {
             .collect();
 
         let membuffer = membuffers.clone();
-        let mut has_return = false;
         py.allow_threads(|| {
             let (registry, handles) =
                 WorkerRegistry::create_workers(threads, top_level.clone(), &self.llvm_options, &f);
+            registry.add_task(task);
+            registry.wait_tasks_complete(handles);
 
-            let mut generator = ArtiqCodeGenerator::new("main".to_string(), size_t, self.time_fns);
-            let context = Context::create();
-            let module = context.create_module("main");
+            let mut generator =
+                ArtiqCodeGenerator::new("attributes_writeback".to_string(), size_t, self.time_fns);
+            let context = inkwell::context::Context::create();
+            let module = context.create_module("attributes_writeback");
             let target_machine = self.llvm_options.create_target_machine().unwrap();
             module.set_data_layout(&target_machine.get_target_data().get_data_layout());
             module.set_triple(&target_machine.get_triple());
-            module.add_basic_value_flag(
-                "Debug Info Version",
-                FlagBehavior::Warning,
-                context.i32_type().const_int(3, false),
-            );
-            module.add_basic_value_flag(
-                "Dwarf Version",
-                FlagBehavior::Warning,
-                context.i32_type().const_int(4, false),
-            );
             let builder = context.create_builder();
             let (_, module, _) = gen_func_impl(
                 &context,
@@ -742,27 +663,9 @@ impl Nac3 {
                 &registry,
                 builder,
                 module,
-                task,
+                attributes_writeback_task,
                 |generator, ctx| {
-                    assert_eq!(instance.body.len(), 1, "toplevel module should have 1 statement");
-                    let StmtKind::Expr { value: ref expr, .. } = instance.body[0].node else {
-                        unreachable!("toplevel statement must be an expression")
-                    };
-                    let ExprKind::Call { .. } = expr.node else {
-                        unreachable!("toplevel expression must be a function call")
-                    };
-
-                    let return_obj =
-                        generator.gen_expr(ctx, expr)?.map(|value| (expr.custom.unwrap(), value));
-                    has_return = return_obj.is_some();
-                    registry.wait_tasks_complete(handles);
-                    attributes_writeback(
-                        ctx,
-                        generator,
-                        inner_resolver.as_ref(),
-                        &host_attributes,
-                        return_obj,
-                    )
+                    attributes_writeback(ctx, generator, inner_resolver.as_ref(), &host_attributes)
                 },
             )
             .unwrap();
@@ -771,23 +674,35 @@ impl Nac3 {
             membuffer.lock().push(buffer);
         });
 
-        embedding_map.setattr("expects_return", has_return).unwrap();
-
         // Link all modules into `main`.
         let buffers = membuffers.lock();
         let main = context
-            .create_module_from_ir(MemoryBuffer::create_from_memory_range(
-                buffers.last().unwrap(),
-                "main",
-            ))
+            .create_module_from_ir(MemoryBuffer::create_from_memory_range(&buffers[0], "main"))
             .unwrap();
-        for buffer in buffers.iter().rev().skip(1) {
+        for buffer in buffers.iter().skip(1) {
             let other = context
                 .create_module_from_ir(MemoryBuffer::create_from_memory_range(buffer, "main"))
                 .unwrap();
 
             main.link_in_module(other).map_err(|err| CompileError::new_err(err.to_string()))?;
         }
+        let builder = context.create_builder();
+        let modinit_return = main
+            .get_function("__modinit__")
+            .unwrap()
+            .get_last_basic_block()
+            .unwrap()
+            .get_terminator()
+            .unwrap();
+        builder.position_before(&modinit_return);
+        builder
+            .build_call(
+                main.get_function("attributes_writeback").unwrap(),
+                &[],
+                "attributes_writeback",
+            )
+            .unwrap();
+
         main.link_in_module(irrt).map_err(|err| CompileError::new_err(err.to_string()))?;
 
         let mut function_iter = main.get_first_function();
@@ -822,20 +737,6 @@ impl Nac3 {
             panic!("Failed to run optimization for module `main`: {}", err.to_string());
         }
 
-        Python::with_gil(|py| {
-            let string_store = self.string_store.read();
-            let mut string_store_vec = string_store.iter().collect::<Vec<_>>();
-            string_store_vec.sort_by(|(_s1, key1), (_s2, key2)| key1.cmp(key2));
-            for (s, key) in string_store_vec {
-                let embed_key: i32 = helper.store_str.call1(py, (s,)).unwrap().extract(py).unwrap();
-                assert_eq!(
-                    embed_key, *key,
-                    "string {s} is out of sync between embedding map (key={embed_key}) and \
-                    the internal string store (key={key})"
-                );
-            }
-        });
-
         link_fn(&main)
     }
 
@@ -888,41 +789,6 @@ impl Nac3 {
     }
 }
 
-/// Retrieves the Name.id from a decorator, supports decorators with arguments.
-fn decorator_id_string(decorator: &Located<ExprKind>) -> Option<String> {
-    if let ExprKind::Name { id, .. } = decorator.node {
-        // Bare decorator
-        return Some(id.to_string());
-    } else if let ExprKind::Call { func, .. } = &decorator.node {
-        // Decorators that are calls (e.g. "@rpc()") have Call for the node,
-        // need to extract the id from within.
-        if let ExprKind::Name { id, .. } = func.node {
-            return Some(id.to_string());
-        }
-    }
-    None
-}
-
-/// Retrieves flags from a decorator, if any.
-fn decorator_get_flags(decorator: &Located<ExprKind>) -> Vec<Constant> {
-    let mut flags = vec![];
-    if let ExprKind::Call { keywords, .. } = &decorator.node {
-        for keyword in keywords {
-            if keyword.node.arg != Some("flags".into()) {
-                continue;
-            }
-            if let ExprKind::Set { elts } = &keyword.node.value.node {
-                for elt in elts {
-                    if let ExprKind::Constant { value, .. } = &elt.node {
-                        flags.push(value.clone());
-                    }
-                }
-            }
-        }
-    }
-    flags
-}
-
 fn link_with_lld(elf_filename: String, obj_filename: String) -> PyResult<()> {
     let linker_args = vec![
         "-shared".to_string(),
@@ -992,7 +858,7 @@ impl Nac3 {
             Isa::RiscV32IMA => &timeline::NOW_PINNING_TIME_FNS,
             Isa::CortexA9 | Isa::Host => &timeline::EXTERN_TIME_FNS,
         };
-        let (primitive, _) = TopLevelComposer::make_primitives(isa.get_size_type());
+        let primitive: PrimitiveStore = TopLevelComposer::make_primitives(isa.get_size_type()).0;
         let builtins = vec![
             (
                 "now_mu".into(),
@@ -1085,48 +951,6 @@ impl Nac3 {
         let working_directory = tempfile::Builder::new().prefix("nac3-").tempdir().unwrap();
         fs::write(working_directory.path().join("kernel.ld"), include_bytes!("kernel.ld")).unwrap();
 
-        let mut string_store: HashMap<String, i32> = HashMap::default();
-
-        // Keep this list of exceptions in sync with `EXCEPTION_ID_LOOKUP` in `artiq::firmware::ksupport::eh_artiq`
-        // The exceptions declared here must be defined in `artiq.coredevice.exceptions`
-        // Verify synchronization by running the test cases in `artiq.test.coredevice.test_exceptions`
-        let runtime_exception_names = [
-            "RTIOUnderflow",
-            "RTIOOverflow",
-            "RTIODestinationUnreachable",
-            "DMAError",
-            "I2CError",
-            "CacheError",
-            "SPIError",
-            "SubkernelError",
-            "0:AssertionError",
-            "0:AttributeError",
-            "0:IndexError",
-            "0:IOError",
-            "0:KeyError",
-            "0:NotImplementedError",
-            "0:OverflowError",
-            "0:RuntimeError",
-            "0:TimeoutError",
-            "0:TypeError",
-            "0:ValueError",
-            "0:ZeroDivisionError",
-            "0:LinAlgError",
-            "UnwrapNoneError",
-        ];
-
-        // Preallocate runtime exception names
-        for (i, name) in runtime_exception_names.iter().enumerate() {
-            let exn_name = if name.find(':').is_none() {
-                format!("0:artiq.coredevice.exceptions.{name}")
-            } else {
-                (*name).to_string()
-            };
-
-            let id = i32::try_from(i).unwrap();
-            string_store.insert(exn_name, id);
-        }
-
         Ok(Nac3 {
             isa,
             time_fns,
@@ -1136,7 +960,7 @@ impl Nac3 {
             top_levels: Vec::default(),
             pyid_to_def: Arc::default(),
             working_directory,
-            string_store: Arc::new(string_store.into()),
+            string_store: Arc::default(),
             exception_ids: Arc::default(),
             deferred_eval_store: DeferredEvaluationStore::new(),
             llvm_options: CodeGenLLVMOptions {
@@ -1146,12 +970,7 @@ impl Nac3 {
         })
     }
 
-    fn analyze(
-        &mut self,
-        functions: &PySet,
-        classes: &PySet,
-        content_modules: &PySet,
-    ) -> PyResult<()> {
+    fn analyze(&mut self, functions: &PySet, classes: &PySet) -> PyResult<()> {
         let (modules, class_ids) =
             Python::with_gil(|py| -> PyResult<(HashMap<u64, PyObject>, HashSet<u64>)> {
                 let mut modules: HashMap<u64, PyObject> = HashMap::new();
@@ -1161,21 +980,13 @@ impl Nac3 {
                 let getmodule_fn = PyModule::import(py, "inspect")?.getattr("getmodule")?;
 
                 for function in functions {
-                    let module: PyObject = getmodule_fn.call1((function,))?.extract()?;
-                    if !module.is_none(py) {
-                        modules.insert(id_fn.call1((&module,))?.extract()?, module);
-                    }
+                    let module = getmodule_fn.call1((function,))?.extract()?;
+                    modules.insert(id_fn.call1((&module,))?.extract()?, module);
                 }
                 for class in classes {
-                    let module: PyObject = getmodule_fn.call1((class,))?.extract()?;
-                    if !module.is_none(py) {
-                        modules.insert(id_fn.call1((&module,))?.extract()?, module);
-                    }
-                    class_ids.insert(id_fn.call1((class,))?.extract()?);
-                }
-                for module in content_modules {
-                    let module: PyObject = module.extract()?;
+                    let module = getmodule_fn.call1((class,))?.extract()?;
                     modules.insert(id_fn.call1((&module,))?.extract()?, module);
+                    class_ids.insert(id_fn.call1((class,))?.extract()?);
                 }
                 Ok((modules, class_ids))
             })?;

nac3artiq/src/symbol_resolver.rs

@@ -1,32 +1,14 @@
-use std::{
-    collections::{HashMap, HashSet},
-    sync::{
-        atomic::{AtomicBool, Ordering::Relaxed},
-        Arc,
-    },
+use inkwell::{
+    types::{BasicType, BasicTypeEnum},
+    values::BasicValueEnum,
+    AddressSpace,
 };
-
 use itertools::Itertools;
-use parking_lot::RwLock;
-use pyo3::{
-    types::{PyDict, PyTuple},
-    PyAny, PyErr, PyObject, PyResult, Python,
-};
-
-use super::PrimitivePythonId;
 use nac3core::{
     codegen::{
-        types::{ndarray::NDArrayType, ProxyType},
-        values::ndarray::make_contiguous_strides,
+        classes::{NDArrayType, ProxyType},
         CodeGenContext, CodeGenerator,
     },
-    inkwell::{
-        module::Linkage,
-        types::{BasicType, BasicTypeEnum},
-        values::BasicValueEnum,
-        AddressSpace,
-    },
-    nac3parser::ast::{self, StrRef},
     symbol_resolver::{StaticValue, SymbolResolver, SymbolValue, ValueEnum},
     toplevel::{
         helper::PrimDef,
@@ -38,6 +20,21 @@ use nac3core::{
         typedef::{into_var_map, iter_type_vars, Type, TypeEnum, TypeVar, Unifier, VarMap},
     },
 };
+use nac3parser::ast::{self, StrRef};
+use parking_lot::{Mutex, RwLock};
+use pyo3::{
+    types::{PyDict, PyTuple},
+    PyAny, PyObject, PyResult, Python,
+};
+use std::{
+    collections::{HashMap, HashSet},
+    sync::{
+        atomic::{AtomicBool, Ordering::Relaxed},
+        Arc,
+    },
+};
+
+use crate::PrimitivePythonId;
 
 pub enum PrimitiveValue {
     I32(i32),
@@ -82,6 +79,7 @@ pub struct InnerResolver {
     pub id_to_primitive: RwLock<HashMap<u64, PrimitiveValue>>,
     pub field_to_val: RwLock<HashMap<ResolverField, Option<PyFieldHandle>>>,
     pub global_value_ids: Arc<RwLock<HashMap<u64, PyObject>>>,
+    pub class_names: Mutex<HashMap<StrRef, Type>>,
     pub pyid_to_def: Arc<RwLock<HashMap<u64, DefinitionId>>>,
     pub pyid_to_type: Arc<RwLock<HashMap<u64, Type>>>,
     pub primitive_ids: PrimitivePythonId,
@@ -135,8 +133,6 @@ impl StaticValue for PythonValue {
                         format!("{}_const", self.id).as_str(),
                     );
                     global.set_constant(true);
-                    // Set linkage of global to private to avoid name collisions
-                    global.set_linkage(Linkage::Private);
                     global.set_initializer(&ctx.ctx.const_struct(
                         &[ctx.ctx.i32_type().const_int(u64::from(id), false).into()],
                         false,
@@ -167,7 +163,7 @@ impl StaticValue for PythonValue {
                 PrimitiveValue::Bool(val) => {
                     ctx.ctx.i8_type().const_int(u64::from(*val), false).into()
                 }
-                PrimitiveValue::Str(val) => ctx.gen_string(generator, val).into(),
+                PrimitiveValue::Str(val) => ctx.ctx.const_string(val.as_bytes(), true).into(),
             });
         }
         if let Some(global) = ctx.module.get_global(&self.id.to_string()) {
@@ -981,7 +977,7 @@ impl InnerResolver {
         } else if ty_id == self.primitive_ids.string || ty_id == self.primitive_ids.np_str_ {
             let val: String = obj.extract().unwrap();
             self.id_to_primitive.write().insert(id, PrimitiveValue::Str(val.clone()));
-            Ok(Some(ctx.gen_string(generator, val).into()))
+            Ok(Some(ctx.ctx.const_string(val.as_bytes(), true).into()))
         } else if ty_id == self.primitive_ids.float || ty_id == self.primitive_ids.float64 {
             let val: f64 = obj.extract().unwrap();
             self.id_to_primitive.write().insert(id, PrimitiveValue::F64(val));
@@ -1085,19 +1081,18 @@ impl InnerResolver {
             } else {
                 unreachable!("must be ndarray")
             };
-            let (ndarray_dtype, _) = unpack_ndarray_var_tys(&mut ctx.unifier, ndarray_ty);
+            let (ndarray_dtype, ndarray_ndims) =
+                unpack_ndarray_var_tys(&mut ctx.unifier, ndarray_ty);
 
-            let llvm_i8 = ctx.ctx.i8_type();
-            let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
             let llvm_usize = generator.get_size_type(ctx.ctx);
-            let llvm_ndarray = NDArrayType::from_unifier_type(generator, ctx, ndarray_ty);
-            let dtype = llvm_ndarray.element_type();
+            let ndarray_dtype_llvm_ty = ctx.get_llvm_type(generator, ndarray_dtype);
+            let ndarray_llvm_ty = NDArrayType::new(generator, ctx.ctx, ndarray_dtype_llvm_ty);
 
             {
                 if self.global_value_ids.read().contains_key(&id) {
                     let global = ctx.module.get_global(&id_str).unwrap_or_else(|| {
                         ctx.module.add_global(
-                            llvm_ndarray.as_base_type().get_element_type().into_struct_type(),
+                            ndarray_llvm_ty.as_underlying_type(),
                             Some(AddressSpace::default()),
                             &id_str,
                         )
@@ -1107,41 +1102,40 @@ impl InnerResolver {
                 self.global_value_ids.write().insert(id, obj.into());
             }
 
-            let ndims = llvm_ndarray.ndims().unwrap();
+            let TypeEnum::TLiteral { values, .. } = &*ctx.unifier.get_ty_immutable(ndarray_ndims)
+            else {
+                unreachable!("Expected Literal for ndarray_ndims")
+            };
+
+            let ndarray_ndims = if values.len() == 1 {
+                values[0].clone()
+            } else {
+                todo!("Unpacking literal of more than one element unimplemented")
+            };
+            let Ok(ndarray_ndims) = u64::try_from(ndarray_ndims) else {
+                unreachable!("Expected u64 value for ndarray_ndims")
+            };
 
             // Obtain the shape of the ndarray
             let shape_tuple: &PyTuple = obj.getattr("shape")?.downcast()?;
-            assert_eq!(shape_tuple.len(), ndims as usize);
-
-            // The Rust type inferencer cannot figure this out
-            let shape_values = shape_tuple
+            assert_eq!(shape_tuple.len(), ndarray_ndims as usize);
+            let shape_values: Result<Option<Vec<_>>, _> = shape_tuple
                 .iter()
                 .enumerate()
                 .map(|(i, elem)| {
-                    let value = self
-                        .get_obj_value(py, elem, ctx, generator, ctx.primitives.usize())
-                        .map_err(|e| {
-                            super::CompileError::new_err(format!("Error getting element {i}: {e}"))
-                        })?
-                        .unwrap();
-                    let value = value.into_int_value();
-                    Ok(value)
+                    self.get_obj_value(py, elem, ctx, generator, ctx.primitives.usize()).map_err(
+                        |e| super::CompileError::new_err(format!("Error getting element {i}: {e}")),
+                    )
                 })
-                .collect::<Result<Vec<_>, PyErr>>()?;
-
-            // Also use this opportunity to get the constant values of `shape_values` for calculating strides.
-            let shape_u64s = shape_values
-                .iter()
-                .map(|dim| {
-                    assert!(dim.is_const());
-                    dim.get_zero_extended_constant().unwrap()
-                })
-                .collect_vec();
-            let shape_values = llvm_usize.const_array(&shape_values);
+                .collect();
+            let shape_values = shape_values?.unwrap();
+            let shape_values = llvm_usize.const_array(
+                &shape_values.into_iter().map(BasicValueEnum::into_int_value).collect_vec(),
+            );
 
             // create a global for ndarray.shape and initialize it using the shape
             let shape_global = ctx.module.add_global(
-                llvm_usize.array_type(ndims as u32),
+                llvm_usize.array_type(ndarray_ndims as u32),
                 Some(AddressSpace::default()),
                 &(id_str.clone() + ".shape"),
             );
@@ -1149,25 +1143,17 @@ impl InnerResolver {
 
             // Obtain the (flattened) elements of the ndarray
             let sz: usize = obj.getattr("size")?.extract()?;
-            let data: Vec<_> = (0..sz)
+            let data: Result<Option<Vec<_>>, _> = (0..sz)
                 .map(|i| {
                     obj.getattr("flat")?.get_item(i).and_then(|elem| {
-                        let value = self
-                            .get_obj_value(py, elem, ctx, generator, ndarray_dtype)
-                            .map_err(|e| {
-                                super::CompileError::new_err(format!(
-                                    "Error getting element {i}: {e}"
-                                ))
-                            })?
-                            .unwrap();
-
-                        assert_eq!(value.get_type(), dtype);
-                        Ok(value)
+                        self.get_obj_value(py, elem, ctx, generator, ndarray_dtype).map_err(|e| {
+                            super::CompileError::new_err(format!("Error getting element {i}: {e}"))
+                        })
                     })
                 })
-                .try_collect()?;
-            let data = data.into_iter();
-            let data = match dtype {
+                .collect();
+            let data = data?.unwrap().into_iter();
+            let data = match ndarray_dtype_llvm_ty {
                 BasicTypeEnum::ArrayType(ty) => {
                     ty.const_array(&data.map(BasicValueEnum::into_array_value).collect_vec())
                 }
@@ -1192,68 +1178,34 @@ impl InnerResolver {
             };
 
             // create a global for ndarray.data and initialize it using the elements
-            //
-            // NOTE: NDArray's `data` is `u8*`. Here, `data_global` is an array of `dtype`.
-            // We will have to cast it to an `u8*` later.
             let data_global = ctx.module.add_global(
-                dtype.array_type(sz as u32),
+                ndarray_dtype_llvm_ty.array_type(sz as u32),
                 Some(AddressSpace::default()),
                 &(id_str.clone() + ".data"),
             );
             data_global.set_initializer(&data);
 
-            // Get the constant itemsize.
-            let itemsize = dtype.size_of().unwrap();
-            let itemsize = itemsize.get_zero_extended_constant().unwrap();
-
-            // Create the strides needed for ndarray.strides
-            let strides = make_contiguous_strides(itemsize, ndims, &shape_u64s);
-            let strides =
-                strides.into_iter().map(|stride| llvm_usize.const_int(stride, false)).collect_vec();
-            let strides = llvm_usize.const_array(&strides);
-
-            // create a global for ndarray.strides and initialize it
-            let strides_global = ctx.module.add_global(
-                llvm_i8.array_type(ndims as u32),
-                Some(AddressSpace::default()),
-                &format!("${id_str}.strides"),
-            );
-            strides_global.set_initializer(&strides);
-
             // create a global for the ndarray object and initialize it
+            let value = ndarray_llvm_ty.as_underlying_type().const_named_struct(&[
+                llvm_usize.const_int(ndarray_ndims, false).into(),
+                shape_global
+                    .as_pointer_value()
+                    .const_cast(llvm_usize.ptr_type(AddressSpace::default()))
+                    .into(),
+                data_global
+                    .as_pointer_value()
+                    .const_cast(ndarray_dtype_llvm_ty.ptr_type(AddressSpace::default()))
+                    .into(),
+            ]);
 
-            // NOTE: data_global is an array of dtype, we want a `u8*`.
-            let ndarray_data = data_global.as_pointer_value();
-            let ndarray_data = ctx.builder.build_pointer_cast(ndarray_data, llvm_pi8, "").unwrap();
-
-            let ndarray_itemsize = llvm_usize.const_int(itemsize, false);
-
-            let ndarray_ndims = llvm_usize.const_int(ndims, false);
-
-            let ndarray_shape = shape_global.as_pointer_value();
-
-            let ndarray_strides = strides_global.as_pointer_value();
-
-            let ndarray = llvm_ndarray
-                .as_base_type()
-                .get_element_type()
-                .into_struct_type()
-                .const_named_struct(&[
-                    ndarray_itemsize.into(),
-                    ndarray_ndims.into(),
-                    ndarray_shape.into(),
-                    ndarray_strides.into(),
-                    ndarray_data.into(),
-                ]);
-
-            let ndarray_global = ctx.module.add_global(
-                llvm_ndarray.as_base_type().get_element_type().into_struct_type(),
+            let ndarray = ctx.module.add_global(
+                ndarray_llvm_ty.as_underlying_type(),
                 Some(AddressSpace::default()),
                 &id_str,
             );
-            ndarray_global.set_initializer(&ndarray);
+            ndarray.set_initializer(&value);
 
-            Ok(Some(ndarray_global.as_pointer_value().into()))
+            Ok(Some(ndarray.as_pointer_value().into()))
         } else if ty_id == self.primitive_ids.tuple {
             let expected_ty_enum = ctx.unifier.get_ty_immutable(expected_ty);
             let TypeEnum::TTuple { ty, is_vararg_ctx: false } = expected_ty_enum.as_ref() else {
@@ -1514,7 +1466,6 @@ impl SymbolResolver for Resolver {
         &self,
         id: StrRef,
         _: &mut CodeGenContext<'ctx, '_>,
-        _: &mut dyn CodeGenerator,
     ) -> Option<ValueEnum<'ctx>> {
         let sym_value = {
             let id_to_val = self.0.id_to_pyval.read();
@@ -1576,7 +1527,10 @@ impl SymbolResolver for Resolver {
         if let Some(id) = string_store.get(s) {
             *id
         } else {
-            let id = i32::try_from(string_store.len()).unwrap();
+            let id = Python::with_gil(|py| -> PyResult<i32> {
+                self.0.helper.store_str.call1(py, (s,))?.extract(py)
+            })
+            .unwrap();
             string_store.insert(s.into(), id);
             id
         }

nac3artiq/src/timeline.rs

@@ -1,12 +1,9 @@
-use itertools::Either;
-
-use nac3core::{
-    codegen::CodeGenContext,
-    inkwell::{
-        values::{BasicValueEnum, CallSiteValue},
-        AddressSpace, AtomicOrdering,
-    },
+use inkwell::{
+    values::{BasicValueEnum, CallSiteValue},
+    AddressSpace, AtomicOrdering,
 };
+use itertools::Either;
+use nac3core::codegen::CodeGenContext;
 
 /// Functions for manipulating the timeline.
 pub trait TimeFns {
@@ -34,7 +31,7 @@ impl TimeFns for NowPinningTimeFns64 {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -83,7 +80,7 @@ impl TimeFns for NowPinningTimeFns64 {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -112,7 +109,7 @@ impl TimeFns for NowPinningTimeFns64 {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -210,7 +207,7 @@ impl TimeFns for NowPinningTimeFns {
             .unwrap_or_else(|| ctx.module.add_global(i64_type, None, "now"));
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 
@@ -261,7 +258,7 @@ impl TimeFns for NowPinningTimeFns {
         let time_lo = ctx.builder.build_int_truncate(time, i32_type, "time.lo").unwrap();
         let now_hiptr = ctx
             .builder
-            .build_bit_cast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
+            .build_bitcast(now, i32_type.ptr_type(AddressSpace::default()), "now.hi.addr")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap();
 

nac3ast/Cargo.toml

@@ -10,6 +10,7 @@ constant-optimization = ["fold"]
 fold = []
 
 [dependencies]
+lazy_static = "1.5"
 parking_lot = "0.12"
 string-interner = "0.17"
 fxhash = "0.2"

nac3ast/src/ast_gen.rs

@@ -5,12 +5,14 @@ pub use crate::location::Location;
 
 use fxhash::FxBuildHasher;
 use parking_lot::{Mutex, MutexGuard};
-use std::{cell::RefCell, collections::HashMap, fmt, sync::LazyLock};
+use std::{cell::RefCell, collections::HashMap, fmt};
 use string_interner::{symbol::SymbolU32, DefaultBackend, StringInterner};
 
 pub type Interner = StringInterner<DefaultBackend, FxBuildHasher>;
-static INTERNER: LazyLock<Mutex<Interner>> =
-    LazyLock::new(|| Mutex::new(StringInterner::with_hasher(FxBuildHasher::default())));
+lazy_static! {
+    static ref INTERNER: Mutex<Interner> =
+        Mutex::new(StringInterner::with_hasher(FxBuildHasher::default()));
+}
 
 thread_local! {
     static LOCAL_INTERNER: RefCell<HashMap<String, StrRef>> = RefCell::default();

nac3ast/src/lib.rs

@@ -1,4 +1,10 @@
-#![deny(future_incompatible, let_underscore, nonstandard_style, clippy::all)]
+#![deny(
+    future_incompatible,
+    let_underscore,
+    nonstandard_style,
+    rust_2024_compatibility,
+    clippy::all
+)]
 #![warn(clippy::pedantic)]
 #![allow(
     clippy::missing_errors_doc,
@@ -8,6 +14,9 @@
     clippy::wildcard_imports
 )]
 
+#[macro_use]
+extern crate lazy_static;
+
 mod ast_gen;
 mod constant;
 #[cfg(feature = "fold")]

nac3core/Cargo.toml

@@ -1,29 +1,26 @@
+[features]
+test = []
+
 [package]
 name = "nac3core"
 version = "0.1.0"
 authors = ["M-Labs"]
 edition = "2021"
 
-[features]
-default = ["derive"]
-derive = ["dep:nac3core_derive"]
-no-escape-analysis = []
-
 [dependencies]
 itertools = "0.13"
 crossbeam = "0.8"
-indexmap = "2.6"
+indexmap = "2.2"
 parking_lot = "0.12"
-rayon = "1.10"
-nac3core_derive = { path = "nac3core_derive", optional = true }
+rayon = "1.8"
 nac3parser = { path = "../nac3parser" }
-strum = "0.26"
-strum_macros = "0.26"
+strum = "0.26.2"
+strum_macros = "0.26.4"
 
 [dependencies.inkwell]
-version = "0.5"
+version = "0.4"
 default-features = false
-features = ["llvm14-0-prefer-dynamic", "target-x86", "target-arm", "target-riscv", "no-libffi-linking"]
+features = ["llvm14-0", "target-x86", "target-arm", "target-riscv", "no-libffi-linking"]
 
 [dev-dependencies]
 test-case = "1.2.0"

nac3core/build.rs

@@ -1,32 +1,49 @@
+use regex::Regex;
 use std::{
     env,
     fs::File,
     io::Write,
-    path::Path,
+    path::{Path, PathBuf},
     process::{Command, Stdio},
 };
 
-use regex::Regex;
+const CMD_IRRT_CLANG: &str = "clang-irrt";
+const CMD_IRRT_CLANG_TEST: &str = "clang-irrt-test";
+const CMD_IRRT_LLVM_AS: &str = "llvm-as-irrt";
 
-fn main() {
-    let out_dir = env::var("OUT_DIR").unwrap();
-    let out_dir = Path::new(&out_dir);
-    let irrt_dir = Path::new("irrt");
+fn get_out_dir() -> PathBuf {
+    PathBuf::from(env::var("OUT_DIR").unwrap())
+}
 
-    let irrt_cpp_path = irrt_dir.join("irrt.cpp");
+fn get_irrt_dir() -> &'static Path {
+    Path::new("irrt")
+}
+
+/// Compile `irrt.cpp` for use in `src/codegen`
+fn compile_irrt_cpp() {
+    let out_dir = get_out_dir();
+    let irrt_dir = get_irrt_dir();
+
+    let (opt_flag, debug_assert_flag) = match env::var("PROFILE").as_deref() {
+        Ok("debug") => ("-O0", "-DIRRT_DEBUG_ASSERT=true"),
+        Ok("release") => ("-O3", "-DIRRT_DEBUG_ASSERT=false"),
+        flavor => panic!("Unknown or missing build flavor {flavor:?}"),
+    };
 
     /*
      * HACK: Sadly, clang doesn't let us emit generic LLVM bitcode.
      * Compiling for WASM32 and filtering the output with regex is the closest we can get.
      */
-    let mut flags: Vec<&str> = vec![
+    let irrt_cpp_path = irrt_dir.join("irrt.cpp");
+    let flags: &[&str] = &[
         "--target=wasm32",
         "-x",
         "c++",
-        "-std=c++20",
         "-fno-discard-value-names",
         "-fno-exceptions",
         "-fno-rtti",
+        opt_flag,
+        debug_assert_flag,
         "-emit-llvm",
         "-S",
         "-Wall",
@@ -38,26 +55,16 @@ fn main() {
         irrt_cpp_path.to_str().unwrap(),
     ];
 
-    match env::var("PROFILE").as_deref() {
-        Ok("debug") => {
-            flags.push("-O0");
-            flags.push("-DIRRT_DEBUG_ASSERT");
-        }
-        Ok("release") => {
-            flags.push("-O3");
-        }
-        flavor => panic!("Unknown or missing build flavor {flavor:?}"),
-    }
-
     // Tell Cargo to rerun if any file under `irrt_dir` (recursive) changes
     println!("cargo:rerun-if-changed={}", irrt_dir.to_str().unwrap());
 
     // Compile IRRT and capture the LLVM IR output
-    let output = Command::new("clang-irrt")
+    let output = Command::new(CMD_IRRT_CLANG)
         .args(flags)
         .output()
-        .inspect(|o| {
+        .map(|o| {
             assert!(o.status.success(), "{}", std::str::from_utf8(&o.stderr).unwrap());
+            o
         })
         .unwrap();
 
@@ -98,7 +105,9 @@ fn main() {
         file.write_all(filtered_output.as_bytes()).unwrap();
     }
 
-    let mut llvm_as = Command::new("llvm-as-irrt")
+    // Assemble the emitted and filtered IR to .bc
+    // That .bc will be integrated into nac3core's codegen
+    let mut llvm_as = Command::new(CMD_IRRT_LLVM_AS)
         .stdin(Stdio::piped())
         .arg("-o")
         .arg(out_dir.join("irrt.bc"))
@@ -107,3 +116,48 @@ fn main() {
     llvm_as.stdin.as_mut().unwrap().write_all(filtered_output.as_bytes()).unwrap();
     assert!(llvm_as.wait().unwrap().success());
 }
+
+/// Compile `irrt_test.cpp` for testing
+fn compile_irrt_test_cpp() {
+    let out_dir = get_out_dir();
+    let irrt_dir = get_irrt_dir();
+
+    let exe_path = out_dir.join("irrt_test.out"); // Output path of the compiled test executable
+    let irrt_test_cpp_path = irrt_dir.join("irrt_test.cpp");
+    let flags: &[&str] = &[
+        irrt_test_cpp_path.to_str().unwrap(),
+        "-x",
+        "c++",
+        "-I",
+        irrt_dir.to_str().unwrap(),
+        "-g",
+        "-fno-discard-value-names",
+        "-O0",
+        "-Wall",
+        "-Wextra",
+        "-Werror=return-type",
+        "-lm", // for `tgamma()`, `lgamma()`
+        "-o",
+        exe_path.to_str().unwrap(),
+    ];
+
+    Command::new(CMD_IRRT_CLANG_TEST)
+        .args(flags)
+        .output()
+        .map(|o| {
+            assert!(o.status.success(), "{}", std::str::from_utf8(&o.stderr).unwrap());
+            o
+        })
+        .unwrap();
+    println!("cargo:rerun-if-changed={}", irrt_dir.to_str().unwrap());
+}
+
+fn main() {
+    compile_irrt_cpp();
+
+    // https://github.com/rust-lang/cargo/issues/2549
+    // `cargo test -F test` to also build `irrt_test.cpp
+    if cfg!(feature = "test") {
+        compile_irrt_test_cpp();
+    }
+}

nac3core/irrt/irrt.cpp

@@ -1,10 +1,10 @@
-#include "irrt/exception.hpp"
-#include "irrt/list.hpp"
-#include "irrt/math.hpp"
-#include "irrt/ndarray.hpp"
-#include "irrt/range.hpp"
-#include "irrt/slice.hpp"
-#include "irrt/ndarray/basic.hpp"
-#include "irrt/ndarray/def.hpp"
-#include "irrt/ndarray/iter.hpp"
-#include "irrt/ndarray/indexing.hpp"
+#define IRRT_DEFINE_TYPEDEF_INTS
+#include <irrt_everything.hpp>
+
+/*
+ * All IRRT implementations.
+ *
+ * We don't have pre-compiled objects, so we are writing all implementations in
+ * headers and concatenate them with `#include` into one massive source file that
+ * contains all the IRRT stuff.
+ */

nac3core/irrt/irrt/core.hpp

@@ -0,0 +1,349 @@
+#pragma once
+
+#include <irrt/int_defs.hpp>
+#include <irrt/util.hpp>
+
+// NDArray indices are always `uint32_t`.
+using NDIndexInt = uint32_t;
+// The type of an index or a value describing the length of a
+// range/slice is always `int32_t`.
+using SliceIndex = int32_t;
+
+namespace {
+// adapted from GNU Scientific Library:
+// https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
+// need to make sure `exp >= 0` before calling this function
+template <typename T>
+T __nac3_int_exp_impl(T base, T exp) {
+    T res = 1;
+    /* repeated squaring method */
+    do {
+        if (exp & 1) {
+            res *= base; /* for n odd */
+        }
+        exp >>= 1;
+        base *= base;
+    } while (exp);
+    return res;
+}
+
+template <typename SizeT>
+SizeT __nac3_ndarray_calc_size_impl(const SizeT* list_data, SizeT list_len,
+                                    SizeT begin_idx, SizeT end_idx) {
+    __builtin_assume(end_idx <= list_len);
+
+    SizeT num_elems = 1;
+    for (SizeT i = begin_idx; i < end_idx; ++i) {
+        SizeT val = list_data[i];
+        __builtin_assume(val > 0);
+        num_elems *= val;
+    }
+    return num_elems;
+}
+
+template <typename SizeT>
+void __nac3_ndarray_calc_nd_indices_impl(SizeT index, const SizeT* dims,
+                                         SizeT num_dims, NDIndexInt* idxs) {
+    SizeT stride = 1;
+    for (SizeT dim = 0; dim < num_dims; dim++) {
+        SizeT i = num_dims - dim - 1;
+        __builtin_assume(dims[i] > 0);
+        idxs[i] = (index / stride) % dims[i];
+        stride *= dims[i];
+    }
+}
+
+template <typename SizeT>
+SizeT __nac3_ndarray_flatten_index_impl(const SizeT* dims, SizeT num_dims,
+                                        const NDIndexInt* indices,
+                                        SizeT num_indices) {
+    SizeT idx = 0;
+    SizeT stride = 1;
+    for (SizeT i = 0; i < num_dims; ++i) {
+        SizeT ri = num_dims - i - 1;
+        if (ri < num_indices) {
+            idx += stride * indices[ri];
+        }
+
+        __builtin_assume(dims[i] > 0);
+        stride *= dims[ri];
+    }
+    return idx;
+}
+
+template <typename SizeT>
+void __nac3_ndarray_calc_broadcast_impl(const SizeT* lhs_dims, SizeT lhs_ndims,
+                                        const SizeT* rhs_dims, SizeT rhs_ndims,
+                                        SizeT* out_dims) {
+    SizeT max_ndims = lhs_ndims > rhs_ndims ? lhs_ndims : rhs_ndims;
+
+    for (SizeT i = 0; i < max_ndims; ++i) {
+        const SizeT* lhs_dim_sz =
+            i < lhs_ndims ? &lhs_dims[lhs_ndims - i - 1] : nullptr;
+        const SizeT* rhs_dim_sz =
+            i < rhs_ndims ? &rhs_dims[rhs_ndims - i - 1] : nullptr;
+
+        SizeT* out_dim = &out_dims[max_ndims - i - 1];
+
+        if (lhs_dim_sz == nullptr) {
+            *out_dim = *rhs_dim_sz;
+        } else if (rhs_dim_sz == nullptr) {
+            *out_dim = *lhs_dim_sz;
+        } else if (*lhs_dim_sz == 1) {
+            *out_dim = *rhs_dim_sz;
+        } else if (*rhs_dim_sz == 1) {
+            *out_dim = *lhs_dim_sz;
+        } else if (*lhs_dim_sz == *rhs_dim_sz) {
+            *out_dim = *lhs_dim_sz;
+        } else {
+            __builtin_unreachable();
+        }
+    }
+}
+
+template <typename SizeT>
+void __nac3_ndarray_calc_broadcast_idx_impl(const SizeT* src_dims,
+                                            SizeT src_ndims,
+                                            const NDIndexInt* in_idx,
+                                            NDIndexInt* out_idx) {
+    for (SizeT i = 0; i < src_ndims; ++i) {
+        SizeT src_i = src_ndims - i - 1;
+        out_idx[src_i] = src_dims[src_i] == 1 ? 0 : in_idx[src_i];
+    }
+}
+}  // namespace
+
+extern "C" {
+#define DEF_nac3_int_exp_(T)                   \
+    T __nac3_int_exp_##T(T base, T exp) {      \
+        return __nac3_int_exp_impl(base, exp); \
+    }
+
+DEF_nac3_int_exp_(int32_t);
+DEF_nac3_int_exp_(int64_t);
+DEF_nac3_int_exp_(uint32_t);
+DEF_nac3_int_exp_(uint64_t);
+
+SliceIndex __nac3_slice_index_bound(SliceIndex i, const SliceIndex len) {
+    if (i < 0) {
+        i = len + i;
+    }
+    if (i < 0) {
+        return 0;
+    } else if (i > len) {
+        return len;
+    }
+    return i;
+}
+
+SliceIndex __nac3_range_slice_len(const SliceIndex start, const SliceIndex end,
+                                  const SliceIndex step) {
+    SliceIndex diff = end - start;
+    if (diff > 0 && step > 0) {
+        return ((diff - 1) / step) + 1;
+    } else if (diff < 0 && step < 0) {
+        return ((diff + 1) / step) + 1;
+    } else {
+        return 0;
+    }
+}
+
+// Handle list assignment and dropping part of the list when
+// both dest_step and src_step are +1.
+// - All the index must *not* be out-of-bound or negative,
+// - The end index is *inclusive*,
+// - The length of src and dest slice size should already
+// be checked: if dest.step == 1 then len(src) <= len(dest) else
+// len(src) == len(dest)
+SliceIndex __nac3_list_slice_assign_var_size(
+    SliceIndex dest_start, SliceIndex dest_end, SliceIndex dest_step,
+    uint8_t* dest_arr, SliceIndex dest_arr_len, SliceIndex src_start,
+    SliceIndex src_end, SliceIndex src_step, uint8_t* src_arr,
+    SliceIndex src_arr_len, const SliceIndex size) {
+    /* if dest_arr_len == 0, do nothing since we do not support
+     * extending list
+     */
+    if (dest_arr_len == 0) return dest_arr_len;
+    /* if both step is 1, memmove directly, handle the dropping of
+     * the list, and shrink size */
+    if (src_step == dest_step && dest_step == 1) {
+        const SliceIndex src_len =
+            (src_end >= src_start) ? (src_end - src_start + 1) : 0;
+        const SliceIndex dest_len =
+            (dest_end >= dest_start) ? (dest_end - dest_start + 1) : 0;
+        if (src_len > 0) {
+            __builtin_memmove(dest_arr + dest_start * size,
+                              src_arr + src_start * size, src_len * size);
+        }
+        if (dest_len > 0) {
+            /* dropping */
+            __builtin_memmove(dest_arr + (dest_start + src_len) * size,
+                              dest_arr + (dest_end + 1) * size,
+                              (dest_arr_len - dest_end - 1) * size);
+        }
+        /* shrink size */
+        return dest_arr_len - (dest_len - src_len);
+    }
+    /* if two range overlaps, need alloca */
+    uint8_t need_alloca =
+        (dest_arr == src_arr) &&
+        !(max(dest_start, dest_end) < min(src_start, src_end) ||
+          max(src_start, src_end) < min(dest_start, dest_end));
+    if (need_alloca) {
+        uint8_t* tmp =
+            reinterpret_cast<uint8_t*>(__builtin_alloca(src_arr_len * size));
+        __builtin_memcpy(tmp, src_arr, src_arr_len * size);
+        src_arr = tmp;
+    }
+    SliceIndex src_ind = src_start;
+    SliceIndex dest_ind = dest_start;
+    for (; (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end);
+         src_ind += src_step, dest_ind += dest_step) {
+        /* for constant optimization */
+        if (size == 1) {
+            __builtin_memcpy(dest_arr + dest_ind, src_arr + src_ind, 1);
+        } else if (size == 4) {
+            __builtin_memcpy(dest_arr + dest_ind * 4, src_arr + src_ind * 4, 4);
+        } else if (size == 8) {
+            __builtin_memcpy(dest_arr + dest_ind * 8, src_arr + src_ind * 8, 8);
+        } else {
+            /* memcpy for var size, cannot overlap after previous
+             * alloca */
+            __builtin_memcpy(dest_arr + dest_ind * size,
+                             src_arr + src_ind * size, size);
+        }
+    }
+    /* only dest_step == 1 can we shrink the dest list. */
+    /* size should be ensured prior to calling this function */
+    if (dest_step == 1 && dest_end >= dest_start) {
+        __builtin_memmove(
+            dest_arr + dest_ind * size, dest_arr + (dest_end + 1) * size,
+            (dest_arr_len - dest_end - 1) * size + size + size + size);
+        return dest_arr_len - (dest_end - dest_ind) - 1;
+    }
+    return dest_arr_len;
+}
+
+int32_t __nac3_isinf(double x) { return __builtin_isinf(x); }
+
+int32_t __nac3_isnan(double x) { return __builtin_isnan(x); }
+
+double tgamma(double arg);
+
+double __nac3_gamma(double z) {
+    // Handling for denormals
+    //     | x                 | Python gamma(x) | C tgamma(x) |
+    // --- | ----------------- | --------------- | ----------- |
+    // (1) | nan               | nan             | nan         |
+    // (2) | -inf              | -inf            | inf         |
+    // (3) | inf               | inf             | inf         |
+    // (4) | 0.0               | inf             | inf         |
+    // (5) | {-1.0, -2.0, ...} | inf             | nan         |
+
+    // (1)-(3)
+    if (__builtin_isinf(z) || __builtin_isnan(z)) {
+        return z;
+    }
+
+    double v = tgamma(z);
+
+    // (4)-(5)
+    return __builtin_isinf(v) || __builtin_isnan(v) ? __builtin_inf() : v;
+}
+
+double lgamma(double arg);
+
+double __nac3_gammaln(double x) {
+    // libm's handling of value overflows differs from scipy:
+    // - scipy: gammaln(-inf) -> -inf
+    // - libm : lgamma(-inf) -> inf
+
+    if (__builtin_isinf(x)) {
+        return x;
+    }
+
+    return lgamma(x);
+}
+
+double j0(double x);
+
+double __nac3_j0(double x) {
+    // libm's handling of value overflows differs from scipy:
+    // - scipy: j0(inf) -> nan
+    // - libm : j0(inf) -> 0.0
+
+    if (__builtin_isinf(x)) {
+        return __builtin_nan("");
+    }
+
+    return j0(x);
+}
+
+uint32_t __nac3_ndarray_calc_size(const uint32_t* list_data, uint32_t list_len,
+                                  uint32_t begin_idx, uint32_t end_idx) {
+    return __nac3_ndarray_calc_size_impl(list_data, list_len, begin_idx,
+                                         end_idx);
+}
+
+uint64_t __nac3_ndarray_calc_size64(const uint64_t* list_data,
+                                    uint64_t list_len, uint64_t begin_idx,
+                                    uint64_t end_idx) {
+    return __nac3_ndarray_calc_size_impl(list_data, list_len, begin_idx,
+                                         end_idx);
+}
+
+void __nac3_ndarray_calc_nd_indices(uint32_t index, const uint32_t* dims,
+                                    uint32_t num_dims, NDIndexInt* idxs) {
+    __nac3_ndarray_calc_nd_indices_impl(index, dims, num_dims, idxs);
+}
+
+void __nac3_ndarray_calc_nd_indices64(uint64_t index, const uint64_t* dims,
+                                      uint64_t num_dims, NDIndexInt* idxs) {
+    __nac3_ndarray_calc_nd_indices_impl(index, dims, num_dims, idxs);
+}
+
+uint32_t __nac3_ndarray_flatten_index(const uint32_t* dims, uint32_t num_dims,
+                                      const NDIndexInt* indices,
+                                      uint32_t num_indices) {
+    return __nac3_ndarray_flatten_index_impl(dims, num_dims, indices,
+                                             num_indices);
+}
+
+uint64_t __nac3_ndarray_flatten_index64(const uint64_t* dims, uint64_t num_dims,
+                                        const NDIndexInt* indices,
+                                        uint64_t num_indices) {
+    return __nac3_ndarray_flatten_index_impl(dims, num_dims, indices,
+                                             num_indices);
+}
+
+void __nac3_ndarray_calc_broadcast(const uint32_t* lhs_dims, uint32_t lhs_ndims,
+                                   const uint32_t* rhs_dims, uint32_t rhs_ndims,
+                                   uint32_t* out_dims) {
+    return __nac3_ndarray_calc_broadcast_impl(lhs_dims, lhs_ndims, rhs_dims,
+                                              rhs_ndims, out_dims);
+}
+
+void __nac3_ndarray_calc_broadcast64(const uint64_t* lhs_dims,
+                                     uint64_t lhs_ndims,
+                                     const uint64_t* rhs_dims,
+                                     uint64_t rhs_ndims, uint64_t* out_dims) {
+    return __nac3_ndarray_calc_broadcast_impl(lhs_dims, lhs_ndims, rhs_dims,
+                                              rhs_ndims, out_dims);
+}
+
+void __nac3_ndarray_calc_broadcast_idx(const uint32_t* src_dims,
+                                       uint32_t src_ndims,
+                                       const NDIndexInt* in_idx,
+                                       NDIndexInt* out_idx) {
+    __nac3_ndarray_calc_broadcast_idx_impl(src_dims, src_ndims, in_idx,
+                                           out_idx);
+}
+
+void __nac3_ndarray_calc_broadcast_idx64(const uint64_t* src_dims,
+                                         uint64_t src_ndims,
+                                         const NDIndexInt* in_idx,
+                                         NDIndexInt* out_idx) {
+    __nac3_ndarray_calc_broadcast_idx_impl(src_dims, src_ndims, in_idx,
+                                           out_idx);
+}
+}  // extern "C"

nac3core/irrt/irrt/cslice.hpp

@@ -1,9 +1,9 @@
 #pragma once
 
-#include "irrt/int_types.hpp"
+#include <irrt/int_defs.hpp>
 
-template<typename SizeT>
+template <typename SizeT>
 struct CSlice {
-    void* base;
+    uint8_t* base;
     SizeT len;
 };

nac3core/irrt/irrt/debug.hpp

@@ -1,25 +0,0 @@
-#pragma once
-
-// Set in nac3core/build.rs
-#ifdef IRRT_DEBUG_ASSERT
-#define IRRT_DEBUG_ASSERT_BOOL true
-#else
-#define IRRT_DEBUG_ASSERT_BOOL false
-#endif
-
-#define raise_debug_assert(SizeT, msg, param1, param2, param3) \
-    raise_exception(SizeT, EXN_ASSERTION_ERROR, "IRRT debug assert failed: " msg, param1, param2, param3)
-
-#define debug_assert_eq(SizeT, lhs, rhs)                                           \
-    if constexpr (IRRT_DEBUG_ASSERT_BOOL) {                                        \
-        if ((lhs) != (rhs)) {                                                      \
-            raise_debug_assert(SizeT, "LHS = {0}. RHS = {1}", lhs, rhs, NO_PARAM); \
-        }                                                                          \
-    }
-
-#define debug_assert(SizeT, expr)                                                  \
-    if constexpr (IRRT_DEBUG_ASSERT_BOOL) {                                        \
-        if (!(expr)) {                                                             \
-            raise_debug_assert(SizeT, "Got false.", NO_PARAM, NO_PARAM, NO_PARAM); \
-        }                                                                          \
-    }

nac3core/irrt/irrt/exception.hpp

@@ -1,37 +1,43 @@
 #pragma once
 
-#include "irrt/cslice.hpp"
-#include "irrt/int_types.hpp"
+#include <irrt/cslice.hpp>
+#include <irrt/int_defs.hpp>
+#include <irrt/util.hpp>
 
 /**
  * @brief The int type of ARTIQ exception IDs.
+ *
+ * It is always `int32_t`
  */
-using ExceptionId = int32_t;
+typedef int32_t ExceptionId;
 
 /*
- * Set of exceptions C++ IRRT can use.
+ * A set of exceptions IRRT can use.
  * Must be synchronized with `setup_irrt_exceptions` in `nac3core/src/codegen/irrt/mod.rs`.
+ * All exception IDs are initialized by `setup_irrt_exceptions`.
  */
+#ifdef IRRT_TESTING
+// If we are doing IRRT tests (i.e., running `cargo test -F test`), define them with a fake set of IDs.
+ExceptionId EXN_INDEX_ERROR = 0;
+ExceptionId EXN_VALUE_ERROR = 1;
+ExceptionId EXN_ASSERTION_ERROR = 2;
+ExceptionId EXN_RUNTIME_ERROR = 3;
+ExceptionId EXN_TYPE_ERROR = 4;
+#else
 extern "C" {
 ExceptionId EXN_INDEX_ERROR;
 ExceptionId EXN_VALUE_ERROR;
 ExceptionId EXN_ASSERTION_ERROR;
+ExceptionId EXN_RUNTIME_ERROR;
 ExceptionId EXN_TYPE_ERROR;
 }
-
-/**
- * @brief Extern function to `__nac3_raise`
- *
- * The parameter `err` could be `Exception<int32_t>` or `Exception<int64_t>`. The caller
- * must make sure to pass `Exception`s with the correct `SizeT` depending on the `size_t` of the runtime.
- */
-extern "C" void __nac3_raise(void* err);
+#endif
 
 namespace {
 /**
  * @brief NAC3's Exception struct
  */
-template<typename SizeT>
+template <typename SizeT>
 struct Exception {
     ExceptionId id;
     CSlice<SizeT> filename;
@@ -41,36 +47,51 @@ struct Exception {
     CSlice<SizeT> msg;
     int64_t params[3];
 };
+}  // namespace
 
-constexpr int64_t NO_PARAM = 0;
+// Declare/Define `__nac3_raise`
+#ifdef IRRT_TESTING
+#include <cstdio>
+void __nac3_raise(void* err) {
+    // TODO: Print the error content?
+    printf("__nac3_raise called. Exiting...\n");
+    exit(1);
+}
+#else
+/**
+ * @brief Extern function to `__nac3_raise`
+ *
+ * The parameter `err` could be `Exception<int32_t>` or `Exception<int64_t>`. The caller
+ * must make sure to pass `Exception`s with the correct `SizeT` depending on the `size_t` of the runtime.
+ */
+extern "C" void __nac3_raise(void* err);
+#endif
 
-template<typename SizeT>
-void _raise_exception_helper(ExceptionId id,
-                             const char* filename,
-                             int32_t line,
-                             const char* function,
-                             const char* msg,
-                             int64_t param0,
-                             int64_t param1,
-                             int64_t param2) {
+namespace {
+const int64_t NO_PARAM = 0;
+
+// Helper function to raise an exception with `__nac3_raise`
+// Do not use this function directly. See `raise_exception`.
+template <typename SizeT>
+void _raise_exception_helper(ExceptionId id, const char* filename, int32_t line,
+                             const char* function, const char* msg,
+                             int64_t param0, int64_t param1, int64_t param2) {
     Exception<SizeT> e = {
         .id = id,
-        .filename = {.base = reinterpret_cast<void*>(const_cast<char*>(filename)),
-                     .len = static_cast<SizeT>(__builtin_strlen(filename))},
+        .filename = {.base = (uint8_t*)filename,
+                     .len = (int32_t)cstr_utils::length(filename)},
         .line = line,
         .column = 0,
-        .function = {.base = reinterpret_cast<void*>(const_cast<char*>(function)),
-                     .len = static_cast<SizeT>(__builtin_strlen(function))},
-        .msg = {.base = reinterpret_cast<void*>(const_cast<char*>(msg)),
-                .len = static_cast<SizeT>(__builtin_strlen(msg))},
+        .function = {.base = (uint8_t*)function,
+                     .len = (int32_t)cstr_utils::length(function)},
+        .msg = {.base = (uint8_t*)msg, .len = (int32_t)cstr_utils::length(msg)},
     };
     e.params[0] = param0;
     e.params[1] = param1;
     e.params[2] = param2;
-    __nac3_raise(reinterpret_cast<void*>(&e));
+    __nac3_raise((void*)&e);
     __builtin_unreachable();
 }
-}  // namespace
 
 /**
  * @brief Raise an exception with location details (location in the IRRT source files).
@@ -78,8 +99,25 @@ void _raise_exception_helper(ExceptionId id,
  * @param id The ID of the exception to raise.
  * @param msg A global constant C-string of the error message.
  *
- * `param0` to `param2` are optional format arguments of `msg`. They should be set to
+ * `param0` and `param2` are optional format arguments of `msg`. They should be set to
  * `NO_PARAM` to indicate they are unused.
  */
-#define raise_exception(SizeT, id, msg, param0, param1, param2) \
-    _raise_exception_helper<SizeT>(id, __FILE__, __LINE__, __FUNCTION__, msg, param0, param1, param2)
+#define raise_exception(SizeT, id, msg, param0, param1, param2)               \
+    _raise_exception_helper<SizeT>(id, __FILE__, __LINE__, __FUNCTION__, msg, \
+                                   param0, param1, param2)
+
+/**
+ * @brief Throw a dummy error for testing.
+ */
+template <typename SizeT>
+void throw_dummy_error() {
+    raise_exception(SizeT, EXN_RUNTIME_ERROR, "dummy error", NO_PARAM, NO_PARAM,
+                    NO_PARAM);
+}
+}  // namespace
+
+extern "C" {
+void __nac3_throw_dummy_error() { throw_dummy_error<int32_t>(); }
+
+void __nac3_throw_dummy_error64() { throw_dummy_error<int64_t>(); }
+}

nac3core/irrt/irrt/int_defs.hpp

@@ -0,0 +1,12 @@
+#pragma once
+
+// This is made toggleable since `irrt_test.cpp` itself would include
+// headers that define these typedefs
+#ifdef IRRT_DEFINE_TYPEDEF_INTS
+using int8_t = _BitInt(8);
+using uint8_t = unsigned _BitInt(8);
+using int32_t = _BitInt(32);
+using uint32_t = unsigned _BitInt(32);
+using int64_t = _BitInt(64);
+using uint64_t = unsigned _BitInt(64);
+#endif

nac3core/irrt/irrt/int_types.hpp

@@ -1,27 +0,0 @@
-#pragma once
-
-#if __STDC_VERSION__ >= 202000
-using int8_t = _BitInt(8);
-using uint8_t = unsigned _BitInt(8);
-using int32_t = _BitInt(32);
-using uint32_t = unsigned _BitInt(32);
-using int64_t = _BitInt(64);
-using uint64_t = unsigned _BitInt(64);
-#else
-
-#pragma clang diagnostic push
-#pragma clang diagnostic ignored "-Wdeprecated-type"
-using int8_t = _ExtInt(8);
-using uint8_t = unsigned _ExtInt(8);
-using int32_t = _ExtInt(32);
-using uint32_t = unsigned _ExtInt(32);
-using int64_t = _ExtInt(64);
-using uint64_t = unsigned _ExtInt(64);
-#pragma clang diagnostic pop
-
-#endif
-
-// NDArray indices are always `uint32_t`.
-using NDIndexInt = uint32_t;
-// The type of an index or a value describing the length of a range/slice is always `int32_t`.
-using SliceIndex = int32_t;

nac3core/irrt/irrt/list.hpp

@@ -1,81 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-#include "irrt/math_util.hpp"
-
-extern "C" {
-// Handle list assignment and dropping part of the list when
-// both dest_step and src_step are +1.
-// - All the index must *not* be out-of-bound or negative,
-// - The end index is *inclusive*,
-// - The length of src and dest slice size should already
-// be checked: if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest)
-SliceIndex __nac3_list_slice_assign_var_size(SliceIndex dest_start,
-                                             SliceIndex dest_end,
-                                             SliceIndex dest_step,
-                                             void* dest_arr,
-                                             SliceIndex dest_arr_len,
-                                             SliceIndex src_start,
-                                             SliceIndex src_end,
-                                             SliceIndex src_step,
-                                             void* src_arr,
-                                             SliceIndex src_arr_len,
-                                             const SliceIndex size) {
-    /* if dest_arr_len == 0, do nothing since we do not support extending list */
-    if (dest_arr_len == 0)
-        return dest_arr_len;
-    /* if both step is 1, memmove directly, handle the dropping of the list, and shrink size */
-    if (src_step == dest_step && dest_step == 1) {
-        const SliceIndex src_len = (src_end >= src_start) ? (src_end - src_start + 1) : 0;
-        const SliceIndex dest_len = (dest_end >= dest_start) ? (dest_end - dest_start + 1) : 0;
-        if (src_len > 0) {
-            __builtin_memmove(static_cast<uint8_t*>(dest_arr) + dest_start * size,
-                              static_cast<uint8_t*>(src_arr) + src_start * size, src_len * size);
-        }
-        if (dest_len > 0) {
-            /* dropping */
-            __builtin_memmove(static_cast<uint8_t*>(dest_arr) + (dest_start + src_len) * size,
-                              static_cast<uint8_t*>(dest_arr) + (dest_end + 1) * size,
-                              (dest_arr_len - dest_end - 1) * size);
-        }
-        /* shrink size */
-        return dest_arr_len - (dest_len - src_len);
-    }
-    /* if two range overlaps, need alloca */
-    uint8_t need_alloca = (dest_arr == src_arr)
-                          && !(max(dest_start, dest_end) < min(src_start, src_end)
-                               || max(src_start, src_end) < min(dest_start, dest_end));
-    if (need_alloca) {
-        void* tmp = __builtin_alloca(src_arr_len * size);
-        __builtin_memcpy(tmp, src_arr, src_arr_len * size);
-        src_arr = tmp;
-    }
-    SliceIndex src_ind = src_start;
-    SliceIndex dest_ind = dest_start;
-    for (; (src_step > 0) ? (src_ind <= src_end) : (src_ind >= src_end); src_ind += src_step, dest_ind += dest_step) {
-        /* for constant optimization */
-        if (size == 1) {
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind, static_cast<uint8_t*>(src_arr) + src_ind, 1);
-        } else if (size == 4) {
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind * 4,
-                             static_cast<uint8_t*>(src_arr) + src_ind * 4, 4);
-        } else if (size == 8) {
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind * 8,
-                             static_cast<uint8_t*>(src_arr) + src_ind * 8, 8);
-        } else {
-            /* memcpy for var size, cannot overlap after previous alloca */
-            __builtin_memcpy(static_cast<uint8_t*>(dest_arr) + dest_ind * size,
-                             static_cast<uint8_t*>(src_arr) + src_ind * size, size);
-        }
-    }
-    /* only dest_step == 1 can we shrink the dest list. */
-    /* size should be ensured prior to calling this function */
-    if (dest_step == 1 && dest_end >= dest_start) {
-        __builtin_memmove(static_cast<uint8_t*>(dest_arr) + dest_ind * size,
-                          static_cast<uint8_t*>(dest_arr) + (dest_end + 1) * size,
-                          (dest_arr_len - dest_end - 1) * size);
-        return dest_arr_len - (dest_end - dest_ind) - 1;
-    }
-    return dest_arr_len;
-}
-}  // extern "C"

nac3core/irrt/irrt/math.hpp

@@ -1,93 +0,0 @@
-#pragma once
-
-namespace {
-// adapted from GNU Scientific Library: https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
-// need to make sure `exp >= 0` before calling this function
-template<typename T>
-T __nac3_int_exp_impl(T base, T exp) {
-    T res = 1;
-    /* repeated squaring method */
-    do {
-        if (exp & 1) {
-            res *= base; /* for n odd */
-        }
-        exp >>= 1;
-        base *= base;
-    } while (exp);
-    return res;
-}
-}  // namespace
-
-#define DEF_nac3_int_exp_(T)                   \
-    T __nac3_int_exp_##T(T base, T exp) {      \
-        return __nac3_int_exp_impl(base, exp); \
-    }
-
-extern "C" {
-
-// Putting semicolons here to make clang-format not reformat this into
-// a stair shape.
-DEF_nac3_int_exp_(int32_t);
-DEF_nac3_int_exp_(int64_t);
-DEF_nac3_int_exp_(uint32_t);
-DEF_nac3_int_exp_(uint64_t);
-
-int32_t __nac3_isinf(double x) {
-    return __builtin_isinf(x);
-}
-
-int32_t __nac3_isnan(double x) {
-    return __builtin_isnan(x);
-}
-
-double tgamma(double arg);
-
-double __nac3_gamma(double z) {
-    // Handling for denormals
-    //     | x                 | Python gamma(x) | C tgamma(x) |
-    // --- | ----------------- | --------------- | ----------- |
-    // (1) | nan               | nan             | nan         |
-    // (2) | -inf              | -inf            | inf         |
-    // (3) | inf               | inf             | inf         |
-    // (4) | 0.0               | inf             | inf         |
-    // (5) | {-1.0, -2.0, ...} | inf             | nan         |
-
-    // (1)-(3)
-    if (__builtin_isinf(z) || __builtin_isnan(z)) {
-        return z;
-    }
-
-    double v = tgamma(z);
-
-    // (4)-(5)
-    return __builtin_isinf(v) || __builtin_isnan(v) ? __builtin_inf() : v;
-}
-
-double lgamma(double arg);
-
-double __nac3_gammaln(double x) {
-    // libm's handling of value overflows differs from scipy:
-    // - scipy: gammaln(-inf) -> -inf
-    // - libm : lgamma(-inf) -> inf
-
-    if (__builtin_isinf(x)) {
-        return x;
-    }
-
-    return lgamma(x);
-}
-
-double j0(double x);
-
-double __nac3_j0(double x) {
-    // libm's handling of value overflows differs from scipy:
-    // - scipy: j0(inf) -> nan
-    // - libm : j0(inf) -> 0.0
-
-    if (__builtin_isinf(x)) {
-        return __builtin_nan("");
-    }
-
-    return j0(x);
-}
-}  // namespace

nac3core/irrt/irrt/math_util.hpp

@@ -1,13 +0,0 @@
-#pragma once
-
-namespace {
-template<typename T>
-const T& max(const T& a, const T& b) {
-    return a > b ? a : b;
-}
-
-template<typename T>
-const T& min(const T& a, const T& b) {
-    return a > b ? b : a;
-}
-}  // namespace

nac3core/irrt/irrt/ndarray.hpp

@@ -1,151 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-
-// TODO: To be deleted since NDArray with strides is done.
-
-namespace {
-template<typename SizeT>
-SizeT __nac3_ndarray_calc_size_impl(const SizeT* list_data, SizeT list_len, SizeT begin_idx, SizeT end_idx) {
-    __builtin_assume(end_idx <= list_len);
-
-    SizeT num_elems = 1;
-    for (SizeT i = begin_idx; i < end_idx; ++i) {
-        SizeT val = list_data[i];
-        __builtin_assume(val > 0);
-        num_elems *= val;
-    }
-    return num_elems;
-}
-
-template<typename SizeT>
-void __nac3_ndarray_calc_nd_indices_impl(SizeT index, const SizeT* dims, SizeT num_dims, NDIndexInt* idxs) {
-    SizeT stride = 1;
-    for (SizeT dim = 0; dim < num_dims; dim++) {
-        SizeT i = num_dims - dim - 1;
-        __builtin_assume(dims[i] > 0);
-        idxs[i] = (index / stride) % dims[i];
-        stride *= dims[i];
-    }
-}
-
-template<typename SizeT>
-SizeT __nac3_ndarray_flatten_index_impl(const SizeT* dims,
-                                        SizeT num_dims,
-                                        const NDIndexInt* indices,
-                                        SizeT num_indices) {
-    SizeT idx = 0;
-    SizeT stride = 1;
-    for (SizeT i = 0; i < num_dims; ++i) {
-        SizeT ri = num_dims - i - 1;
-        if (ri < num_indices) {
-            idx += stride * indices[ri];
-        }
-
-        __builtin_assume(dims[i] > 0);
-        stride *= dims[ri];
-    }
-    return idx;
-}
-
-template<typename SizeT>
-void __nac3_ndarray_calc_broadcast_impl(const SizeT* lhs_dims,
-                                        SizeT lhs_ndims,
-                                        const SizeT* rhs_dims,
-                                        SizeT rhs_ndims,
-                                        SizeT* out_dims) {
-    SizeT max_ndims = lhs_ndims > rhs_ndims ? lhs_ndims : rhs_ndims;
-
-    for (SizeT i = 0; i < max_ndims; ++i) {
-        const SizeT* lhs_dim_sz = i < lhs_ndims ? &lhs_dims[lhs_ndims - i - 1] : nullptr;
-        const SizeT* rhs_dim_sz = i < rhs_ndims ? &rhs_dims[rhs_ndims - i - 1] : nullptr;
-        SizeT* out_dim = &out_dims[max_ndims - i - 1];
-
-        if (lhs_dim_sz == nullptr) {
-            *out_dim = *rhs_dim_sz;
-        } else if (rhs_dim_sz == nullptr) {
-            *out_dim = *lhs_dim_sz;
-        } else if (*lhs_dim_sz == 1) {
-            *out_dim = *rhs_dim_sz;
-        } else if (*rhs_dim_sz == 1) {
-            *out_dim = *lhs_dim_sz;
-        } else if (*lhs_dim_sz == *rhs_dim_sz) {
-            *out_dim = *lhs_dim_sz;
-        } else {
-            __builtin_unreachable();
-        }
-    }
-}
-
-template<typename SizeT>
-void __nac3_ndarray_calc_broadcast_idx_impl(const SizeT* src_dims,
-                                            SizeT src_ndims,
-                                            const NDIndexInt* in_idx,
-                                            NDIndexInt* out_idx) {
-    for (SizeT i = 0; i < src_ndims; ++i) {
-        SizeT src_i = src_ndims - i - 1;
-        out_idx[src_i] = src_dims[src_i] == 1 ? 0 : in_idx[src_i];
-    }
-}
-}  // namespace
-
-extern "C" {
-uint32_t __nac3_ndarray_calc_size(const uint32_t* list_data, uint32_t list_len, uint32_t begin_idx, uint32_t end_idx) {
-    return __nac3_ndarray_calc_size_impl(list_data, list_len, begin_idx, end_idx);
-}
-
-uint64_t
-__nac3_ndarray_calc_size64(const uint64_t* list_data, uint64_t list_len, uint64_t begin_idx, uint64_t end_idx) {
-    return __nac3_ndarray_calc_size_impl(list_data, list_len, begin_idx, end_idx);
-}
-
-void __nac3_ndarray_calc_nd_indices(uint32_t index, const uint32_t* dims, uint32_t num_dims, NDIndexInt* idxs) {
-    __nac3_ndarray_calc_nd_indices_impl(index, dims, num_dims, idxs);
-}
-
-void __nac3_ndarray_calc_nd_indices64(uint64_t index, const uint64_t* dims, uint64_t num_dims, NDIndexInt* idxs) {
-    __nac3_ndarray_calc_nd_indices_impl(index, dims, num_dims, idxs);
-}
-
-uint32_t
-__nac3_ndarray_flatten_index(const uint32_t* dims, uint32_t num_dims, const NDIndexInt* indices, uint32_t num_indices) {
-    return __nac3_ndarray_flatten_index_impl(dims, num_dims, indices, num_indices);
-}
-
-uint64_t __nac3_ndarray_flatten_index64(const uint64_t* dims,
-                                        uint64_t num_dims,
-                                        const NDIndexInt* indices,
-                                        uint64_t num_indices) {
-    return __nac3_ndarray_flatten_index_impl(dims, num_dims, indices, num_indices);
-}
-
-void __nac3_ndarray_calc_broadcast(const uint32_t* lhs_dims,
-                                   uint32_t lhs_ndims,
-                                   const uint32_t* rhs_dims,
-                                   uint32_t rhs_ndims,
-                                   uint32_t* out_dims) {
-    return __nac3_ndarray_calc_broadcast_impl(lhs_dims, lhs_ndims, rhs_dims, rhs_ndims, out_dims);
-}
-
-void __nac3_ndarray_calc_broadcast64(const uint64_t* lhs_dims,
-                                     uint64_t lhs_ndims,
-                                     const uint64_t* rhs_dims,
-                                     uint64_t rhs_ndims,
-                                     uint64_t* out_dims) {
-    return __nac3_ndarray_calc_broadcast_impl(lhs_dims, lhs_ndims, rhs_dims, rhs_ndims, out_dims);
-}
-
-void __nac3_ndarray_calc_broadcast_idx(const uint32_t* src_dims,
-                                       uint32_t src_ndims,
-                                       const NDIndexInt* in_idx,
-                                       NDIndexInt* out_idx) {
-    __nac3_ndarray_calc_broadcast_idx_impl(src_dims, src_ndims, in_idx, out_idx);
-}
-
-void __nac3_ndarray_calc_broadcast_idx64(const uint64_t* src_dims,
-                                         uint64_t src_ndims,
-                                         const NDIndexInt* in_idx,
-                                         NDIndexInt* out_idx) {
-    __nac3_ndarray_calc_broadcast_idx_impl(src_dims, src_ndims, in_idx, out_idx);
-}
-}  // namespace

nac3core/irrt/irrt/ndarray/basic.hpp

@@ -1,20 +1,20 @@
 #pragma once
 
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
+#include <irrt/exception.hpp>
+#include <irrt/int_defs.hpp>
+#include <irrt/ndarray/def.hpp>
 
 namespace {
 namespace ndarray {
 namespace basic {
+namespace util {
 /**
- * @brief Assert that `shape` does not contain negative dimensions.
+ * @brief Asserts that `shape` does not contain negative dimensions.
  *
  * @param ndims Number of dimensions in `shape`
  * @param shape The shape to check on
  */
-template<typename SizeT>
+template <typename SizeT>
 void assert_shape_no_negative(SizeT ndims, const SizeT* shape) {
     for (SizeT axis = 0; axis < ndims; axis++) {
         if (shape[axis] < 0) {
@@ -27,41 +27,15 @@ void assert_shape_no_negative(SizeT ndims, const SizeT* shape) {
 }
 
 /**
- * @brief Assert that two shapes are the same in the context of writing output to an ndarray.
- */
-template<typename SizeT>
-void assert_output_shape_same(SizeT ndarray_ndims,
-                              const SizeT* ndarray_shape,
-                              SizeT output_ndims,
-                              const SizeT* output_shape) {
-    if (ndarray_ndims != output_ndims) {
-        // There is no corresponding NumPy error message like this.
-        raise_exception(SizeT, EXN_VALUE_ERROR, "Cannot write output of ndims {0} to an ndarray with ndims {1}",
-                        output_ndims, ndarray_ndims, NO_PARAM);
-    }
-
-    for (SizeT axis = 0; axis < ndarray_ndims; axis++) {
-        if (ndarray_shape[axis] != output_shape[axis]) {
-            // There is no corresponding NumPy error message like this.
-            raise_exception(SizeT, EXN_VALUE_ERROR,
-                            "Mismatched dimensions on axis {0}, output has "
-                            "dimension {1}, but destination ndarray has dimension {2}.",
-                            axis, output_shape[axis], ndarray_shape[axis]);
-        }
-    }
-}
-
-/**
- * @brief Return the number of elements of an ndarray given its shape.
+ * @brief Returns the number of elements of an ndarray given its shape.
  *
  * @param ndims Number of dimensions in `shape`
  * @param shape The shape of the ndarray
  */
-template<typename SizeT>
+template <typename SizeT>
 SizeT calc_size_from_shape(SizeT ndims, const SizeT* shape) {
     SizeT size = 1;
-    for (SizeT axis = 0; axis < ndims; axis++)
-        size *= shape[axis];
+    for (SizeT axis = 0; axis < ndims; axis++) size *= shape[axis];
     return size;
 }
 
@@ -73,8 +47,9 @@ SizeT calc_size_from_shape(SizeT ndims, const SizeT* shape) {
  * @param indices The returned indices indexing the ndarray with shape `shape`.
  * @param nth The index of the element of interest.
  */
-template<typename SizeT>
-void set_indices_by_nth(SizeT ndims, const SizeT* shape, SizeT* indices, SizeT nth) {
+template <typename SizeT>
+void set_indices_by_nth(SizeT ndims, const SizeT* shape, SizeT* indices,
+                        SizeT nth) {
     for (SizeT i = 0; i < ndims; i++) {
         SizeT axis = ndims - i - 1;
         SizeT dim = shape[axis];
@@ -83,15 +58,16 @@ void set_indices_by_nth(SizeT ndims, const SizeT* shape, SizeT* indices, SizeT n
         nth /= dim;
     }
 }
+}  // namespace util
 
 /**
  * @brief Return the number of elements of an `ndarray`
  *
  * This function corresponds to `<an_ndarray>.size`
  */
-template<typename SizeT>
+template <typename SizeT>
 SizeT size(const NDArray<SizeT>* ndarray) {
-    return calc_size_from_shape(ndarray->ndims, ndarray->shape);
+    return util::calc_size_from_shape(ndarray->ndims, ndarray->shape);
 }
 
 /**
@@ -99,7 +75,7 @@ SizeT size(const NDArray<SizeT>* ndarray) {
  *
  * This function corresponds to `<an_ndarray>.nbytes`.
  */
-template<typename SizeT>
+template <typename SizeT>
 SizeT nbytes(const NDArray<SizeT>* ndarray) {
     return size(ndarray) * ndarray->itemsize;
 }
@@ -109,37 +85,32 @@ SizeT nbytes(const NDArray<SizeT>* ndarray) {
  *
  * This function corresponds to `<an_ndarray>.__len__`.
  *
- * @param dst_length The length.
+ * @param dst_length The returned result
  */
-template<typename SizeT>
+template <typename SizeT>
 SizeT len(const NDArray<SizeT>* ndarray) {
-    if (ndarray->ndims != 0) {
+    // numpy prohibits `__len__` on unsized objects
+    if (ndarray->ndims == 0) {
+        raise_exception(SizeT, EXN_TYPE_ERROR, "len() of unsized object",
+                        NO_PARAM, NO_PARAM, NO_PARAM);
+    } else {
         return ndarray->shape[0];
     }
-
-    // numpy prohibits `__len__` on unsized objects
-    raise_exception(SizeT, EXN_TYPE_ERROR, "len() of unsized object", NO_PARAM, NO_PARAM, NO_PARAM);
-    __builtin_unreachable();
 }
 
 /**
  * @brief Return a boolean indicating if `ndarray` is (C-)contiguous.
  *
- * You may want to see ndarray's rules for C-contiguity:
- * https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45
+ * You may want to see: ndarray's rules for C-contiguity: https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45
  */
-template<typename SizeT>
+template <typename SizeT>
 bool is_c_contiguous(const NDArray<SizeT>* ndarray) {
-    // References:
-    // - tinynumpy's implementation:
-    // https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L102
-    // - ndarray's flags["C_CONTIGUOUS"]:
-    // https://numpy.org/doc/stable/reference/generated/numpy.ndarray.flags.html#numpy.ndarray.flags
-    // - ndarray's rules for C-contiguity:
-    // https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45
+    // Other references:
+    // - tinynumpy's implementation: https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L102
+    // - ndarray's flags["C_CONTIGUOUS"]: https://numpy.org/doc/stable/reference/generated/numpy.ndarray.flags.html#numpy.ndarray.flags
+    // - ndarray's rules for C-contiguity: https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45
 
-    // From
-    // https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45:
+    // From https://github.com/numpy/numpy/blob/df256d0d2f3bc6833699529824781c58f9c6e697/numpy/core/src/multiarray/flagsobject.c#L95C1-L99C45:
     //
     // The traditional rule is that for an array to be flagged as C contiguous,
     // the following must hold:
@@ -165,7 +136,8 @@ bool is_c_contiguous(const NDArray<SizeT>* ndarray) {
 
     for (SizeT i = 1; i < ndarray->ndims; i++) {
         SizeT axis_i = ndarray->ndims - i - 1;
-        if (ndarray->strides[axis_i] != ndarray->shape[axis_i + 1] * ndarray->strides[axis_i + 1]) {
+        if (ndarray->strides[axis_i] !=
+            ndarray->shape[axis_i + 1] + ndarray->strides[axis_i + 1]) {
             return false;
         }
     }
@@ -174,45 +146,47 @@ bool is_c_contiguous(const NDArray<SizeT>* ndarray) {
 }
 
 /**
- * @brief Return the pointer to the element indexed by `indices` along the ndarray's axes.
- *
- * This function does no bound check.
+ * @brief Return the pointer to the element indexed by `indices`.
  */
-template<typename SizeT>
-void* get_pelement_by_indices(const NDArray<SizeT>* ndarray, const SizeT* indices) {
-    void* element = ndarray->data;
+template <typename SizeT>
+uint8_t* get_pelement_by_indices(const NDArray<SizeT>* ndarray,
+                                 const SizeT* indices) {
+    uint8_t* element = ndarray->data;
     for (SizeT dim_i = 0; dim_i < ndarray->ndims; dim_i++)
-        element = static_cast<uint8_t*>(element) + indices[dim_i] * ndarray->strides[dim_i];
+        element += indices[dim_i] * ndarray->strides[dim_i];
     return element;
 }
 
+int counter = 0;
+
 /**
- * @brief Return the pointer to the nth (0-based) element of `ndarray` in flattened view.
+ * @brief Return the pointer to the nth (0-based) element in a flattened view of `ndarray`.
  *
  * This function does no bound check.
  */
-template<typename SizeT>
-void* get_nth_pelement(const NDArray<SizeT>* ndarray, SizeT nth) {
-    void* element = ndarray->data;
+template <typename SizeT>
+uint8_t* get_nth_pelement(const NDArray<SizeT>* ndarray, SizeT nth) {
+    uint8_t* element = ndarray->data;
     for (SizeT i = 0; i < ndarray->ndims; i++) {
         SizeT axis = ndarray->ndims - i - 1;
         SizeT dim = ndarray->shape[axis];
-        element = static_cast<uint8_t*>(element) + ndarray->strides[axis] * (nth % dim);
+        element += ndarray->strides[axis] * (nth % dim);
         nth /= dim;
     }
     return element;
 }
 
 /**
- * @brief Update the strides of an ndarray given an ndarray `shape` to be contiguous.
+ * @brief Update the strides of an ndarray given an ndarray `shape`
+ * and assuming that the ndarray is fully c-contagious.
  *
  * You might want to read https://ajcr.net/stride-guide-part-1/.
  */
-template<typename SizeT>
+template <typename SizeT>
 void set_strides_by_shape(NDArray<SizeT>* ndarray) {
     SizeT stride_product = 1;
     for (SizeT i = 0; i < ndarray->ndims; i++) {
-        SizeT axis = ndarray->ndims - i - 1;
+        int axis = ndarray->ndims - i - 1;
         ndarray->strides[axis] = stride_product * ndarray->itemsize;
         stride_product *= ndarray->shape[axis];
     }
@@ -224,8 +198,9 @@ void set_strides_by_shape(NDArray<SizeT>* ndarray) {
  * @param pelement Pointer to the element in `ndarray` to be set.
  * @param pvalue Pointer to the value `pelement` will be set to.
  */
-template<typename SizeT>
-void set_pelement_value(NDArray<SizeT>* ndarray, void* pelement, const void* pvalue) {
+template <typename SizeT>
+void set_pelement_value(NDArray<SizeT>* ndarray, uint8_t* pelement,
+                        const uint8_t* pvalue) {
     __builtin_memcpy(pelement, pvalue, ndarray->itemsize);
 }
 
@@ -234,17 +209,17 @@ void set_pelement_value(NDArray<SizeT>* ndarray, void* pelement, const void* pva
  *
  * Both ndarrays will be viewed in their flatten views when copying the elements.
  */
-template<typename SizeT>
+template <typename SizeT>
 void copy_data(const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
-    // TODO: Make this faster with memcpy when we see a contiguous segment.
-    // TODO: Handle overlapping.
+    // TODO: Make this faster with memcpy
 
-    debug_assert_eq(SizeT, src_ndarray->itemsize, dst_ndarray->itemsize);
+    __builtin_assume(src_ndarray->itemsize == dst_ndarray->itemsize);
 
     for (SizeT i = 0; i < size(src_ndarray); i++) {
         auto src_element = ndarray::basic::get_nth_pelement(src_ndarray, i);
         auto dst_element = ndarray::basic::get_nth_pelement(dst_ndarray, i);
-        ndarray::basic::set_pelement_value(dst_ndarray, dst_element, src_element);
+        ndarray::basic::set_pelement_value(dst_ndarray, dst_element,
+                                           src_element);
     }
 }
 }  // namespace basic
@@ -254,26 +229,14 @@ void copy_data(const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
 extern "C" {
 using namespace ndarray::basic;
 
-void __nac3_ndarray_util_assert_shape_no_negative(int32_t ndims, int32_t* shape) {
-    assert_shape_no_negative(ndims, shape);
+void __nac3_ndarray_util_assert_shape_no_negative(int32_t ndims,
+                                                  int32_t* shape) {
+    util::assert_shape_no_negative(ndims, shape);
 }
 
-void __nac3_ndarray_util_assert_shape_no_negative64(int64_t ndims, int64_t* shape) {
-    assert_shape_no_negative(ndims, shape);
-}
-
-void __nac3_ndarray_util_assert_output_shape_same(int32_t ndarray_ndims,
-                                                  const int32_t* ndarray_shape,
-                                                  int32_t output_ndims,
-                                                  const int32_t* output_shape) {
-    assert_output_shape_same(ndarray_ndims, ndarray_shape, output_ndims, output_shape);
-}
-
-void __nac3_ndarray_util_assert_output_shape_same64(int64_t ndarray_ndims,
-                                                    const int64_t* ndarray_shape,
-                                                    int64_t output_ndims,
-                                                    const int64_t* output_shape) {
-    assert_output_shape_same(ndarray_ndims, ndarray_shape, output_ndims, output_shape);
+void __nac3_ndarray_util_assert_shape_no_negative64(int64_t ndims,
+                                                    int64_t* shape) {
+    util::assert_shape_no_negative(ndims, shape);
 }
 
 uint32_t __nac3_ndarray_size(NDArray<int32_t>* ndarray) {
@@ -292,13 +255,9 @@ uint64_t __nac3_ndarray_nbytes64(NDArray<int64_t>* ndarray) {
     return nbytes(ndarray);
 }
 
-int32_t __nac3_ndarray_len(NDArray<int32_t>* ndarray) {
-    return len(ndarray);
-}
+int32_t __nac3_ndarray_len(NDArray<int32_t>* ndarray) { return len(ndarray); }
 
-int64_t __nac3_ndarray_len64(NDArray<int64_t>* ndarray) {
-    return len(ndarray);
-}
+int64_t __nac3_ndarray_len64(NDArray<int64_t>* ndarray) { return len(ndarray); }
 
 bool __nac3_ndarray_is_c_contiguous(NDArray<int32_t>* ndarray) {
     return is_c_contiguous(ndarray);
@@ -308,22 +267,16 @@ bool __nac3_ndarray_is_c_contiguous64(NDArray<int64_t>* ndarray) {
     return is_c_contiguous(ndarray);
 }
 
-void* __nac3_ndarray_get_nth_pelement(const NDArray<int32_t>* ndarray, int32_t nth) {
+uint8_t* __nac3_ndarray_get_nth_pelement(const NDArray<int32_t>* ndarray,
+                                         int32_t nth) {
     return get_nth_pelement(ndarray, nth);
 }
 
-void* __nac3_ndarray_get_nth_pelement64(const NDArray<int64_t>* ndarray, int64_t nth) {
+uint8_t* __nac3_ndarray_get_nth_pelement64(const NDArray<int64_t>* ndarray,
+                                           int64_t nth) {
     return get_nth_pelement(ndarray, nth);
 }
 
-void* __nac3_ndarray_get_pelement_by_indices(const NDArray<int32_t>* ndarray, int32_t* indices) {
-    return get_pelement_by_indices(ndarray, indices);
-}
-
-void* __nac3_ndarray_get_pelement_by_indices64(const NDArray<int64_t>* ndarray, int64_t* indices) {
-    return get_pelement_by_indices(ndarray, indices);
-}
-
 void __nac3_ndarray_set_strides_by_shape(NDArray<int32_t>* ndarray) {
     set_strides_by_shape(ndarray);
 }
@@ -332,11 +285,13 @@ void __nac3_ndarray_set_strides_by_shape64(NDArray<int64_t>* ndarray) {
     set_strides_by_shape(ndarray);
 }
 
-void __nac3_ndarray_copy_data(NDArray<int32_t>* src_ndarray, NDArray<int32_t>* dst_ndarray) {
+void __nac3_ndarray_copy_data(NDArray<int32_t>* src_ndarray,
+                              NDArray<int32_t>* dst_ndarray) {
     copy_data(src_ndarray, dst_ndarray);
 }
 
-void __nac3_ndarray_copy_data64(NDArray<int64_t>* src_ndarray, NDArray<int64_t>* dst_ndarray) {
+void __nac3_ndarray_copy_data64(NDArray<int64_t>* src_ndarray,
+                                NDArray<int64_t>* dst_ndarray) {
     copy_data(src_ndarray, dst_ndarray);
 }
 }

nac3core/irrt/irrt/ndarray/broadcast.hpp

@@ -0,0 +1,157 @@
+#pragma once
+
+#include <irrt/int_defs.hpp>
+#include <irrt/ndarray/def.hpp>
+#include <irrt/slice.hpp>
+
+namespace {
+template <typename SizeT>
+struct ShapeEntry {
+    SizeT ndims;
+    SizeT* shape;
+};
+}  // namespace
+
+namespace {
+namespace ndarray {
+namespace broadcast {
+namespace util {
+/**
+ * @brief Return true if `src_shape` can broadcast to `dst_shape`.
+ *
+ * See https://numpy.org/doc/stable/user/basics.broadcasting.html
+ */
+template <typename SizeT>
+bool can_broadcast_shape_to(SizeT target_ndims, const SizeT* target_shape,
+                            SizeT src_ndims, const SizeT* src_shape) {
+    if (src_ndims > target_ndims) {
+        return false;
+    }
+
+    for (SizeT i = 0; i < src_ndims; i++) {
+        SizeT target_dim = target_shape[target_ndims - i - 1];
+        SizeT src_dim = src_shape[src_ndims - i - 1];
+        if (!(src_dim == 1 || target_dim == src_dim)) {
+            return false;
+        }
+    }
+    return true;
+}
+
+/**
+ * @brief Performs `np.broadcast_shapes`
+ */
+template <typename SizeT>
+void broadcast_shapes(SizeT num_shapes, const ShapeEntry<SizeT>* shapes,
+                      SizeT dst_ndims, SizeT* dst_shape) {
+    // `dst_ndims` must be `max([shape.ndims for shape in shapes])`, but the caller has to calculate it/provide it
+    //   for this function since they should already know in order to allocate `dst_shape` in the first place.
+    // `dst_shape` must be pre-allocated.
+    // `dst_shape` does not have to be initialized
+    for (SizeT dst_axis = 0; dst_axis < dst_ndims; dst_axis++) {
+        dst_shape[dst_axis] = 1;
+    }
+
+    for (SizeT i = 0; i < num_shapes; i++) {
+        ShapeEntry<SizeT> entry = shapes[i];
+
+        for (SizeT j = 0; j < entry.ndims; j++) {
+            SizeT entry_axis = entry.ndims - j - 1;
+            SizeT dst_axis = dst_ndims - j - 1;
+
+            SizeT entry_dim = entry.shape[entry_axis];
+            SizeT dst_dim = dst_shape[dst_axis];
+
+            if (dst_dim == 1) {
+                dst_shape[dst_axis] = entry_dim;
+            } else if (entry_dim == 1 || entry_dim == dst_dim) {
+                // Do nothing
+            } else {
+                raise_exception(SizeT, EXN_VALUE_ERROR,
+                                "shape mismatch: objects cannot be broadcast "
+                                "to a single shape.",
+                                NO_PARAM, NO_PARAM, NO_PARAM);
+            }
+        }
+    }
+}
+}  // namespace util
+
+/**
+ * @brief Perform `np.broadcast_to(<ndarray>, <target_shape>)` and appropriate assertions.
+ *
+ * This function attempts to broadcast `src_ndarray` to a new shape defined by `dst_ndarray.shape`,
+ * and return the result by modifying `dst_ndarray`.
+ *
+ * # Notes on `dst_ndarray`
+ * The caller is responsible for allocating space for the resulting ndarray.
+ * Here is what this function expects from `dst_ndarray` when called:
+ *   - `dst_ndarray->data` does not have to be initialized.
+ *   - `dst_ndarray->itemsize` does not have to be initialized.
+ *   - `dst_ndarray->ndims` must be initialized, determining the length of `dst_ndarray->shape`
+ *   - `dst_ndarray->shape` must be allocated, and must contain the desired target broadcast shape.
+ *   - `dst_ndarray->strides` must be allocated, through it can contain uninitialized values.
+ * When this function call ends:
+ *   - `dst_ndarray->data` is set to `src_ndarray->data` (`dst_ndarray` is just a view to `src_ndarray`)
+ *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`
+ *   - `dst_ndarray->ndims` is unchanged.
+ *   - `dst_ndarray->shape` is unchanged.
+ *   - `dst_ndarray->strides` is updated accordingly by how ndarray broadcast_to works.
+ */
+template <typename SizeT>
+void broadcast_to(const NDArray<SizeT>* src_ndarray,
+                  NDArray<SizeT>* dst_ndarray) {
+    if (!ndarray::broadcast::util::can_broadcast_shape_to(
+            dst_ndarray->ndims, dst_ndarray->shape, src_ndarray->ndims,
+            src_ndarray->shape)) {
+        raise_exception(SizeT, EXN_VALUE_ERROR,
+                        "operands could not be broadcast together",
+                        dst_ndarray->shape[0], src_ndarray->shape[0], NO_PARAM);
+    }
+
+    dst_ndarray->data = src_ndarray->data;
+    dst_ndarray->itemsize = src_ndarray->itemsize;
+
+    for (SizeT i = 0; i < dst_ndarray->ndims; i++) {
+        SizeT src_axis = src_ndarray->ndims - i - 1;
+        SizeT dst_axis = dst_ndarray->ndims - i - 1;
+        if (src_axis < 0 || (src_ndarray->shape[src_axis] == 1 &&
+                             dst_ndarray->shape[dst_axis] != 1)) {
+            // Freeze the steps in-place
+            dst_ndarray->strides[dst_axis] = 0;
+        } else {
+            dst_ndarray->strides[dst_axis] = src_ndarray->strides[src_axis];
+        }
+    }
+}
+}  // namespace broadcast
+}  // namespace ndarray
+}  // namespace
+
+extern "C" {
+using namespace ndarray::broadcast;
+
+void __nac3_ndarray_broadcast_to(NDArray<int32_t>* src_ndarray,
+                                 NDArray<int32_t>* dst_ndarray) {
+    broadcast_to(src_ndarray, dst_ndarray);
+}
+
+void __nac3_ndarray_broadcast_to64(NDArray<int64_t>* src_ndarray,
+                                   NDArray<int64_t>* dst_ndarray) {
+    broadcast_to(src_ndarray, dst_ndarray);
+}
+
+void __nac3_ndarray_broadcast_shapes(int32_t num_shapes,
+                                     const ShapeEntry<int32_t>* shapes,
+                                     int32_t dst_ndims, int32_t* dst_shape) {
+    ndarray::broadcast::util::broadcast_shapes(num_shapes, shapes, dst_ndims,
+                                               dst_shape);
+}
+
+void __nac3_ndarray_broadcast_shapes64(int64_t num_shapes,
+                                       const ShapeEntry<int64_t>* shapes,
+                                       int64_t dst_ndims, int64_t* dst_shape) {
+    ndarray::broadcast::util::broadcast_shapes(num_shapes, shapes, dst_ndims,
+                                               dst_shape);
+}
+}

nac3core/irrt/irrt/ndarray/def.hpp

@@ -1,22 +1,20 @@
 #pragma once
 
-#include "irrt/int_types.hpp"
-
 namespace {
 /**
  * @brief The NDArray object
  *
- * Official numpy implementation:
- * https://github.com/numpy/numpy/blob/735a477f0bc2b5b84d0e72d92f224bde78d4e069/doc/source/reference/c-api/types-and-structures.rst#pyarrayinterface
- *
- * Note that this implementation is based on `PyArrayInterface` rather of `PyArrayObject`. The
- * difference between `PyArrayInterface` and `PyArrayObject` (relevant to our implementation) is
- * that `PyArrayInterface` *has* `itemsize` and uses `void*` for its `data`, whereas `PyArrayObject`
- * does not require `itemsize` (probably using `strides[-1]` instead) and uses `char*` for its
- * `data`. There are also minor differences in the struct layout.
+ * The official numpy implementations: https://github.com/numpy/numpy/blob/735a477f0bc2b5b84d0e72d92f224bde78d4e069/doc/source/reference/c-api/types-and-structures.rst
  */
-template<typename SizeT>
+template <typename SizeT>
 struct NDArray {
+    /**
+     * @brief The underlying data this `ndarray` is pointing to.
+     *
+     * Must be set to `nullptr` to indicate that this NDArray's `data` is uninitialized.
+     */
+    uint8_t* data;
+
     /**
      * @brief The number of bytes of a single element in `data`.
      */
@@ -39,13 +37,8 @@ struct NDArray {
      *
      * The stride values are in units of bytes, not number of elements.
      *
-     * Note that `strides` can have negative values or contain 0.
+     * Note that `strides` can have negative values.
      */
     SizeT* strides;
-
-    /**
-     * @brief The underlying data this `ndarray` is pointing to.
-     */
-    void* data;
 };
 }  // namespace

nac3core/irrt/irrt/ndarray/indexing.hpp

@@ -1,11 +1,10 @@
 #pragma once
 
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/basic.hpp"
-#include "irrt/ndarray/def.hpp"
-#include "irrt/range.hpp"
-#include "irrt/slice.hpp"
+#include <irrt/exception.hpp>
+#include <irrt/int_defs.hpp>
+#include <irrt/ndarray/basic.hpp>
+#include <irrt/ndarray/def.hpp>
+#include <irrt/slice.hpp>
 
 namespace {
 typedef uint8_t NDIndexType;
@@ -13,14 +12,14 @@ typedef uint8_t NDIndexType;
 /**
  * @brief A single element index
  *
- * `data` points to a `int32_t`.
+ * `data` points to a `SliceIndex`.
  */
-const NDIndexType ND_INDEX_TYPE_SINGLE_ELEMENT = 0;
 
+const NDIndexType ND_INDEX_TYPE_SINGLE_ELEMENT = 0;
 /**
  * @brief A slice index
  *
- * `data` points to a `Slice<int32_t>`.
+ * `data` points to a `UserRange`.
  */
 const NDIndexType ND_INDEX_TYPE_SLICE = 1;
 
@@ -40,25 +39,19 @@ const NDIndexType ND_INDEX_TYPE_ELLIPSIS = 3;
 
 /**
  * @brief An index used in ndarray indexing
- *
- * That is:
- * ```
- * my_ndarray[::-1, 3, ..., np.newaxis]
- *            ^^^^  ^  ^^^  ^^^^^^^^^^ each of these is represented by an NDIndex.
- * ```
  */
 struct NDIndex {
     /**
      * @brief Enum tag to specify the type of index.
      *
-     * Please see the comment of each enum constant.
+     * Please see comments of each enum constant.
      */
     NDIndexType type;
 
     /**
      * @brief The accompanying data associated with `type`.
      *
-     * Please see the comment of each enum constant.
+     * Please see comments of each enum constant.
      */
     uint8_t* data;
 };
@@ -67,12 +60,62 @@ struct NDIndex {
 namespace {
 namespace ndarray {
 namespace indexing {
+namespace util {
+
+/**
+ * @brief Return the expected rank of the resulting ndarray
+ * created by indexing an ndarray of rank `ndims` using `indexes`.
+ *
+ * An IndexError is raised if the indexes are invalid.
+ */
+template <typename SizeT>
+SizeT validate_and_deduce_ndims_after_indexing(SizeT ndims, SizeT num_indexes,
+                                               const NDIndex* indexes) {
+    // There may be ellipsis `...` in `indexes`. There can only be 0 or 1 ellipsis.
+    SizeT num_ellipsis = 0;
+
+    for (SizeT i = 0; i < num_indexes; i++) {
+        if (indexes[i].type == ND_INDEX_TYPE_SINGLE_ELEMENT) {
+            ndims--;
+        } else if (indexes[i].type == ND_INDEX_TYPE_SLICE) {
+            // Nothing
+        } else if (indexes[i].type == ND_INDEX_TYPE_NEWAXIS) {
+            ndims++;
+        } else if (indexes[i].type == ND_INDEX_TYPE_ELLIPSIS) {
+            // `...` doesn't do anything to `ndims`.
+            num_ellipsis++;
+            if (num_ellipsis > 1) {
+                raise_exception(
+                    SizeT, EXN_INDEX_ERROR,
+                    "an index can only have a single ellipsis ('...')",
+                    NO_PARAM, NO_PARAM, NO_PARAM);
+            }
+        } else {
+            __builtin_unreachable();
+        }
+    }
+
+    if (num_indexes - num_ellipsis > ndims) {
+        raise_exception(SizeT, EXN_INDEX_ERROR,
+                        "too many indices for array: array is {0}-dimensional, "
+                        "but {1} were indexed",
+                        ndims, num_indexes, NO_PARAM);
+    }
+
+    return ndims;
+}
+}  // namespace util
+
 /**
  * @brief Perform ndarray "basic indexing" (https://numpy.org/doc/stable/user/basics.indexing.html#basic-indexing)
  *
- * This function is very similar to performing `dst_ndarray = src_ndarray[indices]` in Python.
+ * This is function very similar to performing `dst_ndarray = src_ndarray[indexes]` in Python (where the variables
+ * can all be found in the parameter of this function).
  *
- * This function also does proper assertions on `indices` to check for out of bounds access and more.
+ * In other words, this function takes in an ndarray (`src_ndarray`), index it with `indexes`, and return the
+ * indexed array (by writing the result to `dst_ndarray`).
+ *
+ * This function also does proper assertions on `indexes`.
  *
  * # Notes on `dst_ndarray`
  * The caller is responsible for allocating space for the resulting ndarray.
@@ -80,91 +123,104 @@ namespace indexing {
  *   - `dst_ndarray->data` does not have to be initialized.
  *   - `dst_ndarray->itemsize` does not have to be initialized.
  *   - `dst_ndarray->ndims` must be initialized, and it must be equal to the expected `ndims` of the `dst_ndarray` after
- *       indexing `src_ndarray` with `indices`.
+ *       indexing `src_ndarray` with `indexes`.
  *   - `dst_ndarray->shape` must be allocated, through it can contain uninitialized values.
  *   - `dst_ndarray->strides` must be allocated, through it can contain uninitialized values.
  * When this function call ends:
- *   - `dst_ndarray->data` is set to `src_ndarray->data`.
- *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`.
+ *   - `dst_ndarray->data` is set to `src_ndarray->data` (`dst_ndarray` is just a view to `src_ndarray`)
+ *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`
  *   - `dst_ndarray->ndims` is unchanged.
  *   - `dst_ndarray->shape` is updated according to how `src_ndarray` is indexed.
  *   - `dst_ndarray->strides` is updated accordingly by how ndarray indexing works.
  *
- * @param indices indices to index `src_ndarray`, ordered in the same way you would write them in Python.
+ * @param indexes Indexes to index `src_ndarray`, ordered in the same way you would write them in Python.
  * @param src_ndarray The NDArray to be indexed.
  * @param dst_ndarray The resulting NDArray after indexing. Further details in the comments above,
  */
-template<typename SizeT>
-void index(SizeT num_indices, const NDIndex* indices, const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
-    // Validate `indices`.
+template <typename SizeT>
+void index(SizeT num_indexes, const NDIndex* indexes,
+           const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray) {
+    // First, validate `indexes`.
 
     // Expected value of `dst_ndarray->ndims`.
     SizeT expected_dst_ndims = src_ndarray->ndims;
     // To check for "too many indices for array: array is ?-dimensional, but ? were indexed"
     SizeT num_indexed = 0;
-    // There may be ellipsis `...` in `indices`. There can only be 0 or 1 ellipsis.
+    // There may be ellipsis `...` in `indexes`. There can only be 0 or 1 ellipsis.
     SizeT num_ellipsis = 0;
 
-    for (SizeT i = 0; i < num_indices; i++) {
-        if (indices[i].type == ND_INDEX_TYPE_SINGLE_ELEMENT) {
+    for (SizeT i = 0; i < num_indexes; i++) {
+        if (indexes[i].type == ND_INDEX_TYPE_SINGLE_ELEMENT) {
             expected_dst_ndims--;
             num_indexed++;
-        } else if (indices[i].type == ND_INDEX_TYPE_SLICE) {
+        } else if (indexes[i].type == ND_INDEX_TYPE_SLICE) {
             num_indexed++;
-        } else if (indices[i].type == ND_INDEX_TYPE_NEWAXIS) {
+        } else if (indexes[i].type == ND_INDEX_TYPE_NEWAXIS) {
             expected_dst_ndims++;
-        } else if (indices[i].type == ND_INDEX_TYPE_ELLIPSIS) {
+        } else if (indexes[i].type == ND_INDEX_TYPE_ELLIPSIS) {
             num_ellipsis++;
             if (num_ellipsis > 1) {
-                raise_exception(SizeT, EXN_INDEX_ERROR, "an index can only have a single ellipsis ('...')", NO_PARAM,
-                                NO_PARAM, NO_PARAM);
+                raise_exception(
+                    SizeT, EXN_INDEX_ERROR,
+                    "an index can only have a single ellipsis ('...')",
+                    NO_PARAM, NO_PARAM, NO_PARAM);
             }
         } else {
             __builtin_unreachable();
         }
     }
 
-    debug_assert_eq(SizeT, expected_dst_ndims, dst_ndarray->ndims);
+    if (IRRT_DEBUG_ASSERT) {
+        if (expected_dst_ndims != dst_ndarray->ndims) {
+            raise_exception(
+                SizeT, EXN_ASSERTION_ERROR,
+                "IRRT expected_dst_ndims is {0}, but dst_ndarray->ndims is {1}",
+                expected_dst_ndims, dst_ndarray->ndims, NO_PARAM);
+        }
+    }
 
     if (src_ndarray->ndims - num_indexed < 0) {
         raise_exception(SizeT, EXN_INDEX_ERROR,
                         "too many indices for array: array is {0}-dimensional, "
                         "but {1} were indexed",
-                        src_ndarray->ndims, num_indices, NO_PARAM);
+                        src_ndarray->ndims, num_indexes, NO_PARAM);
     }
 
     dst_ndarray->data = src_ndarray->data;
     dst_ndarray->itemsize = src_ndarray->itemsize;
 
-    // Reference code:
-    // https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L652
+    // Reference code: https://github.com/wadetb/tinynumpy/blob/0d23d22e07062ffab2afa287374c7b366eebdda1/tinynumpy/tinynumpy.py#L652
     SizeT src_axis = 0;
     SizeT dst_axis = 0;
 
-    for (int32_t i = 0; i < num_indices; i++) {
-        const NDIndex* index = &indices[i];
+    for (SliceIndex i = 0; i < num_indexes; i++) {
+        const NDIndex* index = &indexes[i];
         if (index->type == ND_INDEX_TYPE_SINGLE_ELEMENT) {
-            SizeT input = (SizeT) * ((int32_t*)index->data);
+            SliceIndex input = *((SliceIndex*)index->data);
+            SliceIndex k = slice::resolve_index_in_length(
+                src_ndarray->shape[src_axis], input);
 
-            SizeT k = slice::resolve_index_in_length(src_ndarray->shape[src_axis], input);
-            if (k == -1) {
+            if (k == slice::OUT_OF_BOUNDS) {
                 raise_exception(SizeT, EXN_INDEX_ERROR,
                                 "index {0} is out of bounds for axis {1} "
                                 "with size {2}",
                                 input, src_axis, src_ndarray->shape[src_axis]);
             }
 
-            dst_ndarray->data = static_cast<uint8_t*>(dst_ndarray->data) + k * src_ndarray->strides[src_axis];
+            dst_ndarray->data += k * src_ndarray->strides[src_axis];
 
             src_axis++;
         } else if (index->type == ND_INDEX_TYPE_SLICE) {
-            Slice<int32_t>* slice = (Slice<int32_t>*)index->data;
+            UserSlice* input = (UserSlice*)index->data;
 
-            Range<int32_t> range = slice->indices_checked<SizeT>(src_ndarray->shape[src_axis]);
+            Slice slice;
+            input->indices_checked<SizeT>(src_ndarray->shape[src_axis], &slice);
 
-            dst_ndarray->data = static_cast<uint8_t*>(dst_ndarray->data) + (SizeT)range.start * src_ndarray->strides[src_axis];
-            dst_ndarray->strides[dst_axis] = ((SizeT)range.step) * src_ndarray->strides[src_axis];
-            dst_ndarray->shape[dst_axis] = (SizeT)range.len<SizeT>();
+            dst_ndarray->data +=
+                (SizeT)slice.start * src_ndarray->strides[src_axis];
+            dst_ndarray->strides[dst_axis] =
+                ((SizeT)slice.step) * src_ndarray->strides[src_axis];
+            dst_ndarray->shape[dst_axis] = (SizeT)slice.len();
 
             dst_axis++;
             src_axis++;
@@ -194,8 +250,18 @@ void index(SizeT num_indices, const NDIndex* indices, const NDArray<SizeT>* src_
         dst_ndarray->strides[dst_axis] = src_ndarray->strides[src_axis];
     }
 
-    debug_assert_eq(SizeT, src_ndarray->ndims, src_axis);
-    debug_assert_eq(SizeT, dst_ndarray->ndims, dst_axis);
+    if (IRRT_DEBUG_ASSERT) {
+        if (dst_ndarray->ndims != dst_axis) {
+            raise_exception(SizeT, EXN_ASSERTION_ERROR,
+                            "IRRT dst_ndarray->ndims ({0}) != dst_axis ({1})",
+                            dst_ndarray->ndims, dst_axis, NO_PARAM);
+        }
+        if (src_ndarray->ndims != src_axis) {
+            raise_exception(SizeT, EXN_ASSERTION_ERROR,
+                            "IRRT src_ndarray->ndims ({0}) != src_axis ({1})",
+                            src_ndarray->ndims, src_axis, NO_PARAM);
+        }
+    }
 }
 }  // namespace indexing
 }  // namespace ndarray
@@ -204,17 +270,15 @@ void index(SizeT num_indices, const NDIndex* indices, const NDArray<SizeT>* src_
 extern "C" {
 using namespace ndarray::indexing;
 
-void __nac3_ndarray_index(int32_t num_indices,
-                          NDIndex* indices,
+void __nac3_ndarray_index(int32_t num_indexes, NDIndex* indexes,
                           NDArray<int32_t>* src_ndarray,
                           NDArray<int32_t>* dst_ndarray) {
-    index(num_indices, indices, src_ndarray, dst_ndarray);
+    index(num_indexes, indexes, src_ndarray, dst_ndarray);
 }
 
-void __nac3_ndarray_index64(int64_t num_indices,
-                            NDIndex* indices,
+void __nac3_ndarray_index64(int64_t num_indexes, NDIndex* indexes,
                             NDArray<int64_t>* src_ndarray,
                             NDArray<int64_t>* dst_ndarray) {
-    index(num_indices, indices, src_ndarray, dst_ndarray);
+    index(num_indexes, indexes, src_ndarray, dst_ndarray);
 }
 }

nac3core/irrt/irrt/ndarray/iter.hpp

@@ -1,146 +0,0 @@
-#pragma once
-
-#include "irrt/int_types.hpp"
-#include "irrt/ndarray/def.hpp"
-
-namespace {
-/**
- * @brief Helper struct to enumerate through an ndarray *efficiently*.
- *
- * Example usage (in pseudo-code):
- * ```
- * // Suppose my_ndarray has been initialized, with shape [2, 3] and dtype `double`
- * NDIter nditer;
- * nditer.initialize(my_ndarray);
- * while (nditer.has_element()) {
- *    // This body is run 6 (= my_ndarray.size) times.
- *
- *    // [0, 0] -> [0, 1] -> [0, 2] -> [1, 0] -> [1, 1] -> [1, 2] -> end
- *    print(nditer.indices);
- *
- *    // 0 -> 1 -> 2 -> 3 -> 4 -> 5
- *    print(nditer.nth);
- *
- *    // <1st element> -> <2nd element> -> ... -> <6th element> -> end
- *    print(*((double *) nditer.element))
- *
- *    nditer.next(); // Go to next element.
- * }
- * ```
- *
- * Interesting cases:
- * - If `my_ndarray.ndims` == 0, there is one iteration.
- * - If `my_ndarray.shape` contains zeroes, there are no iterations.
- */
-template<typename SizeT>
-struct NDIter {
-    // Information about the ndarray being iterated over.
-    SizeT ndims;
-    SizeT* shape;
-    SizeT* strides;
-
-    /**
-     * @brief The current indices.
-     *
-     * Must be allocated by the caller.
-     */
-    SizeT* indices;
-
-    /**
-     * @brief The nth (0-based) index of the current indices.
-     *
-     * Initially this is 0.
-     */
-    SizeT nth;
-
-    /**
-     * @brief Pointer to the current element.
-     *
-     * Initially this points to first element of the ndarray.
-     */
-    void* element;
-
-    /**
-     * @brief Cache for the product of shape.
-     *
-     * Could be 0 if `shape` has 0s in it.
-     */
-    SizeT size;
-
-    void initialize(SizeT ndims, SizeT* shape, SizeT* strides, void* element, SizeT* indices) {
-        this->ndims = ndims;
-        this->shape = shape;
-        this->strides = strides;
-
-        this->indices = indices;
-        this->element = element;
-
-        // Compute size
-        this->size = 1;
-        for (SizeT i = 0; i < ndims; i++) {
-            this->size *= shape[i];
-        }
-
-        // `indices` starts on all 0s.
-        for (SizeT axis = 0; axis < ndims; axis++)
-            indices[axis] = 0;
-        nth = 0;
-    }
-
-    void initialize_by_ndarray(NDArray<SizeT>* ndarray, SizeT* indices) {
-        // NOTE: ndarray->data is pointing to the first element, and `NDIter`'s `element` should also point to the first
-        // element as well.
-        this->initialize(ndarray->ndims, ndarray->shape, ndarray->strides, ndarray->data, indices);
-    }
-
-    // Is the current iteration valid?
-    // If true, then `element`, `indices` and `nth` contain details about the current element.
-    bool has_element() { return nth < size; }
-
-    // Go to the next element.
-    void next() {
-        for (SizeT i = 0; i < ndims; i++) {
-            SizeT axis = ndims - i - 1;
-            indices[axis]++;
-            if (indices[axis] >= shape[axis]) {
-                indices[axis] = 0;
-
-                // TODO: There is something called backstrides to speedup iteration.
-                // See https://ajcr.net/stride-guide-part-1/, and
-                // https://docs.scipy.org/doc/numpy-1.13.0/reference/c-api.types-and-structures.html#c.PyArrayIterObject.PyArrayIterObject.backstrides.
-                element = static_cast<void*>(reinterpret_cast<uint8_t*>(element) - strides[axis] * (shape[axis] - 1));
-            } else {
-                element = static_cast<void*>(reinterpret_cast<uint8_t*>(element) + strides[axis]);
-                break;
-            }
-        }
-        nth++;
-    }
-};
-}  // namespace
-
-extern "C" {
-void __nac3_nditer_initialize(NDIter<int32_t>* iter, NDArray<int32_t>* ndarray, int32_t* indices) {
-    iter->initialize_by_ndarray(ndarray, indices);
-}
-
-void __nac3_nditer_initialize64(NDIter<int64_t>* iter, NDArray<int64_t>* ndarray, int64_t* indices) {
-    iter->initialize_by_ndarray(ndarray, indices);
-}
-
-bool __nac3_nditer_has_element(NDIter<int32_t>* iter) {
-    return iter->has_element();
-}
-
-bool __nac3_nditer_has_element64(NDIter<int64_t>* iter) {
-    return iter->has_element();
-}
-
-void __nac3_nditer_next(NDIter<int32_t>* iter) {
-    iter->next();
-}
-
-void __nac3_nditer_next64(NDIter<int64_t>* iter) {
-    iter->next();
-}
-}

nac3core/irrt/irrt/ndarray/product.hpp

@@ -0,0 +1,26 @@
+
+namespace {
+namespace ndarray {
+namespace matmul {
+namespace util {
+
+template <typename SizeT>
+void broadcast_shape(SizeT a_ndims, SizeT* a_shape, SizeT b_ndims,
+                     SizeT* b_shape, SizeT* dst_shape) {
+    __builtin_assume(!(a_ndims == 1 && b_ndims == 1));
+    __builtin_assume(a_ndims >= 1);
+    __builtin_assume(b_ndims >= 1);
+}
+}  // namespace util
+
+template <typename SizeT, typename T>
+void matmul_at_least_2d(NDArray<SizeT>* a_ndarray, NDArray<SizeT>* b_ndarray,
+            NDArray<SizeT>* dst_ndarray) {
+    __builtin_assume(sizeof(T) == a_ndarray->itemsize);
+    __builtin_assume(sizeof(T) == b_ndarray->itemsize);
+
+    // See https://numpy.org/doc/stable/reference/generated/numpy.matmul.html#numpy-matmul
+}
+}  // namespace matmul
+}  // namespace ndarray
+}  // namespace

nac3core/irrt/irrt/ndarray/reshape.hpp

@@ -0,0 +1,111 @@
+#pragma once
+
+#include <irrt/int_defs.hpp>
+#include <irrt/ndarray/def.hpp>
+
+namespace {
+namespace ndarray {
+namespace reshape {
+namespace util {
+
+/**
+ * @brief Perform assertions on and resolve unknown dimensions in `new_shape` in `np.reshape(<ndarray>, new_shape)`
+ *
+ * If `new_shape` indeed contains unknown dimensions (specified with `-1`, just like numpy), `new_shape` will be
+ * modified to contain the resolved dimension.
+ *
+ * To perform assertions on and resolve unknown dimensions in `new_shape`, we don't need the actual
+ * `<ndarray>` object itself, but only the `.size` of the `<ndarray>`.
+ *
+ * @param size The `.size` of `<ndarray>`
+ * @param new_ndims Number of elements in `new_shape`
+ * @param new_shape Target shape to reshape to
+ */
+template <typename SizeT>
+void resolve_and_check_new_shape(SizeT size, SizeT new_ndims,
+                                 SizeT* new_shape) {
+    // Is there a -1 in `new_shape`?
+    bool neg1_exists = false;
+    // Location of -1, only initialized if `neg1_exists` is true
+    SizeT neg1_axis_i;
+    // The computed ndarray size of `new_shape`
+    SizeT new_size = 1;
+
+    for (SizeT axis_i = 0; axis_i < new_ndims; axis_i++) {
+        SizeT dim = new_shape[axis_i];
+        if (dim < 0) {
+            if (dim == -1) {
+                if (neg1_exists) {
+                    // Multiple `-1` found. Throw an error.
+                    raise_exception(SizeT, EXN_VALUE_ERROR,
+                                    "can only specify one unknown dimension",
+                                    NO_PARAM, NO_PARAM, NO_PARAM);
+                } else {
+                    neg1_exists = true;
+                    neg1_axis_i = axis_i;
+                }
+            } else {
+                // TODO: What? In `np.reshape` any negative dimensions is
+                // treated like its `-1`.
+                //
+                // Try running `np.zeros((3, 4)).reshape((-999, 2))`
+                //
+                // It is not documented by numpy.
+                // Throw an error for now...
+
+                raise_exception(
+                    SizeT, EXN_VALUE_ERROR,
+                    "Found non -1 negative dimension {0} on axis {1}", dim,
+                    axis_i, NO_PARAM);
+            }
+        } else {
+            new_size *= dim;
+        }
+    }
+
+    bool can_reshape;
+    if (neg1_exists) {
+        // Let `x` be the unknown dimension
+        // Solve `x * <new_size> = <size>`
+        if (new_size == 0 && size == 0) {
+            // `x` has infinitely many solutions
+            can_reshape = false;
+        } else if (new_size == 0 && size != 0) {
+            // `x` has no solutions
+            can_reshape = false;
+        } else if (size % new_size != 0) {
+            // `x` has no integer solutions
+            can_reshape = false;
+        } else {
+            can_reshape = true;
+            new_shape[neg1_axis_i] = size / new_size;  // Resolve dimension
+        }
+    } else {
+        can_reshape = (new_size == size);
+    }
+
+    if (!can_reshape) {
+        raise_exception(SizeT, EXN_VALUE_ERROR,
+                        "cannot reshape array of size {0} into given shape",
+                        size, NO_PARAM, NO_PARAM);
+    }
+}
+}  // namespace util
+}  // namespace reshape
+}  // namespace ndarray
+}  // namespace
+
+extern "C" {
+void __nac3_ndarray_resolve_and_check_new_shape(int32_t size, int32_t new_ndims,
+                                                int32_t* new_shape) {
+    ndarray::reshape::util::resolve_and_check_new_shape(size, new_ndims,
+                                                        new_shape);
+}
+
+void __nac3_ndarray_resolve_and_check_new_shape64(int64_t size,
+                                                  int64_t new_ndims,
+                                                  int64_t* new_shape) {
+    ndarray::reshape::util::resolve_and_check_new_shape(size, new_ndims,
+                                                        new_shape);
+}
+}

nac3core/irrt/irrt/ndarray/transpose.hpp

@@ -0,0 +1,148 @@
+#pragma once
+
+#include <irrt/int_defs.hpp>
+#include <irrt/ndarray/def.hpp>
+#include <irrt/slice.hpp>
+
+/*
+ * Notes on `np.transpose(<array>, <axes>)`
+ *
+ * TODO: `axes`, if specified, can actually contain negative indices,
+ * but it is not documented in numpy.
+ *
+ * Supporting it for now.
+ */
+
+namespace {
+namespace ndarray {
+namespace transpose {
+namespace util {
+
+/**
+ * @brief Do assertions on `<axes>` in `np.transpose(<array>, <axes>)`.
+ *
+ * Note that `np.transpose`'s `<axe>` argument is optional. If the argument
+ * is specified but the user, use this function to do assertions on it.
+ *
+ * @param ndims The number of dimensions of `<array>`
+ * @param num_axes Number of elements in `<axes>` as specified by the user.
+ *   This should be equal to `ndims`. If not, a "ValueError: axes don't match array" is thrown.
+ * @param axes The user specified `<axes>`.
+ */
+template <typename SizeT>
+void assert_transpose_axes(SizeT ndims, SizeT num_axes, const SizeT* axes) {
+    if (ndims != num_axes) {
+        raise_exception(SizeT, EXN_VALUE_ERROR, "axes don't match array",
+                        NO_PARAM, NO_PARAM, NO_PARAM);
+    }
+
+    // TODO: Optimize this
+    bool* axe_specified = (bool*)__builtin_alloca(sizeof(bool) * ndims);
+    for (SizeT i = 0; i < ndims; i++) axe_specified[i] = false;
+
+    for (SizeT i = 0; i < ndims; i++) {
+        SizeT axis = slice::resolve_index_in_length(ndims, axes[i]);
+        if (axis == slice::OUT_OF_BOUNDS) {
+            // TODO: numpy actually throws a `numpy.exceptions.AxisError`
+            raise_exception(
+                SizeT, EXN_VALUE_ERROR,
+                "axis {0} is out of bounds for array of dimension {1}", axis,
+                ndims, NO_PARAM);
+        }
+
+        if (axe_specified[axis]) {
+            raise_exception(SizeT, EXN_VALUE_ERROR,
+                            "repeated axis in transpose", NO_PARAM, NO_PARAM,
+                            NO_PARAM);
+        }
+
+        axe_specified[axis] = true;
+    }
+}
+}  // namespace util
+
+/**
+ * @brief Create a transpose view of `src_ndarray` and perform proper assertions.
+ *
+ * This function is very similar to doing `dst_ndarray = np.transpose(src_ndarray, <axes>)`.
+ * If `<axes>` is supposed to be `None`, caller can pass in a `nullptr` to `<axes>`.
+ *
+ * The transpose view created is returned by modifying `dst_ndarray`.
+ *
+ * The caller is responsible for setting up `dst_ndarray` before calling this function.
+ * Here is what this function expects from `dst_ndarray` when called:
+ *   - `dst_ndarray->data` does not have to be initialized.
+ *   - `dst_ndarray->itemsize` does not have to be initialized.
+ *   - `dst_ndarray->ndims` must be initialized, must be equal to `src_ndarray->ndims`.
+ *   - `dst_ndarray->shape` must be allocated, through it can contain uninitialized values.
+ *   - `dst_ndarray->strides` must be allocated, through it can contain uninitialized values.
+ * When this function call ends:
+ *   - `dst_ndarray->data` is set to `src_ndarray->data` (`dst_ndarray` is just a view to `src_ndarray`)
+ *   - `dst_ndarray->itemsize` is set to `src_ndarray->itemsize`
+ *   - `dst_ndarray->ndims` is unchanged
+ *   - `dst_ndarray->shape` is updated according to how `np.transpose` works
+ *   - `dst_ndarray->strides` is updated according to how `np.transpose` works
+ *
+ * @param src_ndarray The NDArray to build a transpose view on
+ * @param dst_ndarray The resulting NDArray after transpose. Further details in the comments above,
+ * @param num_axes Number of elements in axes. Unused if `axes` is nullptr.
+ * @param axes Axes permutation. Set it to `nullptr` if `<axes>` is `None`.
+ */
+template <typename SizeT>
+void transpose(const NDArray<SizeT>* src_ndarray, NDArray<SizeT>* dst_ndarray,
+               SizeT num_axes, const SizeT* axes) {
+    __builtin_assume(src_ndarray->ndims == dst_ndarray->ndims);
+    const auto ndims = src_ndarray->ndims;
+
+    if (axes != nullptr) util::assert_transpose_axes(ndims, num_axes, axes);
+
+    dst_ndarray->data = src_ndarray->data;
+    dst_ndarray->itemsize = src_ndarray->itemsize;
+
+    // Check out https://ajcr.net/stride-guide-part-2/ to see how `np.transpose` works behind the scenes.
+    if (axes == nullptr) {
+        // `np.transpose(<array>, axes=None)`
+
+        /*
+         * Minor note: `np.transpose(<array>, axes=None)` is equivalent to
+         * `np.transpose(<array>, axes=[N-1, N-2, ..., 0])` - basically it
+         * is reversing the order of strides and shape.
+         *
+         * This is a fast implementation to handle this special (but very common) case.
+         */
+
+        for (SizeT axis = 0; axis < ndims; axis++) {
+            dst_ndarray->shape[axis] = src_ndarray->shape[ndims - axis - 1];
+            dst_ndarray->strides[axis] = src_ndarray->strides[ndims - axis - 1];
+        }
+    } else {
+        // `np.transpose(<array>, <axes>)`
+
+        // Permute strides and shape according to `axes`, while resolving negative indices in `axes`
+        for (SizeT axis = 0; axis < ndims; axis++) {
+            // `i` cannot be OUT_OF_BOUNDS because of assertions
+            SizeT i = slice::resolve_index_in_length(ndims, axes[axis]);
+
+            dst_ndarray->shape[axis] = src_ndarray->shape[i];
+            dst_ndarray->strides[axis] = src_ndarray->strides[i];
+        }
+    }
+}
+}  // namespace transpose
+}  // namespace ndarray
+}  // namespace
+
+extern "C" {
+using namespace ndarray::transpose;
+void __nac3_ndarray_transpose(const NDArray<int32_t>* src_ndarray,
+                              NDArray<int32_t>* dst_ndarray, int32_t num_axes,
+                              const int32_t* axes) {
+    transpose(src_ndarray, dst_ndarray, num_axes, axes);
+}
+
+void __nac3_ndarray_transpose64(const NDArray<int64_t>* src_ndarray,
+                                NDArray<int64_t>* dst_ndarray, int64_t num_axes,
+                                const int64_t* axes) {
+    transpose(src_ndarray, dst_ndarray, num_axes, axes);
+}
+}

nac3core/irrt/irrt/range.hpp

@@ -1,47 +0,0 @@
-#pragma once
-
-#include "irrt/debug.hpp"
-#include "irrt/int_types.hpp"
-
-namespace {
-namespace range {
-template<typename T>
-T len(T start, T stop, T step) {
-    // Reference:
-    // https://github.com/python/cpython/blob/9dbd12375561a393eaec4b21ee4ac568a407cdb0/Objects/rangeobject.c#L933
-    if (step > 0 && start < stop)
-        return 1 + (stop - 1 - start) / step;
-    else if (step < 0 && start > stop)
-        return 1 + (start - 1 - stop) / (-step);
-    else
-        return 0;
-}
-}  // namespace range
-
-/**
- * @brief A Python range.
- */
-template<typename T>
-struct Range {
-    T start;
-    T stop;
-    T step;
-
-    /**
-     * @brief Calculate the `len()` of this range.
-     */
-    template<typename SizeT>
-    T len() {
-        debug_assert(SizeT, step != 0);
-        return range::len(start, stop, step);
-    }
-};
-}  // namespace
-
-extern "C" {
-using namespace range;
-
-SliceIndex __nac3_range_slice_len(const SliceIndex start, const SliceIndex end, const SliceIndex step) {
-    return len(start, end, step);
-}
-}

nac3core/irrt/irrt/slice.hpp

@@ -1,67 +1,79 @@
 #pragma once
 
-#include "irrt/debug.hpp"
-#include "irrt/exception.hpp"
-#include "irrt/int_types.hpp"
-#include "irrt/math_util.hpp"
-#include "irrt/range.hpp"
+#include <irrt/int_defs.hpp>
+#include <irrt/slice.hpp>
+#include <irrt/util.hpp>
+
+#include "exception.hpp"
+
+// The type of an index or a value describing the length of a
+// range/slice is always `int32_t`.
+using SliceIndex = int32_t;
 
 namespace {
+
+/**
+ * @brief A Python-like slice with resolved indices.
+ *
+ * "Resolved indices" means that `start` and `stop` must be positive and are
+ * bound to a known length.
+ */
+struct Slice {
+    SliceIndex start;
+    SliceIndex stop;
+    SliceIndex step;
+
+    /**
+     * @brief Calculate and return the length / the number of the slice.
+     * 
+     * If this were a Python range, this function would be `len(range(start, stop, step))`.
+     */
+    SliceIndex len() {
+        SliceIndex diff = stop - start;
+        if (diff > 0 && step > 0) {
+            return ((diff - 1) / step) + 1;
+        } else if (diff < 0 && step < 0) {
+            return ((diff + 1) / step) + 1;
+        } else {
+            return 0;
+        }
+    }
+};
+
 namespace slice {
 /**
- * @brief Resolve a possibly negative index in a list of a known length.
+ * @brief Resolve a slice index under a given length like Python indexing.
+ *
+ * In Python, if you have a `list` of length 100, `list[-1]` resolves to
+ * `list[99]`, so `resolve_index_in_length_clamped(100, -1)` returns `99`.
+ *
+ * If `length` is 0, 0 is returned for any value of `index`.
+ *
+ * If `index` is out of bounds, clamps the returned value between `0` and
+ * `length - 1` (inclusive).
  *
- * Returns -1 if the resolved index is out of the list's bounds.
  */
-template<typename T>
-T resolve_index_in_length(T length, T index) {
-    T resolved = index < 0 ? length + index : index;
-    if (0 <= resolved && resolved < length) {
-        return resolved;
+SliceIndex resolve_index_in_length_clamped(SliceIndex length,
+                                           SliceIndex index) {
+    if (index < 0) {
+        return max<SliceIndex>(length + index, 0);
     } else {
-        return -1;
+        return min<SliceIndex>(length, index);
     }
 }
 
+const SliceIndex OUT_OF_BOUNDS = -1;
+
 /**
- * @brief Resolve a slice as a range.
- *
- * This is equivalent to `range(*slice(start, stop, step).indices(length))` in Python.
+ * @brief Like `resolve_index_in_length_clamped`, but returns `OUT_OF_BOUNDS`
+ * if `index` is out of bounds.
  */
-template<typename T>
-void indices(bool start_defined,
-             T start,
-             bool stop_defined,
-             T stop,
-             bool step_defined,
-             T step,
-             T length,
-             T* range_start,
-             T* range_stop,
-             T* range_step) {
-    // Reference: https://github.com/python/cpython/blob/main/Objects/sliceobject.c#L388
-    *range_step = step_defined ? step : 1;
-    bool step_is_negative = *range_step < 0;
-
-    T lower, upper;
-    if (step_is_negative) {
-        lower = -1;
-        upper = length - 1;
+SliceIndex resolve_index_in_length(SliceIndex length, SliceIndex index) {
+    SliceIndex resolved = index < 0 ? length + index : index;
+    if (0 <= resolved && resolved < length) {
+        return resolved;
     } else {
-        lower = 0;
-        upper = length;
-    }
-
-    if (start_defined) {
-        *range_start = start < 0 ? max(lower, start + length) : min(upper, start);
-    } else {
-        *range_start = step_is_negative ? upper : lower;
-    }
-
-    if (stop_defined) {
-        *range_stop = stop < 0 ? max(lower, stop + length) : min(upper, stop);
-    } else {
-        *range_stop = step_is_negative ? lower : upper;
+        return OUT_OF_BOUNDS;
     }
 }
 }  // namespace slice
@@ -69,18 +81,17 @@ void indices(bool start_defined,
 /**
  * @brief A Python-like slice with **unresolved** indices.
  */
-template<typename T>
-struct Slice {
+struct UserSlice {
     bool start_defined;
-    T start;
+    SliceIndex start;
 
     bool stop_defined;
-    T stop;
+    SliceIndex stop;
 
     bool step_defined;
-    T step;
+    SliceIndex step;
 
-    Slice() { this->reset(); }
+    UserSlice() { this->reset(); }
 
     void reset() {
         this->start_defined = false;
@@ -88,69 +99,67 @@ struct Slice {
         this->step_defined = false;
     }
 
-    void set_start(T start) {
+    void set_start(SliceIndex start) {
         this->start_defined = true;
         this->start = start;
     }
 
-    void set_stop(T stop) {
+    void set_stop(SliceIndex stop) {
         this->stop_defined = true;
         this->stop = stop;
     }
 
-    void set_step(T step) {
+    void set_step(SliceIndex step) {
         this->step_defined = true;
         this->step = step;
     }
 
     /**
-     * @brief Resolve this slice as a range.
+     * @brief Resolve this slice.
      *
-     * In Python, this would be `range(*slice(start, stop, step).indices(length))`.
+     * In Python, this would be `slice(start, stop, step).indices(length)`.
+     *
+     * @return A `Slice` with the resolved indices.
      */
-    template<typename SizeT>
-    Range<T> indices(T length) {
-        // Reference:
-        // https://github.com/python/cpython/blob/main/Objects/sliceobject.c#L388
-        debug_assert(SizeT, length >= 0);
+    Slice indices(SliceIndex length) {
+        Slice result;
+
+        result.step = step_defined ? step : 1;
+        bool step_is_negative = result.step < 0;
+
+        if (start_defined) {
+            result.start =
+                slice::resolve_index_in_length_clamped(length, start);
+        } else {
+            result.start = step_is_negative ? length - 1 : 0;
+        }
+
+        if (stop_defined) {
+            result.stop = slice::resolve_index_in_length_clamped(length, stop);
+        } else {
+            result.stop = step_is_negative ? -1 : length;
+        }
 
-        Range<T> result;
-        slice::indices(start_defined, start, stop_defined, stop, step_defined, step, length, &result.start,
-                       &result.stop, &result.step);
         return result;
     }
 
     /**
      * @brief Like `.indices()` but with assertions.
      */
-    template<typename SizeT>
-    Range<T> indices_checked(T length) {
-        // TODO: Switch to `SizeT length`
-
+    template <typename SizeT>
+    void indices_checked(SliceIndex length, Slice* result) {
         if (length < 0) {
-            raise_exception(SizeT, EXN_VALUE_ERROR, "length should not be negative, got {0}", length, NO_PARAM,
-                            NO_PARAM);
+            raise_exception(SizeT, EXN_VALUE_ERROR,
+                            "length should not be negative, got {0}", length,
+                            NO_PARAM, NO_PARAM);
         }
 
         if (this->step_defined && this->step == 0) {
-            raise_exception(SizeT, EXN_VALUE_ERROR, "slice step cannot be zero", NO_PARAM, NO_PARAM, NO_PARAM);
+            raise_exception(SizeT, EXN_VALUE_ERROR, "slice step cannot be zero",
+                            NO_PARAM, NO_PARAM, NO_PARAM);
         }
 
-        return this->indices<SizeT>(length);
+        *result = this->indices(length);
     }
 };
 }  // namespace
-
-extern "C" {
-SliceIndex __nac3_slice_index_bound(SliceIndex i, const SliceIndex len) {
-    if (i < 0) {
-        i = len + i;
-    }
-    if (i < 0) {
-        return 0;
-    } else if (i > len) {
-        return len;
-    }
-    return i;
-}
-}

nac3core/irrt/irrt/util.hpp

@@ -0,0 +1,101 @@
+#pragma once
+
+namespace {
+template <typename T>
+const T& max(const T& a, const T& b) {
+    return a > b ? a : b;
+}
+
+template <typename T>
+const T& min(const T& a, const T& b) {
+    return a > b ? b : a;
+}
+
+template <typename T>
+bool arrays_match(int len, T* as, T* bs) {
+    for (int i = 0; i < len; i++) {
+        if (as[i] != bs[i]) return false;
+    }
+    return true;
+}
+
+namespace cstr_utils {
+/**
+ * @brief Return true if `str` is empty.
+ */
+bool is_empty(const char* str) { return str[0] == '\0'; }
+
+/**
+ * @brief Implementation of `strcmp()`
+ */
+int8_t compare(const char* a, const char* b) {
+    uint32_t i = 0;
+    while (true) {
+        if (a[i] < b[i]) {
+            return -1;
+        } else if (a[i] > b[i]) {
+            return 1;
+        } else {
+            if (a[i] == '\0') {
+                return 0;
+            } else {
+                i++;
+            }
+        }
+    }
+}
+
+/**
+ * @brief Return true two strings have the same content.
+ */
+int8_t equal(const char* a, const char* b) { return compare(a, b) == 0; }
+
+/**
+ * @brief Implementation of `strlen()`.
+ */
+uint32_t length(const char* str) {
+    uint32_t length = 0;
+    while (*str != '\0') {
+        length++;
+        str++;
+    }
+    return length;
+}
+
+/**
+ * @brief Copy a null-terminated string to a buffer with limited size and guaranteed null-termination.
+ *
+ * `dst_max_size` must be greater than 0, otherwise this function has undefined behavior.
+ *
+ * This function attempts to copy everything from `src` from `dst`, and *always* null-terminates `dst`.
+ *
+ * If the size of `dst` is too small, the final byte (`dst[dst_max_size - 1]`) of `dst` will be set to
+ * the null terminator.
+ *
+ * @param src String to copy from.
+ * @param dst Buffer to copy string to.
+ * @param dst_max_size
+ *   Number of bytes of this buffer, including the space needed for the null terminator.
+ *   Must be greater than 0.
+ * @return If `dst` is too small to contain everything in `src`.
+ */
+bool copy(const char* src, char* dst, uint32_t dst_max_size) {
+    for (uint32_t i = 0; i < dst_max_size; i++) {
+        bool is_last = i + 1 == dst_max_size;
+        if (is_last && src[i] != '\0') {
+            dst[i] = '\0';
+            return false;
+        }
+
+        if (src[i] == '\0') {
+            dst[i] = '\0';
+            return true;
+        }
+
+        dst[i] = src[i];
+    }
+
+    __builtin_unreachable();
+}
+}  // namespace cstr_utils
+}  // namespace

nac3core/irrt/irrt_everything.hpp

@@ -0,0 +1,16 @@
+#pragma once
+
+#ifndef IRRT_DEBUG_ASSERT
+#error IRRT_DEBUG_ASSERT flag is missing!! Please define it to 'false' or 'true'.
+#endif
+
+#include <irrt/core.hpp>
+#include <irrt/exception.hpp>
+#include <irrt/int_defs.hpp>
+#include <irrt/ndarray/basic.hpp>
+#include <irrt/ndarray/broadcast.hpp>
+#include <irrt/ndarray/def.hpp>
+#include <irrt/ndarray/indexing.hpp>
+#include <irrt/ndarray/reshape.hpp>
+#include <irrt/ndarray/transpose.hpp>
+#include <irrt/util.hpp>

nac3core/irrt/irrt_test.cpp

@@ -0,0 +1,25 @@
+// This file will be compiled like a real C++ program,
+// and we do have the luxury to use the standard libraries.
+// That is if the nix flakes do not have issues... especially on msys2...
+
+#include <cstdint>
+#include <cstdio>
+#include <cstdlib>
+
+// Special macro to inform `#include <irrt/*>` that we are testing.
+#define IRRT_TESTING
+
+// Note that failure unit tests are not supported.
+
+#include <test/test_core.hpp>
+#include <test/test_ndarray_basic.hpp>
+#include <test/test_ndarray_broadcast.hpp>
+#include <test/test_ndarray_indexing.hpp>
+
+int main() {
+    test::core::run();
+    test::ndarray_basic::run();
+    test::ndarray_indexing::run();
+    test::ndarray_broadcast::run();
+    return 0;
+}

nac3core/irrt/test/includes.hpp

@@ -0,0 +1,11 @@
+#pragma once
+
+#include <cstdint>
+#include <cstdio>
+#include <cstdlib>
+#include <irrt_everything.hpp>
+#include <test/util.hpp>
+
+/*
+    Include this header for every test_*.cpp
+*/

nac3core/irrt/test/test_core.hpp

@@ -0,0 +1,16 @@
+#pragma once
+
+#include <test/includes.hpp>
+
+namespace test {
+namespace core {
+void test_int_exp() {
+    BEGIN_TEST();
+
+    assert_values_match(125L, __nac3_int_exp_impl<int64_t>(5, 3));
+    assert_values_match(3125L, __nac3_int_exp_impl<int64_t>(5, 5));
+}
+
+void run() { test_int_exp(); }
+}  // namespace core
+}  // namespace test

nac3core/irrt/test/test_ndarray_basic.hpp

@@ -0,0 +1,30 @@
+#pragma once
+
+#include <test/includes.hpp>
+
+namespace test {
+namespace ndarray_basic {
+void test_calc_size_from_shape_normal() {
+    // Test shapes with normal values
+    BEGIN_TEST();
+
+    int64_t shape[4] = {2, 3, 5, 7};
+    assert_values_match(
+        210L, ndarray::basic::util::calc_size_from_shape<int64_t>(4, shape));
+}
+
+void test_calc_size_from_shape_has_zero() {
+    // Test shapes with 0 in them
+    BEGIN_TEST();
+
+    int64_t shape[4] = {2, 0, 5, 7};
+    assert_values_match(
+        0L, ndarray::basic::util::calc_size_from_shape<int64_t>(4, shape));
+}
+
+void run() {
+    test_calc_size_from_shape_normal();
+    test_calc_size_from_shape_has_zero();
+}
+}  // namespace ndarray_basic
+}  // namespace test

nac3core/irrt/test/test_ndarray_broadcast.hpp

@@ -0,0 +1,127 @@
+#pragma once
+
+#include <test/includes.hpp>
+
+namespace test {
+namespace ndarray_broadcast {
+void test_can_broadcast_shape() {
+    BEGIN_TEST();
+
+    assert_values_match(true,
+                        ndarray::broadcast::util::can_broadcast_shape_to(
+                            1, (int32_t[]){3}, 5, (int32_t[]){1, 1, 1, 1, 3}));
+    assert_values_match(false, ndarray::broadcast::util::can_broadcast_shape_to(
+                                   1, (int32_t[]){3}, 2, (int32_t[]){3, 1}));
+    assert_values_match(true, ndarray::broadcast::util::can_broadcast_shape_to(
+                                  1, (int32_t[]){3}, 1, (int32_t[]){3}));
+    assert_values_match(false, ndarray::broadcast::util::can_broadcast_shape_to(
+                                   1, (int32_t[]){1}, 1, (int32_t[]){3}));
+    assert_values_match(true, ndarray::broadcast::util::can_broadcast_shape_to(
+                                  1, (int32_t[]){1}, 1, (int32_t[]){1}));
+    assert_values_match(
+        true, ndarray::broadcast::util::can_broadcast_shape_to(
+                  3, (int32_t[]){256, 256, 3}, 3, (int32_t[]){256, 1, 3}));
+    assert_values_match(true,
+                        ndarray::broadcast::util::can_broadcast_shape_to(
+                            3, (int32_t[]){256, 256, 3}, 1, (int32_t[]){3}));
+    assert_values_match(false,
+                        ndarray::broadcast::util::can_broadcast_shape_to(
+                            3, (int32_t[]){256, 256, 3}, 1, (int32_t[]){2}));
+    assert_values_match(true,
+                        ndarray::broadcast::util::can_broadcast_shape_to(
+                            3, (int32_t[]){256, 256, 3}, 1, (int32_t[]){1}));
+
+    // In cases when the shapes contain zero(es)
+    assert_values_match(true, ndarray::broadcast::util::can_broadcast_shape_to(
+                                  1, (int32_t[]){0}, 1, (int32_t[]){1}));
+    assert_values_match(false, ndarray::broadcast::util::can_broadcast_shape_to(
+                                   1, (int32_t[]){0}, 1, (int32_t[]){2}));
+    assert_values_match(true,
+                        ndarray::broadcast::util::can_broadcast_shape_to(
+                            4, (int32_t[]){0, 4, 0, 0}, 1, (int32_t[]){1}));
+    assert_values_match(
+        true, ndarray::broadcast::util::can_broadcast_shape_to(
+                  4, (int32_t[]){0, 4, 0, 0}, 4, (int32_t[]){1, 1, 1, 1}));
+    assert_values_match(
+        true, ndarray::broadcast::util::can_broadcast_shape_to(
+                  4, (int32_t[]){0, 4, 0, 0}, 4, (int32_t[]){1, 4, 1, 1}));
+    assert_values_match(false, ndarray::broadcast::util::can_broadcast_shape_to(
+                                   2, (int32_t[]){4, 3}, 2, (int32_t[]){0, 3}));
+    assert_values_match(false, ndarray::broadcast::util::can_broadcast_shape_to(
+                                   2, (int32_t[]){4, 3}, 2, (int32_t[]){0, 0}));
+}
+
+void test_ndarray_broadcast() {
+    /*
+    # array = np.array([[19.9, 29.9, 39.9, 49.9]], dtype=np.float64)
+    # >>> [[19.9 29.9 39.9 49.9]]
+    #
+    # array = np.broadcast_to(array, (2, 3, 4))
+    # >>> [[[19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]]
+    # >>>  [[19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]
+    # >>>   [19.9 29.9 39.9 49.9]]]
+    #
+    # assery array.strides == (0, 0, 8)
+
+    */
+    BEGIN_TEST();
+
+    double in_data[4] = {19.9, 29.9, 39.9, 49.9};
+    const int32_t in_ndims = 2;
+    int32_t in_shape[in_ndims] = {1, 4};
+    int32_t in_strides[in_ndims] = {};
+    NDArray<int32_t> ndarray = {.data = (uint8_t*)in_data,
+                                .itemsize = sizeof(double),
+                                .ndims = in_ndims,
+                                .shape = in_shape,
+                                .strides = in_strides};
+    ndarray::basic::set_strides_by_shape(&ndarray);
+
+    const int32_t dst_ndims = 3;
+    int32_t dst_shape[dst_ndims] = {2, 3, 4};
+    int32_t dst_strides[dst_ndims] = {};
+    NDArray<int32_t> dst_ndarray = {
+        .ndims = dst_ndims, .shape = dst_shape, .strides = dst_strides};
+
+    ndarray::broadcast::broadcast_to(&ndarray, &dst_ndarray);
+
+    assert_arrays_match(dst_ndims, ((int32_t[]){0, 0, 8}), dst_ndarray.strides);
+
+    assert_values_match(19.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 0, 0}))));
+    assert_values_match(29.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 0, 1}))));
+    assert_values_match(39.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 0, 2}))));
+    assert_values_match(49.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 0, 3}))));
+    assert_values_match(19.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 1, 0}))));
+    assert_values_match(29.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 1, 1}))));
+    assert_values_match(39.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 1, 2}))));
+    assert_values_match(49.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){0, 1, 3}))));
+    assert_values_match(49.9,
+                        *((double*)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, ((int32_t[]){1, 2, 3}))));
+}
+
+void run() {
+    test_can_broadcast_shape();
+    test_ndarray_broadcast();
+}
+}  // namespace ndarray_broadcast
+}  // namespace test

nac3core/irrt/test/test_ndarray_indexing.hpp

@@ -0,0 +1,165 @@
+#pragma once
+
+#include <test/includes.hpp>
+
+namespace test {
+namespace ndarray_indexing {
+void test_normal_1() {
+    /*
+        Reference Python code:
+        ```python
+        ndarray = np.arange(12, dtype=np.float64).reshape((3, 4));
+        # array([[ 0.,  1.,  2.,  3.],
+        #        [ 4.,  5.,  6.,  7.],
+        #        [ 8.,  9., 10., 11.]])
+
+        dst_ndarray = ndarray[-2:, 1::2]
+        # array([[ 5.,  7.],
+        #        [ 9., 11.]])
+
+        assert dst_ndarray.shape == (2, 2)
+        assert dst_ndarray.strides == (32, 16)
+        assert dst_ndarray[0, 0] == 5.0
+        assert dst_ndarray[0, 1] == 7.0
+        assert dst_ndarray[1, 0] == 9.0
+        assert dst_ndarray[1, 1] == 11.0
+        ```
+    */
+    BEGIN_TEST();
+
+    // Prepare src_ndarray
+    double src_data[12] = {0.0, 1.0, 2.0, 3.0, 4.0,  5.0,
+                           6.0, 7.0, 8.0, 9.0, 10.0, 11.0};
+    int64_t src_itemsize = sizeof(double);
+    const int64_t src_ndims = 2;
+    int64_t src_shape[src_ndims] = {3, 4};
+    int64_t src_strides[src_ndims] = {};
+    NDArray<int64_t> src_ndarray = {.data = (uint8_t *)src_data,
+                                    .itemsize = src_itemsize,
+                                    .ndims = src_ndims,
+                                    .shape = src_shape,
+                                    .strides = src_strides};
+    ndarray::basic::set_strides_by_shape(&src_ndarray);
+
+    // Prepare dst_ndarray
+    const int64_t dst_ndims = 2;
+    int64_t dst_shape[dst_ndims] = {999, 999};    // Empty values
+    int64_t dst_strides[dst_ndims] = {999, 999};  // Empty values
+    NDArray<int64_t> dst_ndarray = {.data = nullptr,
+                                    .ndims = dst_ndims,
+                                    .shape = dst_shape,
+                                    .strides = dst_strides};
+
+    // Create the subscripts in `ndarray[-2::, 1::2]`
+    UserSlice subscript_1;
+    subscript_1.set_start(-2);
+
+    UserSlice subscript_2;
+    subscript_2.set_start(1);
+    subscript_2.set_step(2);
+
+    const int64_t num_indexes = 2;
+    NDIndex indexes[num_indexes] = {
+        {.type = ND_INDEX_TYPE_SLICE, .data = (uint8_t *)&subscript_1},
+        {.type = ND_INDEX_TYPE_SLICE, .data = (uint8_t *)&subscript_2}};
+
+    ndarray::indexing::index(num_indexes, indexes, &src_ndarray, &dst_ndarray);
+
+    int64_t expected_shape[dst_ndims] = {2, 2};
+    int64_t expected_strides[dst_ndims] = {32, 16};
+
+    assert_arrays_match(dst_ndims, expected_shape, dst_ndarray.shape);
+    assert_arrays_match(dst_ndims, expected_strides, dst_ndarray.strides);
+
+    // dst_ndarray[0, 0]
+    assert_values_match(5.0,
+                        *((double *)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, (int64_t[dst_ndims]){0, 0})));
+    // dst_ndarray[0, 1]
+    assert_values_match(7.0,
+                        *((double *)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, (int64_t[dst_ndims]){0, 1})));
+    // dst_ndarray[1, 0]
+    assert_values_match(9.0,
+                        *((double *)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, (int64_t[dst_ndims]){1, 0})));
+    // dst_ndarray[1, 1]
+    assert_values_match(11.0,
+                        *((double *)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, (int64_t[dst_ndims]){1, 1})));
+}
+
+void test_normal_2() {
+    /*
+        ```python
+        ndarray = np.arange(12, dtype=np.float64).reshape((3, 4))
+        # array([[ 0.,  1.,  2.,  3.],
+        #        [ 4.,  5.,  6.,  7.],
+        #        [ 8.,  9., 10., 11.]])
+
+        dst_ndarray = ndarray[2, ::-2]
+        # array([11.,  9.])
+
+        assert dst_ndarray.shape == (2,)
+        assert dst_ndarray.strides == (-16,)
+        assert dst_ndarray[0] == 11.0
+        assert dst_ndarray[1] == 9.0
+        ```
+    */
+    BEGIN_TEST();
+
+    // Prepare src_ndarray
+    double src_data[12] = {0.0, 1.0, 2.0, 3.0, 4.0,  5.0,
+                           6.0, 7.0, 8.0, 9.0, 10.0, 11.0};
+    int64_t src_itemsize = sizeof(double);
+    const int64_t src_ndims = 2;
+    int64_t src_shape[src_ndims] = {3, 4};
+    int64_t src_strides[src_ndims] = {};
+    NDArray<int64_t> src_ndarray = {.data = (uint8_t *)src_data,
+                                    .itemsize = src_itemsize,
+                                    .ndims = src_ndims,
+                                    .shape = src_shape,
+                                    .strides = src_strides};
+    ndarray::basic::set_strides_by_shape(&src_ndarray);
+
+    // Prepare dst_ndarray
+    const int64_t dst_ndims = 1;
+    int64_t dst_shape[dst_ndims] = {999};    // Empty values
+    int64_t dst_strides[dst_ndims] = {999};  // Empty values
+    NDArray<int64_t> dst_ndarray = {.data = nullptr,
+                                    .ndims = dst_ndims,
+                                    .shape = dst_shape,
+                                    .strides = dst_strides};
+
+    // Create the subscripts in `ndarray[2, ::-2]`
+    int64_t subscript_1 = 2;
+
+    UserSlice subscript_2;
+    subscript_2.set_step(-2);
+
+    const int64_t num_indexes = 2;
+    NDIndex indexes[num_indexes] = {
+        {.type = ND_INDEX_TYPE_SINGLE_ELEMENT, .data = (uint8_t *)&subscript_1},
+        {.type = ND_INDEX_TYPE_SLICE, .data = (uint8_t *)&subscript_2}};
+
+    ndarray::indexing::index(num_indexes, indexes, &src_ndarray, &dst_ndarray);
+
+    int64_t expected_shape[dst_ndims] = {2};
+    int64_t expected_strides[dst_ndims] = {-16};
+    assert_arrays_match(dst_ndims, expected_shape, dst_ndarray.shape);
+    assert_arrays_match(dst_ndims, expected_strides, dst_ndarray.strides);
+
+    assert_values_match(11.0,
+                        *((double *)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, (int64_t[dst_ndims]){0})));
+    assert_values_match(9.0,
+                        *((double *)ndarray::basic::get_pelement_by_indices(
+                            &dst_ndarray, (int64_t[dst_ndims]){1})));
+}
+
+void run() {
+    test_normal_1();
+    test_normal_2();
+}
+}  // namespace ndarray_indexing
+}  // namespace test

nac3core/irrt/test/util.hpp

@@ -0,0 +1,131 @@
+#pragma once
+
+#include <cstdio>
+#include <cstdlib>
+
+template <class T>
+void print_value(const T& value);
+
+template <>
+void print_value(const bool& value) {
+    printf("%s", value ? "true" : "false");
+}
+
+template <>
+void print_value(const int8_t& value) {
+    printf("%d", value);
+}
+
+template <>
+void print_value(const int32_t& value) {
+    printf("%d", value);
+}
+
+template <>
+void print_value(const int64_t& value) {
+    printf("%d", value);
+}
+
+template <>
+void print_value(const uint8_t& value) {
+    printf("%u", value);
+}
+
+template <>
+void print_value(const uint32_t& value) {
+    printf("%u", value);
+}
+
+template <>
+void print_value(const uint64_t& value) {
+    printf("%d", value);
+}
+
+template <>
+void print_value(const float& value) {
+    printf("%f", value);
+}
+
+template <>
+void print_value(const double& value) {
+    printf("%f", value);
+}
+
+void __begin_test(const char* function_name, const char* file, int line) {
+    printf("######### Running %s @ %s:%d\n", function_name, file, line);
+}
+
+#define BEGIN_TEST() __begin_test(__FUNCTION__, __FILE__, __LINE__)
+
+void test_fail() {
+    printf("[!] Test failed. Exiting with status code 1.\n");
+    exit(1);
+}
+
+template <typename T>
+void debug_print_array(int len, const T* as) {
+    printf("[");
+    for (int i = 0; i < len; i++) {
+        if (i != 0) printf(", ");
+        print_value(as[i]);
+    }
+    printf("]");
+}
+
+void print_assertion_passed(const char* file, int line) {
+    printf("[*] Assertion passed on %s:%d\n", file, line);
+}
+
+void print_assertion_failed(const char* file, int line) {
+    printf("[!] Assertion failed on %s:%d\n", file, line);
+}
+
+void __assert_true(const char* file, int line, bool cond) {
+    if (cond) {
+        print_assertion_passed(file, line);
+    } else {
+        print_assertion_failed(file, line);
+        test_fail();
+    }
+}
+
+#define assert_true(cond) __assert_true(__FILE__, __LINE__, cond)
+
+template <typename T>
+void __assert_arrays_match(const char* file, int line, int len,
+                           const T* expected, const T* got) {
+    if (arrays_match(len, expected, got)) {
+        print_assertion_passed(file, line);
+    } else {
+        print_assertion_failed(file, line);
+        printf("Expect = ");
+        debug_print_array(len, expected);
+        printf("\n");
+        printf("   Got = ");
+        debug_print_array(len, got);
+        printf("\n");
+        test_fail();
+    }
+}
+
+#define assert_arrays_match(len, expected, got) \
+    __assert_arrays_match(__FILE__, __LINE__, len, expected, got)
+
+template <typename T>
+void __assert_values_match(const char* file, int line, T expected, T got) {
+    if (expected == got) {
+        print_assertion_passed(file, line);
+    } else {
+        print_assertion_failed(file, line);
+        printf("Expect = ");
+        print_value(expected);
+        printf("\n");
+        printf("   Got = ");
+        print_value(got);
+        printf("\n");
+        test_fail();
+    }
+}
+
+#define assert_values_match(expected, got) \
+    __assert_values_match(__FILE__, __LINE__, expected, got)

nac3core/nac3core_derive/Cargo.toml

@@ -1,21 +0,0 @@
-[package]
-name = "nac3core_derive"
-version = "0.1.0"
-edition = "2021"
-
-[lib]
-proc-macro = true
-
-[[test]]
-name = "structfields_tests"
-path = "tests/structfields_test.rs"
-
-[dev-dependencies]
-nac3core = { path = ".." }
-trybuild = { version = "1.0", features = ["diff"] }
-
-[dependencies]
-proc-macro2 = "1.0"
-proc-macro-error = "1.0"
-syn = "2.0"
-quote = "1.0"

nac3core/nac3core_derive/src/lib.rs

@@ -1,320 +0,0 @@
-use proc_macro::TokenStream;
-use proc_macro_error::{abort, proc_macro_error};
-use quote::quote;
-use syn::{
-    parse_macro_input, spanned::Spanned, Data, DataStruct, Expr, ExprField, ExprMethodCall,
-    ExprPath, GenericArgument, Ident, LitStr, Path, PathArguments, Type, TypePath,
-};
-
-/// Extracts all generic arguments of a [`Type`] into a [`Vec`].
-///
-/// Returns [`Some`] of a possibly-empty [`Vec`] if the path of `ty` matches with
-/// `expected_ty_name`, otherwise returns [`None`].
-fn extract_generic_args(expected_ty_name: &'static str, ty: &Type) -> Option<Vec<GenericArgument>> {
-    let Type::Path(TypePath { qself: None, path, .. }) = ty else {
-        return None;
-    };
-
-    let segments = &path.segments;
-    if segments.len() != 1 {
-        return None;
-    };
-
-    let segment = segments.iter().next().unwrap();
-    if segment.ident != expected_ty_name {
-        return None;
-    }
-
-    let PathArguments::AngleBracketed(path_args) = &segment.arguments else {
-        return Some(Vec::new());
-    };
-    let args = &path_args.args;
-
-    Some(args.iter().cloned().collect::<Vec<_>>())
-}
-
-/// Maps a `path` matching one of the `target_idents` into the `replacement` [`Ident`].
-fn map_path_to_ident(path: &Path, target_idents: &[&str], replacement: &str) -> Option<Ident> {
-    path.require_ident()
-        .ok()
-        .filter(|ident| target_idents.iter().any(|target| ident == target))
-        .map(|ident| Ident::new(replacement, ident.span()))
-}
-
-/// Extracts the left-hand side of a dot-expression.
-fn extract_dot_operand(expr: &Expr) -> Option<&Expr> {
-    match expr {
-        Expr::MethodCall(ExprMethodCall { receiver: operand, .. })
-        | Expr::Field(ExprField { base: operand, .. }) => Some(operand),
-        _ => None,
-    }
-}
-
-/// Replaces the top-level receiver of a dot-expression with an [`Ident`], returning `Some(&mut expr)` if the
-/// replacement is performed.
-///
-/// The top-level receiver is the left-most receiver expression, e.g. the top-level receiver of `a.b.c.foo()` is `a`.
-fn replace_top_level_receiver(expr: &mut Expr, ident: Ident) -> Option<&mut Expr> {
-    if let Expr::MethodCall(ExprMethodCall { receiver: operand, .. })
-    | Expr::Field(ExprField { base: operand, .. }) = expr
-    {
-        return if extract_dot_operand(operand).is_some() {
-            if replace_top_level_receiver(operand, ident).is_some() {
-                Some(expr)
-            } else {
-                None
-            }
-        } else {
-            *operand = Box::new(Expr::Path(ExprPath {
-                attrs: Vec::default(),
-                qself: None,
-                path: ident.into(),
-            }));
-
-            Some(expr)
-        };
-    }
-
-    None
-}
-
-/// Iterates all operands to the left-hand side of the `.` of an [expression][`Expr`], i.e. the container operand of all
-/// [`Expr::Field`] and the receiver operand of all [`Expr::MethodCall`].
-///
-/// The iterator will return the operand expressions in reverse order of appearance. For example, `a.b.c.func()` will
-/// return `vec![c, b, a]`.
-fn iter_dot_operands(expr: &Expr) -> impl Iterator<Item = &Expr> {
-    let mut o = extract_dot_operand(expr);
-
-    std::iter::from_fn(move || {
-        let this = o;
-        o = o.as_ref().and_then(|o| extract_dot_operand(o));
-
-        this
-    })
-}
-
-/// Normalizes a value expression for use when creating an instance of this structure, returning a
-/// [`proc_macro2::TokenStream`] of tokens representing the normalized expression.
-fn normalize_value_expr(expr: &Expr) -> proc_macro2::TokenStream {
-    match &expr {
-        Expr::Path(ExprPath { qself: None, path, .. }) => {
-            if let Some(ident) = map_path_to_ident(path, &["usize", "size_t"], "llvm_usize") {
-                quote! { #ident }
-            } else {
-                abort!(
-                    path,
-                    format!(
-                        "Expected one of `size_t`, `usize`, or an implicit call expression in #[value_type(...)], found {}", 
-                        quote!(#expr).to_string(),
-                    )
-                )
-            }
-        }
-
-        Expr::Call(_) => {
-            quote! { ctx.#expr }
-        }
-
-        Expr::MethodCall(_) => {
-            let base_receiver = iter_dot_operands(expr).last();
-
-            match base_receiver {
-                // `usize.{...}`, `size_t.{...}` -> Rewrite the identifiers to `llvm_usize`
-                Some(Expr::Path(ExprPath { qself: None, path, .. }))
-                    if map_path_to_ident(path, &["usize", "size_t"], "llvm_usize").is_some() =>
-                {
-                    let ident =
-                        map_path_to_ident(path, &["usize", "size_t"], "llvm_usize").unwrap();
-
-                    let mut expr = expr.clone();
-                    let expr = replace_top_level_receiver(&mut expr, ident).unwrap();
-
-                    quote!(#expr)
-                }
-
-                // `ctx.{...}`, `context.{...}` -> Rewrite the identifiers to `ctx`
-                Some(Expr::Path(ExprPath { qself: None, path, .. }))
-                    if map_path_to_ident(path, &["ctx", "context"], "ctx").is_some() =>
-                {
-                    let ident = map_path_to_ident(path, &["ctx", "context"], "ctx").unwrap();
-
-                    let mut expr = expr.clone();
-                    let expr = replace_top_level_receiver(&mut expr, ident).unwrap();
-
-                    quote!(#expr)
-                }
-
-                // No reserved identifier prefix -> Prepend `ctx.` to the entire expression
-                _ => quote! { ctx.#expr },
-            }
-        }
-
-        _ => {
-            abort!(
-                expr,
-                format!(
-                    "Expected one of `size_t`, `usize`, or an implicit call expression in #[value_type(...)], found {}", 
-                    quote!(#expr).to_string(),
-                )
-            )
-        }
-    }
-}
-
-/// Derives an implementation of `codegen::types::structure::StructFields`.
-///
-/// The benefit of using `#[derive(StructFields)]` is that all index- or order-dependent logic required by
-/// `impl StructFields` is automatically generated by this implementation, including the field index as required by
-/// `StructField::new` and the fields as returned by `StructFields::to_vec`.
-///
-/// # Prerequisites
-///
-/// In order to derive from [`StructFields`], you must implement (or derive) [`Eq`] and [`Copy`] as required by
-/// `StructFields`.
-///
-/// Moreover, `#[derive(StructFields)]` can only be used for `struct`s with named fields, and may only contain fields
-/// with either `StructField` or [`PhantomData`] types.
-///
-/// # Attributes for [`StructFields`]
-///
-/// Each `StructField` field must be declared with the `#[value_type(...)]` attribute. The argument of `value_type`
-/// accepts one of the following:
-///
-/// - An expression returning an instance of `inkwell::types::BasicType` (with or without the receiver `ctx`/`context`).
-/// For example, `context.i8_type()`, `ctx.i8_type()`, and `i8_type()` all refer to `i8`.
-/// - The reserved identifiers `usize` and `size_t` referring to an `inkwell::types::IntType` of the platform-dependent
-/// integer size. `usize` and `size_t` can also be used as the receiver to other method calls, e.g.
-/// `usize.array_type(3)`.
-///
-/// # Example
-///
-/// The following is an example of an LLVM slice implemented using `#[derive(StructFields)]`.
-///
-/// ```rust,ignore
-/// use nac3core::{
-///     codegen::types::structure::StructField,
-///     inkwell::{
-///         values::{IntValue, PointerValue},
-///         AddressSpace,
-///     },
-/// };
-/// use nac3core_derive::StructFields;
-///
-/// // All classes that implement StructFields must also implement Eq and Copy
-/// #[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-/// pub struct SliceValue<'ctx> {
-///     // Declares ptr have a value type of i8*
-///     //
-///     // Can also be written as `ctx.i8_type().ptr_type(...)` or `context.i8_type().ptr_type(...)`
-///     #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-///     ptr: StructField<'ctx, PointerValue<'ctx>>,
-///
-///     // Declares len have a value type of usize, depending on the target compilation platform
-///     #[value_type(usize)]
-///     len: StructField<'ctx, IntValue<'ctx>>,
-/// }
-/// ```
-#[proc_macro_derive(StructFields, attributes(value_type))]
-#[proc_macro_error]
-pub fn derive(input: TokenStream) -> TokenStream {
-    let input = parse_macro_input!(input as syn::DeriveInput);
-    let ident = &input.ident;
-
-    let Data::Struct(DataStruct { fields, .. }) = &input.data else {
-        abort!(input, "Only structs with named fields are supported");
-    };
-    if let Err(err_span) =
-        fields
-            .iter()
-            .try_for_each(|field| if field.ident.is_some() { Ok(()) } else { Err(field.span()) })
-    {
-        abort!(err_span, "Only structs with named fields are supported");
-    };
-
-    // Check if struct<'ctx>
-    if input.generics.params.len() != 1 {
-        abort!(input.generics, "Expected exactly 1 generic parameter")
-    }
-
-    let phantom_info = fields
-        .iter()
-        .filter(|field| extract_generic_args("PhantomData", &field.ty).is_some())
-        .map(|field| field.ident.as_ref().unwrap())
-        .cloned()
-        .collect::<Vec<_>>();
-
-    let field_info = fields
-        .iter()
-        .filter(|field| extract_generic_args("PhantomData", &field.ty).is_none())
-        .map(|field| {
-            let ident = field.ident.as_ref().unwrap();
-            let ty = &field.ty;
-
-            let Some(_) = extract_generic_args("StructField", ty) else {
-                abort!(field, "Only StructField and PhantomData are allowed")
-            };
-
-            let attrs = &field.attrs;
-            let Some(value_type_attr) =
-                attrs.iter().find(|attr| attr.path().is_ident("value_type"))
-            else {
-                abort!(field, "Expected #[value_type(...)] attribute for field");
-            };
-
-            let Ok(value_type_expr) = value_type_attr.parse_args::<Expr>() else {
-                abort!(value_type_attr, "Expected expression in #[value_type(...)]");
-            };
-
-            let value_expr_toks = normalize_value_expr(&value_type_expr);
-
-            (ident.clone(), value_expr_toks)
-        })
-        .collect::<Vec<_>>();
-
-    // `<*>::new` impl of `StructField` and `PhantomData` for `StructFields::new`
-    let phantoms_create = phantom_info
-        .iter()
-        .map(|id| quote! { #id: ::std::marker::PhantomData })
-        .collect::<Vec<_>>();
-    let fields_create = field_info
-        .iter()
-        .map(|(id, ty)| {
-            let id_lit = LitStr::new(&id.to_string(), id.span());
-            quote! {
-                #id: ::nac3core::codegen::types::structure::StructField::create(
-                    &mut counter,
-                    #id_lit,
-                    #ty,
-                )
-            }
-        })
-        .collect::<Vec<_>>();
-
-    // `.into()` impl of `StructField` for `StructFields::to_vec`
-    let fields_into =
-        field_info.iter().map(|(id, _)| quote! { self.#id.into() }).collect::<Vec<_>>();
-
-    let impl_block = quote! {
-        impl<'ctx> ::nac3core::codegen::types::structure::StructFields<'ctx> for #ident<'ctx> {
-            fn new(ctx: impl ::nac3core::inkwell::context::AsContextRef<'ctx>, llvm_usize: ::nac3core::inkwell::types::IntType<'ctx>) -> Self {
-                let ctx = unsafe { ::nac3core::inkwell::context::ContextRef::new(ctx.as_ctx_ref()) };
-
-                let mut counter = ::nac3core::codegen::types::structure::FieldIndexCounter::default();
-
-                #ident {
-                    #(#fields_create),*
-                    #(#phantoms_create),*
-                }
-            }
-
-            fn to_vec(&self) -> ::std::vec::Vec<(&'static str, ::nac3core::inkwell::types::BasicTypeEnum<'ctx>)> {
-                vec![
-                    #(#fields_into),*
-                ]
-            }
-        }
-    };
-
-    impl_block.into()
-}

nac3core/nac3core_derive/tests/structfields_empty.rs

@@ -1,9 +0,0 @@
-use nac3core_derive::StructFields;
-use std::marker::PhantomData;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct EmptyValue<'ctx> {
-    _phantom: PhantomData<&'ctx ()>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_ndarray.rs

@@ -1,20 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        values::{IntValue, PointerValue},
-        AddressSpace,
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct NDArrayValue<'ctx> {
-    #[value_type(usize)]
-    ndims: StructField<'ctx, IntValue<'ctx>>,
-    #[value_type(usize.ptr_type(AddressSpace::default()))]
-    shape: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    data: StructField<'ctx, PointerValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        values::{IntValue, PointerValue},
-        AddressSpace,
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(usize)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice_context.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        values::{IntValue, PointerValue},
-        AddressSpace,
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(context.i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(usize)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice_ctx.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        values::{IntValue, PointerValue},
-        AddressSpace,
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(ctx.i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(usize)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_slice_sizet.rs

@@ -1,18 +0,0 @@
-use nac3core::{
-    codegen::types::structure::StructField,
-    inkwell::{
-        values::{IntValue, PointerValue},
-        AddressSpace,
-    },
-};
-use nac3core_derive::StructFields;
-
-#[derive(PartialEq, Eq, Clone, Copy, StructFields)]
-pub struct SliceValue<'ctx> {
-    #[value_type(i8_type().ptr_type(AddressSpace::default()))]
-    ptr: StructField<'ctx, PointerValue<'ctx>>,
-    #[value_type(size_t)]
-    len: StructField<'ctx, IntValue<'ctx>>,
-}
-
-fn main() {}

nac3core/nac3core_derive/tests/structfields_test.rs

@@ -1,10 +0,0 @@
-#[test]
-fn test_parse_empty() {
-    let t = trybuild::TestCases::new();
-    t.pass("tests/structfields_empty.rs");
-    t.pass("tests/structfields_slice.rs");
-    t.pass("tests/structfields_slice_ctx.rs");
-    t.pass("tests/structfields_slice_context.rs");
-    t.pass("tests/structfields_slice_sizet.rs");
-    t.pass("tests/structfields_ndarray.rs");
-}

nac3core/src/codegen/concrete_type.rs

@@ -1,9 +1,3 @@
-use std::collections::HashMap;
-
-use indexmap::IndexMap;
-
-use nac3parser::ast::StrRef;
-
 use crate::{
     symbol_resolver::SymbolValue,
     toplevel::DefinitionId,
@@ -15,6 +9,10 @@ use crate::{
     },
 };
 
+use indexmap::IndexMap;
+use nac3parser::ast::StrRef;
+use std::collections::HashMap;
+
 pub struct ConcreteTypeStore {
     store: Vec<ConcreteTypeEnum>,
 }

nac3core/src/codegen/extern_fns.rs

@@ -1,10 +1,8 @@
-use inkwell::{
-    attributes::{Attribute, AttributeLoc},
-    values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue},
-};
+use inkwell::attributes::{Attribute, AttributeLoc};
+use inkwell::values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue};
 use itertools::Either;
 
-use super::CodeGenContext;
+use crate::codegen::CodeGenContext;
 
 /// Macro to generate extern function
 /// Both function return type and function parameter type are `FloatValue`
@@ -15,11 +13,11 @@ use super::CodeGenContext;
 /// * `$extern_fn:literal`: Name of underlying extern function
 ///
 /// Optional Arguments:
-/// * `$(,$attributes:literal)*)`: Attributes linked with the extern function.
-///   The default attributes are "mustprogress", "nofree", "nounwind", "willreturn", and "writeonly".
-///   These will be used unless other attributes are specified
+/// * `$(,$attributes:literal)*)`: Attributes linked with the extern function
+/// The default attributes are "mustprogress", "nofree", "nounwind", "willreturn", and "writeonly"
+/// These will be used unless other attributes are specified
 /// * `$(,$args:ident)*`: Operands of the extern function
-///   The data type of these operands will be set to `FloatValue`
+/// The data type of these operands will be set to `FloatValue`
 ///  
 macro_rules! generate_extern_fn {
     ("unary", $fn_name:ident, $extern_fn:literal) => {

nac3core/src/codegen/generator.rs

@@ -1,18 +1,16 @@
+use crate::{
+    codegen::{bool_to_i1, bool_to_i8, classes::ArraySliceValue, expr::*, stmt::*, CodeGenContext},
+    symbol_resolver::ValueEnum,
+    toplevel::{DefinitionId, TopLevelDef},
+    typecheck::typedef::{FunSignature, Type},
+};
 use inkwell::{
     context::Context,
     types::{BasicTypeEnum, IntType},
     values::{BasicValueEnum, IntValue, PointerValue},
 };
-
 use nac3parser::ast::{Expr, Stmt, StrRef};
 
-use super::{bool_to_i1, bool_to_i8, expr::*, stmt::*, values::ArraySliceValue, CodeGenContext};
-use crate::{
-    symbol_resolver::ValueEnum,
-    toplevel::{DefinitionId, TopLevelDef},
-    typecheck::typedef::{FunSignature, Type},
-};
-
 pub trait CodeGenerator {
     /// Return the module name for the code generator.
     fn get_name(&self) -> &str;
@@ -59,7 +57,6 @@ pub trait CodeGenerator {
     /// - fun: Function signature, definition ID and the substitution key.
     /// - params: Function parameters. Note that this does not include the object even if the
     ///   function is a class method.
-    ///
     /// Note that this function should check if the function is generated in another thread (due to
     /// possible race condition), see the default implementation for an example.
     fn gen_func_instance<'ctx>(

nac3core/src/codegen/irrt/list.rs

@@ -1,162 +0,0 @@
-use inkwell::{
-    types::BasicTypeEnum,
-    values::{BasicValueEnum, CallSiteValue, IntValue},
-    AddressSpace, IntPredicate,
-};
-use itertools::Either;
-
-use super::calculate_len_for_slice_range;
-use crate::codegen::{
-    macros::codegen_unreachable,
-    values::{ArrayLikeValue, ListValue},
-    CodeGenContext, CodeGenerator,
-};
-
-/// This function handles 'end' **inclusively**.
-/// Order of tuples `assign_idx` and `value_idx` is ('start', 'end', 'step').
-/// Negative index should be handled before entering this function
-pub fn list_slice_assignment<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    ty: BasicTypeEnum<'ctx>,
-    dest_arr: ListValue<'ctx>,
-    dest_idx: (IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>),
-    src_arr: ListValue<'ctx>,
-    src_idx: (IntValue<'ctx>, IntValue<'ctx>, IntValue<'ctx>),
-) {
-    let size_ty = generator.get_size_type(ctx.ctx);
-    let int8_ptr = ctx.ctx.i8_type().ptr_type(AddressSpace::default());
-    let int32 = ctx.ctx.i32_type();
-    let (fun_symbol, elem_ptr_type) = ("__nac3_list_slice_assign_var_size", int8_ptr);
-    let slice_assign_fun = {
-        let ty_vec = vec![
-            int32.into(),         // dest start idx
-            int32.into(),         // dest end idx
-            int32.into(),         // dest step
-            elem_ptr_type.into(), // dest arr ptr
-            int32.into(),         // dest arr len
-            int32.into(),         // src start idx
-            int32.into(),         // src end idx
-            int32.into(),         // src step
-            elem_ptr_type.into(), // src arr ptr
-            int32.into(),         // src arr len
-            int32.into(),         // size
-        ];
-        ctx.module.get_function(fun_symbol).unwrap_or_else(|| {
-            let fn_t = int32.fn_type(ty_vec.as_slice(), false);
-            ctx.module.add_function(fun_symbol, fn_t, None)
-        })
-    };
-
-    let zero = int32.const_zero();
-    let one = int32.const_int(1, false);
-    let dest_arr_ptr = dest_arr.data().base_ptr(ctx, generator);
-    let dest_arr_ptr =
-        ctx.builder.build_pointer_cast(dest_arr_ptr, elem_ptr_type, "dest_arr_ptr_cast").unwrap();
-    let dest_len = dest_arr.load_size(ctx, Some("dest.len"));
-    let dest_len = ctx.builder.build_int_truncate_or_bit_cast(dest_len, int32, "srclen32").unwrap();
-    let src_arr_ptr = src_arr.data().base_ptr(ctx, generator);
-    let src_arr_ptr =
-        ctx.builder.build_pointer_cast(src_arr_ptr, elem_ptr_type, "src_arr_ptr_cast").unwrap();
-    let src_len = src_arr.load_size(ctx, Some("src.len"));
-    let src_len = ctx.builder.build_int_truncate_or_bit_cast(src_len, int32, "srclen32").unwrap();
-
-    // index in bound and positive should be done
-    // assert if dest.step == 1 then len(src) <= len(dest) else len(src) == len(dest), and
-    // throw exception if not satisfied
-    let src_end = ctx
-        .builder
-        .build_select(
-            ctx.builder.build_int_compare(IntPredicate::SLT, src_idx.2, zero, "is_neg").unwrap(),
-            ctx.builder.build_int_sub(src_idx.1, one, "e_min_one").unwrap(),
-            ctx.builder.build_int_add(src_idx.1, one, "e_add_one").unwrap(),
-            "final_e",
-        )
-        .map(BasicValueEnum::into_int_value)
-        .unwrap();
-    let dest_end = ctx
-        .builder
-        .build_select(
-            ctx.builder.build_int_compare(IntPredicate::SLT, dest_idx.2, zero, "is_neg").unwrap(),
-            ctx.builder.build_int_sub(dest_idx.1, one, "e_min_one").unwrap(),
-            ctx.builder.build_int_add(dest_idx.1, one, "e_add_one").unwrap(),
-            "final_e",
-        )
-        .map(BasicValueEnum::into_int_value)
-        .unwrap();
-    let src_slice_len =
-        calculate_len_for_slice_range(generator, ctx, src_idx.0, src_end, src_idx.2);
-    let dest_slice_len =
-        calculate_len_for_slice_range(generator, ctx, dest_idx.0, dest_end, dest_idx.2);
-    let src_eq_dest = ctx
-        .builder
-        .build_int_compare(IntPredicate::EQ, src_slice_len, dest_slice_len, "slice_src_eq_dest")
-        .unwrap();
-    let src_slt_dest = ctx
-        .builder
-        .build_int_compare(IntPredicate::SLT, src_slice_len, dest_slice_len, "slice_src_slt_dest")
-        .unwrap();
-    let dest_step_eq_one = ctx
-        .builder
-        .build_int_compare(
-            IntPredicate::EQ,
-            dest_idx.2,
-            dest_idx.2.get_type().const_int(1, false),
-            "slice_dest_step_eq_one",
-        )
-        .unwrap();
-    let cond_1 = ctx.builder.build_and(dest_step_eq_one, src_slt_dest, "slice_cond_1").unwrap();
-    let cond = ctx.builder.build_or(src_eq_dest, cond_1, "slice_cond").unwrap();
-    ctx.make_assert(
-        generator,
-        cond,
-        "0:ValueError",
-        "attempt to assign sequence of size {0} to slice of size {1} with step size {2}",
-        [Some(src_slice_len), Some(dest_slice_len), Some(dest_idx.2)],
-        ctx.current_loc,
-    );
-
-    let new_len = {
-        let args = vec![
-            dest_idx.0.into(),   // dest start idx
-            dest_idx.1.into(),   // dest end idx
-            dest_idx.2.into(),   // dest step
-            dest_arr_ptr.into(), // dest arr ptr
-            dest_len.into(),     // dest arr len
-            src_idx.0.into(),    // src start idx
-            src_idx.1.into(),    // src end idx
-            src_idx.2.into(),    // src step
-            src_arr_ptr.into(),  // src arr ptr
-            src_len.into(),      // src arr len
-            {
-                let s = match ty {
-                    BasicTypeEnum::FloatType(t) => t.size_of(),
-                    BasicTypeEnum::IntType(t) => t.size_of(),
-                    BasicTypeEnum::PointerType(t) => t.size_of(),
-                    BasicTypeEnum::StructType(t) => t.size_of().unwrap(),
-                    _ => codegen_unreachable!(ctx),
-                };
-                ctx.builder.build_int_truncate_or_bit_cast(s, int32, "size").unwrap()
-            }
-            .into(),
-        ];
-        ctx.builder
-            .build_call(slice_assign_fun, args.as_slice(), "slice_assign")
-            .map(CallSiteValue::try_as_basic_value)
-            .map(|v| v.map_left(BasicValueEnum::into_int_value))
-            .map(Either::unwrap_left)
-            .unwrap()
-    };
-    // update length
-    let need_update =
-        ctx.builder.build_int_compare(IntPredicate::NE, new_len, dest_len, "need_update").unwrap();
-    let current = ctx.builder.get_insert_block().unwrap().get_parent().unwrap();
-    let update_bb = ctx.ctx.append_basic_block(current, "update");
-    let cont_bb = ctx.ctx.append_basic_block(current, "cont");
-    ctx.builder.build_conditional_branch(need_update, update_bb, cont_bb).unwrap();
-    ctx.builder.position_at_end(update_bb);
-    let new_len = ctx.builder.build_int_z_extend_or_bit_cast(new_len, size_ty, "new_len").unwrap();
-    dest_arr.store_size(ctx, generator, new_len);
-    ctx.builder.build_unconditional_branch(cont_bb).unwrap();
-    ctx.builder.position_at_end(cont_bb);
-}

nac3core/src/codegen/irrt/math.rs

@@ -1,152 +0,0 @@
-use inkwell::{
-    values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue},
-    IntPredicate,
-};
-use itertools::Either;
-
-use crate::codegen::{
-    macros::codegen_unreachable,
-    {CodeGenContext, CodeGenerator},
-};
-
-// repeated squaring method adapted from GNU Scientific Library:
-// https://git.savannah.gnu.org/cgit/gsl.git/tree/sys/pow_int.c
-pub fn integer_power<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    base: IntValue<'ctx>,
-    exp: IntValue<'ctx>,
-    signed: bool,
-) -> IntValue<'ctx> {
-    let symbol = match (base.get_type().get_bit_width(), exp.get_type().get_bit_width(), signed) {
-        (32, 32, true) => "__nac3_int_exp_int32_t",
-        (64, 64, true) => "__nac3_int_exp_int64_t",
-        (32, 32, false) => "__nac3_int_exp_uint32_t",
-        (64, 64, false) => "__nac3_int_exp_uint64_t",
-        _ => codegen_unreachable!(ctx),
-    };
-    let base_type = base.get_type();
-    let pow_fun = ctx.module.get_function(symbol).unwrap_or_else(|| {
-        let fn_type = base_type.fn_type(&[base_type.into(), base_type.into()], false);
-        ctx.module.add_function(symbol, fn_type, None)
-    });
-    // throw exception when exp < 0
-    let ge_zero = ctx
-        .builder
-        .build_int_compare(
-            IntPredicate::SGE,
-            exp,
-            exp.get_type().const_zero(),
-            "assert_int_pow_ge_0",
-        )
-        .unwrap();
-    ctx.make_assert(
-        generator,
-        ge_zero,
-        "0:ValueError",
-        "integer power must be positive or zero",
-        [None, None, None],
-        ctx.current_loc,
-    );
-    ctx.builder
-        .build_call(pow_fun, &[base.into(), exp.into()], "call_int_pow")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}
-
-/// Generates a call to `isinf` in IR. Returns an `i1` representing the result.
-pub fn call_isinf<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    v: FloatValue<'ctx>,
-) -> IntValue<'ctx> {
-    let intrinsic_fn = ctx.module.get_function("__nac3_isinf").unwrap_or_else(|| {
-        let fn_type = ctx.ctx.i32_type().fn_type(&[ctx.ctx.f64_type().into()], false);
-        ctx.module.add_function("__nac3_isinf", fn_type, None)
-    });
-
-    let ret = ctx
-        .builder
-        .build_call(intrinsic_fn, &[v.into()], "isinf")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap();
-
-    generator.bool_to_i1(ctx, ret)
-}
-
-/// Generates a call to `isnan` in IR. Returns an `i1` representing the result.
-pub fn call_isnan<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    v: FloatValue<'ctx>,
-) -> IntValue<'ctx> {
-    let intrinsic_fn = ctx.module.get_function("__nac3_isnan").unwrap_or_else(|| {
-        let fn_type = ctx.ctx.i32_type().fn_type(&[ctx.ctx.f64_type().into()], false);
-        ctx.module.add_function("__nac3_isnan", fn_type, None)
-    });
-
-    let ret = ctx
-        .builder
-        .build_call(intrinsic_fn, &[v.into()], "isnan")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap();
-
-    generator.bool_to_i1(ctx, ret)
-}
-
-/// Generates a call to `gamma` in IR. Returns an `f64` representing the result.
-pub fn call_gamma<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    let intrinsic_fn = ctx.module.get_function("__nac3_gamma").unwrap_or_else(|| {
-        let fn_type = llvm_f64.fn_type(&[llvm_f64.into()], false);
-        ctx.module.add_function("__nac3_gamma", fn_type, None)
-    });
-
-    ctx.builder
-        .build_call(intrinsic_fn, &[v.into()], "gamma")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_float_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}
-
-/// Generates a call to `gammaln` in IR. Returns an `f64` representing the result.
-pub fn call_gammaln<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    let intrinsic_fn = ctx.module.get_function("__nac3_gammaln").unwrap_or_else(|| {
-        let fn_type = llvm_f64.fn_type(&[llvm_f64.into()], false);
-        ctx.module.add_function("__nac3_gammaln", fn_type, None)
-    });
-
-    ctx.builder
-        .build_call(intrinsic_fn, &[v.into()], "gammaln")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_float_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}
-
-/// Generates a call to `j0` in IR. Returns an `f64` representing the result.
-pub fn call_j0<'ctx>(ctx: &CodeGenContext<'ctx, '_>, v: FloatValue<'ctx>) -> FloatValue<'ctx> {
-    let llvm_f64 = ctx.ctx.f64_type();
-
-    let intrinsic_fn = ctx.module.get_function("__nac3_j0").unwrap_or_else(|| {
-        let fn_type = llvm_f64.fn_type(&[llvm_f64.into()], false);
-        ctx.module.add_function("__nac3_j0", fn_type, None)
-    });
-
-    ctx.builder
-        .build_call(intrinsic_fn, &[v.into()], "j0")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_float_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}

nac3core/src/codegen/irrt/ndarray/basic.rs

@@ -1,250 +0,0 @@
-use inkwell::{
-    values::{BasicValueEnum, IntValue, PointerValue},
-    AddressSpace,
-};
-
-use crate::codegen::{
-    expr::{create_and_call_function, infer_and_call_function},
-    irrt::get_usize_dependent_function_name,
-    types::ProxyType,
-    values::{ndarray::NDArrayValue, ProxyValue},
-    CodeGenContext, CodeGenerator,
-};
-
-pub fn call_nac3_ndarray_util_assert_shape_no_negative<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndims: IntValue<'ctx>,
-    shape: PointerValue<'ctx>,
-) {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let name = get_usize_dependent_function_name(
-        generator,
-        ctx,
-        "__nac3_ndarray_util_assert_shape_no_negative",
-    );
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[(llvm_usize.into(), ndims.into()), (llvm_pusize.into(), shape.into())],
-        None,
-        None,
-    );
-}
-
-pub fn call_nac3_ndarray_util_assert_output_shape_same<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray_ndims: IntValue<'ctx>,
-    ndarray_shape: PointerValue<'ctx>,
-    output_ndims: IntValue<'ctx>,
-    output_shape: IntValue<'ctx>,
-) {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let name = get_usize_dependent_function_name(
-        generator,
-        ctx,
-        "__nac3_ndarray_util_assert_output_shape_same",
-    );
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[
-            (llvm_usize.into(), ndarray_ndims.into()),
-            (llvm_pusize.into(), ndarray_shape.into()),
-            (llvm_usize.into(), output_ndims.into()),
-            (llvm_pusize.into(), output_shape.into()),
-        ],
-        None,
-        None,
-    );
-}
-
-pub fn call_nac3_ndarray_size<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_size");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[(llvm_ndarray.into(), ndarray.as_base_value().into())],
-        Some("size"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-pub fn call_nac3_ndarray_nbytes<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_nbytes");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[(llvm_ndarray.into(), ndarray.as_base_value().into())],
-        Some("nbytes"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-pub fn call_nac3_ndarray_len<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_len");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_usize.into()),
-        &[(llvm_ndarray.into(), ndarray.as_base_value().into())],
-        Some("len"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-pub fn call_nac3_ndarray_is_c_contiguous<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) -> IntValue<'ctx> {
-    let llvm_i1 = ctx.ctx.bool_type();
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_is_c_contiguous");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_i1.into()),
-        &[(llvm_ndarray.into(), ndarray.as_base_value().into())],
-        Some("is_c_contiguous"),
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-pub fn call_nac3_ndarray_get_nth_pelement<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-    index: IntValue<'ctx>,
-) -> PointerValue<'ctx> {
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_get_nth_pelement");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_pi8.into()),
-        &[(llvm_ndarray.into(), ndarray.as_base_value().into()), (llvm_usize.into(), index.into())],
-        Some("pelement"),
-        None,
-    )
-    .map(BasicValueEnum::into_pointer_value)
-    .unwrap()
-}
-
-pub fn call_nac3_ndarray_get_pelement_by_indices<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-    indices: PointerValue<'ctx>,
-) -> PointerValue<'ctx> {
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name =
-        get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_get_pelement_by_indices");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        Some(llvm_pi8.into()),
-        &[
-            (llvm_ndarray.into(), ndarray.as_base_value().into()),
-            (llvm_pusize.into(), indices.into()),
-        ],
-        Some("pelement"),
-        None,
-    )
-    .map(BasicValueEnum::into_pointer_value)
-    .unwrap()
-}
-
-pub fn call_nac3_ndarray_set_strides_by_shape<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-) {
-    let llvm_ndarray = ndarray.get_type().as_base_type();
-
-    let name =
-        get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_set_strides_by_shape");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[(llvm_ndarray.into(), ndarray.as_base_value().into())],
-        None,
-        None,
-    );
-}
-
-pub fn call_nac3_ndarray_copy_data<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    src_ndarray: NDArrayValue<'ctx>,
-    dst_ndarray: NDArrayValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_copy_data");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[src_ndarray.as_base_value().into(), dst_ndarray.as_base_value().into()],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/indexing.rs

@@ -1,29 +0,0 @@
-use crate::codegen::{
-    expr::infer_and_call_function,
-    irrt::get_usize_dependent_function_name,
-    values::{ndarray::NDArrayValue, ArrayLikeValue, ArraySliceValue, ProxyValue},
-    CodeGenContext, CodeGenerator,
-};
-
-pub fn call_nac3_ndarray_index<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    indices: ArraySliceValue<'ctx>,
-    src_ndarray: NDArrayValue<'ctx>,
-    dst_ndarray: NDArrayValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_ndarray_index");
-    infer_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            indices.size(ctx, generator).into(),
-            indices.base_ptr(ctx, generator).into(),
-            src_ndarray.as_base_value().into(),
-            dst_ndarray.as_base_value().into(),
-        ],
-        None,
-        None,
-    );
-}

nac3core/src/codegen/irrt/ndarray/iter.rs

@@ -1,70 +0,0 @@
-use inkwell::{
-    values::{BasicValueEnum, IntValue},
-    AddressSpace,
-};
-
-use crate::codegen::{
-    expr::{create_and_call_function, infer_and_call_function},
-    irrt::get_usize_dependent_function_name,
-    types::ProxyType,
-    values::{
-        ndarray::{NDArrayValue, NDIterValue},
-        ArrayLikeValue, ArraySliceValue, ProxyValue,
-    },
-    CodeGenContext, CodeGenerator,
-};
-
-pub fn call_nac3_nditer_initialize<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    iter: NDIterValue<'ctx>,
-    ndarray: NDArrayValue<'ctx>,
-    indices: ArraySliceValue<'ctx>,
-) {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_nditer_initialize");
-
-    create_and_call_function(
-        ctx,
-        &name,
-        None,
-        &[
-            (iter.get_type().as_base_type().into(), iter.as_base_value().into()),
-            (ndarray.get_type().as_base_type().into(), ndarray.as_base_value().into()),
-            (llvm_pusize.into(), indices.base_ptr(ctx, generator).into()),
-        ],
-        None,
-        None,
-    );
-}
-
-pub fn call_nac3_nditer_has_element<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    iter: NDIterValue<'ctx>,
-) -> IntValue<'ctx> {
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_nditer_has_element");
-
-    infer_and_call_function(
-        ctx,
-        &name,
-        Some(ctx.ctx.bool_type().into()),
-        &[iter.as_base_value().into()],
-        None,
-        None,
-    )
-    .map(BasicValueEnum::into_int_value)
-    .unwrap()
-}
-
-pub fn call_nac3_nditer_next<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    iter: NDIterValue<'ctx>,
-) {
-    let name = get_usize_dependent_function_name(generator, ctx, "__nac3_nditer_next");
-
-    infer_and_call_function(ctx, &name, None, &[iter.as_base_value().into()], None, None);
-}

nac3core/src/codegen/irrt/ndarray/mod.rs

@@ -1,391 +0,0 @@
-use inkwell::{
-    types::IntType,
-    values::{BasicValueEnum, CallSiteValue, IntValue},
-    AddressSpace, IntPredicate,
-};
-use itertools::Either;
-
-use crate::codegen::{
-    llvm_intrinsics,
-    macros::codegen_unreachable,
-    stmt::gen_for_callback_incrementing,
-    values::{
-        ndarray::NDArrayValue, ArrayLikeIndexer, ArrayLikeValue, ArraySliceValue,
-        TypedArrayLikeAccessor, TypedArrayLikeAdapter, UntypedArrayLikeAccessor,
-    },
-    CodeGenContext, CodeGenerator,
-};
-pub use basic::*;
-pub use indexing::*;
-pub use iter::*;
-
-mod basic;
-mod indexing;
-mod iter;
-
-/// Generates a call to `__nac3_ndarray_calc_size`. Returns an [`IntValue`] representing the
-/// calculated total size.
-///
-/// * `dims` - An [`ArrayLikeIndexer`] containing the size of each dimension.
-/// * `range` - The dimension index to begin and end (exclusively) calculating the dimensions for,
-///   or [`None`] if starting from the first dimension and ending at the last dimension
-///   respectively.
-pub fn call_ndarray_calc_size<'ctx, G, Dims>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    dims: &Dims,
-    (begin, end): (Option<IntValue<'ctx>>, Option<IntValue<'ctx>>),
-) -> IntValue<'ctx>
-where
-    G: CodeGenerator + ?Sized,
-    Dims: ArrayLikeIndexer<'ctx>,
-{
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let ndarray_calc_size_fn_name = match llvm_usize.get_bit_width() {
-        32 => "__nac3_ndarray_calc_size",
-        64 => "__nac3_ndarray_calc_size64",
-        bw => codegen_unreachable!(ctx, "Unsupported size type bit width: {}", bw),
-    };
-    let ndarray_calc_size_fn_t = llvm_usize.fn_type(
-        &[llvm_pusize.into(), llvm_usize.into(), llvm_usize.into(), llvm_usize.into()],
-        false,
-    );
-    let ndarray_calc_size_fn =
-        ctx.module.get_function(ndarray_calc_size_fn_name).unwrap_or_else(|| {
-            ctx.module.add_function(ndarray_calc_size_fn_name, ndarray_calc_size_fn_t, None)
-        });
-
-    let begin = begin.unwrap_or_else(|| llvm_usize.const_zero());
-    let end = end.unwrap_or_else(|| dims.size(ctx, generator));
-    ctx.builder
-        .build_call(
-            ndarray_calc_size_fn,
-            &[
-                dims.base_ptr(ctx, generator).into(),
-                dims.size(ctx, generator).into(),
-                begin.into(),
-                end.into(),
-            ],
-            "",
-        )
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}
-
-/// Generates a call to `__nac3_ndarray_calc_nd_indices`. Returns a [`TypeArrayLikeAdpater`]
-/// containing `i32` indices of the flattened index.
-///
-/// * `index` - The index to compute the multidimensional index for.
-/// * `ndarray` - LLVM pointer to the `NDArray`. This value must be the LLVM representation of an
-///   `NDArray`.
-pub fn call_ndarray_calc_nd_indices<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    index: IntValue<'ctx>,
-    ndarray: NDArrayValue<'ctx>,
-) -> TypedArrayLikeAdapter<'ctx, IntValue<'ctx>> {
-    let llvm_void = ctx.ctx.void_type();
-    let llvm_i32 = ctx.ctx.i32_type();
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pi32 = llvm_i32.ptr_type(AddressSpace::default());
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let ndarray_calc_nd_indices_fn_name = match llvm_usize.get_bit_width() {
-        32 => "__nac3_ndarray_calc_nd_indices",
-        64 => "__nac3_ndarray_calc_nd_indices64",
-        bw => codegen_unreachable!(ctx, "Unsupported size type bit width: {}", bw),
-    };
-    let ndarray_calc_nd_indices_fn =
-        ctx.module.get_function(ndarray_calc_nd_indices_fn_name).unwrap_or_else(|| {
-            let fn_type = llvm_void.fn_type(
-                &[llvm_usize.into(), llvm_pusize.into(), llvm_usize.into(), llvm_pi32.into()],
-                false,
-            );
-
-            ctx.module.add_function(ndarray_calc_nd_indices_fn_name, fn_type, None)
-        });
-
-    let ndarray_num_dims = ndarray.load_ndims(ctx);
-    let ndarray_dims = ndarray.shape();
-
-    let indices = ctx.builder.build_array_alloca(llvm_i32, ndarray_num_dims, "").unwrap();
-
-    ctx.builder
-        .build_call(
-            ndarray_calc_nd_indices_fn,
-            &[
-                index.into(),
-                ndarray_dims.base_ptr(ctx, generator).into(),
-                ndarray_num_dims.into(),
-                indices.into(),
-            ],
-            "",
-        )
-        .unwrap();
-
-    TypedArrayLikeAdapter::from(
-        ArraySliceValue::from_ptr_val(indices, ndarray_num_dims, None),
-        Box::new(|_, v| v.into_int_value()),
-        Box::new(|_, v| v.into()),
-    )
-}
-
-fn call_ndarray_flatten_index_impl<'ctx, G, Indices>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-    indices: &Indices,
-) -> IntValue<'ctx>
-where
-    G: CodeGenerator + ?Sized,
-    Indices: ArrayLikeIndexer<'ctx>,
-{
-    let llvm_i32 = ctx.ctx.i32_type();
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-
-    let llvm_pi32 = llvm_i32.ptr_type(AddressSpace::default());
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    debug_assert_eq!(
-        IntType::try_from(indices.element_type(ctx, generator))
-            .map(IntType::get_bit_width)
-            .unwrap_or_default(),
-        llvm_i32.get_bit_width(),
-        "Expected i32 value for argument `indices` to `call_ndarray_flatten_index_impl`"
-    );
-    debug_assert_eq!(
-        indices.size(ctx, generator).get_type().get_bit_width(),
-        llvm_usize.get_bit_width(),
-        "Expected usize integer value for argument `indices_size` to `call_ndarray_flatten_index_impl`"
-    );
-
-    let ndarray_flatten_index_fn_name = match llvm_usize.get_bit_width() {
-        32 => "__nac3_ndarray_flatten_index",
-        64 => "__nac3_ndarray_flatten_index64",
-        bw => codegen_unreachable!(ctx, "Unsupported size type bit width: {}", bw),
-    };
-    let ndarray_flatten_index_fn =
-        ctx.module.get_function(ndarray_flatten_index_fn_name).unwrap_or_else(|| {
-            let fn_type = llvm_usize.fn_type(
-                &[llvm_pusize.into(), llvm_usize.into(), llvm_pi32.into(), llvm_usize.into()],
-                false,
-            );
-
-            ctx.module.add_function(ndarray_flatten_index_fn_name, fn_type, None)
-        });
-
-    let ndarray_num_dims = ndarray.load_ndims(ctx);
-    let ndarray_dims = ndarray.shape();
-
-    let index = ctx
-        .builder
-        .build_call(
-            ndarray_flatten_index_fn,
-            &[
-                ndarray_dims.base_ptr(ctx, generator).into(),
-                ndarray_num_dims.into(),
-                indices.base_ptr(ctx, generator).into(),
-                indices.size(ctx, generator).into(),
-            ],
-            "",
-        )
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap();
-
-    index
-}
-
-/// Generates a call to `__nac3_ndarray_flatten_index`. Returns the flattened index for the
-/// multidimensional index.
-///
-/// * `ndarray` - LLVM pointer to the `NDArray`. This value must be the LLVM representation of an
-///   `NDArray`.
-/// * `indices` - The multidimensional index to compute the flattened index for.
-pub fn call_ndarray_flatten_index<'ctx, G, Index>(
-    generator: &G,
-    ctx: &CodeGenContext<'ctx, '_>,
-    ndarray: NDArrayValue<'ctx>,
-    indices: &Index,
-) -> IntValue<'ctx>
-where
-    G: CodeGenerator + ?Sized,
-    Index: ArrayLikeIndexer<'ctx>,
-{
-    call_ndarray_flatten_index_impl(generator, ctx, ndarray, indices)
-}
-
-/// Generates a call to `__nac3_ndarray_calc_broadcast`. Returns a tuple containing the number of
-/// dimension and size of each dimension of the resultant `ndarray`.
-pub fn call_ndarray_calc_broadcast<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    lhs: NDArrayValue<'ctx>,
-    rhs: NDArrayValue<'ctx>,
-) -> TypedArrayLikeAdapter<'ctx, IntValue<'ctx>> {
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let ndarray_calc_broadcast_fn_name = match llvm_usize.get_bit_width() {
-        32 => "__nac3_ndarray_calc_broadcast",
-        64 => "__nac3_ndarray_calc_broadcast64",
-        bw => codegen_unreachable!(ctx, "Unsupported size type bit width: {}", bw),
-    };
-    let ndarray_calc_broadcast_fn =
-        ctx.module.get_function(ndarray_calc_broadcast_fn_name).unwrap_or_else(|| {
-            let fn_type = llvm_usize.fn_type(
-                &[
-                    llvm_pusize.into(),
-                    llvm_usize.into(),
-                    llvm_pusize.into(),
-                    llvm_usize.into(),
-                    llvm_pusize.into(),
-                ],
-                false,
-            );
-
-            ctx.module.add_function(ndarray_calc_broadcast_fn_name, fn_type, None)
-        });
-
-    let lhs_ndims = lhs.load_ndims(ctx);
-    let rhs_ndims = rhs.load_ndims(ctx);
-    let min_ndims = llvm_intrinsics::call_int_umin(ctx, lhs_ndims, rhs_ndims, None);
-
-    gen_for_callback_incrementing(
-        generator,
-        ctx,
-        None,
-        llvm_usize.const_zero(),
-        (min_ndims, false),
-        |generator, ctx, _, idx| {
-            let idx = ctx.builder.build_int_sub(min_ndims, idx, "").unwrap();
-            let (lhs_dim_sz, rhs_dim_sz) = unsafe {
-                (
-                    lhs.shape().get_typed_unchecked(ctx, generator, &idx, None),
-                    rhs.shape().get_typed_unchecked(ctx, generator, &idx, None),
-                )
-            };
-
-            let llvm_usize_const_one = llvm_usize.const_int(1, false);
-            let lhs_eqz = ctx
-                .builder
-                .build_int_compare(IntPredicate::EQ, lhs_dim_sz, llvm_usize_const_one, "")
-                .unwrap();
-            let rhs_eqz = ctx
-                .builder
-                .build_int_compare(IntPredicate::EQ, rhs_dim_sz, llvm_usize_const_one, "")
-                .unwrap();
-            let lhs_or_rhs_eqz = ctx.builder.build_or(lhs_eqz, rhs_eqz, "").unwrap();
-
-            let lhs_eq_rhs = ctx
-                .builder
-                .build_int_compare(IntPredicate::EQ, lhs_dim_sz, rhs_dim_sz, "")
-                .unwrap();
-
-            let is_compatible = ctx.builder.build_or(lhs_or_rhs_eqz, lhs_eq_rhs, "").unwrap();
-
-            ctx.make_assert(
-                generator,
-                is_compatible,
-                "0:ValueError",
-                "operands could not be broadcast together",
-                [None, None, None],
-                ctx.current_loc,
-            );
-
-            Ok(())
-        },
-        llvm_usize.const_int(1, false),
-    )
-    .unwrap();
-
-    let max_ndims = llvm_intrinsics::call_int_umax(ctx, lhs_ndims, rhs_ndims, None);
-    let lhs_dims = lhs.shape().base_ptr(ctx, generator);
-    let lhs_ndims = lhs.load_ndims(ctx);
-    let rhs_dims = rhs.shape().base_ptr(ctx, generator);
-    let rhs_ndims = rhs.load_ndims(ctx);
-    let out_dims = ctx.builder.build_array_alloca(llvm_usize, max_ndims, "").unwrap();
-    let out_dims = ArraySliceValue::from_ptr_val(out_dims, max_ndims, None);
-
-    ctx.builder
-        .build_call(
-            ndarray_calc_broadcast_fn,
-            &[
-                lhs_dims.into(),
-                lhs_ndims.into(),
-                rhs_dims.into(),
-                rhs_ndims.into(),
-                out_dims.base_ptr(ctx, generator).into(),
-            ],
-            "",
-        )
-        .unwrap();
-
-    TypedArrayLikeAdapter::from(
-        out_dims,
-        Box::new(|_, v| v.into_int_value()),
-        Box::new(|_, v| v.into()),
-    )
-}
-
-/// Generates a call to `__nac3_ndarray_calc_broadcast_idx`. Returns an [`ArrayAllocaValue`]
-/// containing the indices used for accessing `array` corresponding to the index of the broadcasted
-/// array `broadcast_idx`.
-pub fn call_ndarray_calc_broadcast_index<
-    'ctx,
-    G: CodeGenerator + ?Sized,
-    BroadcastIdx: UntypedArrayLikeAccessor<'ctx>,
->(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    array: NDArrayValue<'ctx>,
-    broadcast_idx: &BroadcastIdx,
-) -> TypedArrayLikeAdapter<'ctx, IntValue<'ctx>> {
-    let llvm_i32 = ctx.ctx.i32_type();
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let llvm_pi32 = llvm_i32.ptr_type(AddressSpace::default());
-    let llvm_pusize = llvm_usize.ptr_type(AddressSpace::default());
-
-    let ndarray_calc_broadcast_fn_name = match llvm_usize.get_bit_width() {
-        32 => "__nac3_ndarray_calc_broadcast_idx",
-        64 => "__nac3_ndarray_calc_broadcast_idx64",
-        bw => codegen_unreachable!(ctx, "Unsupported size type bit width: {}", bw),
-    };
-    let ndarray_calc_broadcast_fn =
-        ctx.module.get_function(ndarray_calc_broadcast_fn_name).unwrap_or_else(|| {
-            let fn_type = llvm_usize.fn_type(
-                &[llvm_pusize.into(), llvm_usize.into(), llvm_pi32.into(), llvm_pi32.into()],
-                false,
-            );
-
-            ctx.module.add_function(ndarray_calc_broadcast_fn_name, fn_type, None)
-        });
-
-    let broadcast_size = broadcast_idx.size(ctx, generator);
-    let out_idx = ctx.builder.build_array_alloca(llvm_i32, broadcast_size, "").unwrap();
-
-    let array_dims = array.shape().base_ptr(ctx, generator);
-    let array_ndims = array.load_ndims(ctx);
-    let broadcast_idx_ptr = unsafe {
-        broadcast_idx.ptr_offset_unchecked(ctx, generator, &llvm_usize.const_zero(), None)
-    };
-
-    ctx.builder
-        .build_call(
-            ndarray_calc_broadcast_fn,
-            &[array_dims.into(), array_ndims.into(), broadcast_idx_ptr.into(), out_idx.into()],
-            "",
-        )
-        .unwrap();
-
-    TypedArrayLikeAdapter::from(
-        ArraySliceValue::from_ptr_val(out_idx, broadcast_size, None),
-        Box::new(|_, v| v.into_int_value()),
-        Box::new(|_, v| v.into()),
-    )
-}

nac3core/src/codegen/irrt/range.rs

@@ -1,42 +0,0 @@
-use inkwell::{
-    values::{BasicValueEnum, CallSiteValue, IntValue},
-    IntPredicate,
-};
-use itertools::Either;
-
-use crate::codegen::{CodeGenContext, CodeGenerator};
-
-pub fn calculate_len_for_slice_range<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    start: IntValue<'ctx>,
-    end: IntValue<'ctx>,
-    step: IntValue<'ctx>,
-) -> IntValue<'ctx> {
-    const SYMBOL: &str = "__nac3_range_slice_len";
-    let len_func = ctx.module.get_function(SYMBOL).unwrap_or_else(|| {
-        let i32_t = ctx.ctx.i32_type();
-        let fn_t = i32_t.fn_type(&[i32_t.into(), i32_t.into(), i32_t.into()], false);
-        ctx.module.add_function(SYMBOL, fn_t, None)
-    });
-
-    // assert step != 0, throw exception if not
-    let not_zero = ctx
-        .builder
-        .build_int_compare(IntPredicate::NE, step, step.get_type().const_zero(), "range_step_ne")
-        .unwrap();
-    ctx.make_assert(
-        generator,
-        not_zero,
-        "0:ValueError",
-        "step must not be zero",
-        [None, None, None],
-        ctx.current_loc,
-    );
-    ctx.builder
-        .build_call(len_func, &[start.into(), end.into(), step.into()], "calc_len")
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}

nac3core/src/codegen/irrt/slice.rs

@@ -1,39 +0,0 @@
-use inkwell::values::{BasicValueEnum, CallSiteValue, IntValue};
-use itertools::Either;
-
-use nac3parser::ast::Expr;
-
-use crate::{
-    codegen::{CodeGenContext, CodeGenerator},
-    typecheck::typedef::Type,
-};
-
-/// this function allows index out of range, since python
-/// allows index out of range in slice (`a = [1,2,3]; a[1:10] == [2,3]`).
-pub fn handle_slice_index_bound<'ctx, G: CodeGenerator>(
-    i: &Expr<Option<Type>>,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    generator: &mut G,
-    length: IntValue<'ctx>,
-) -> Result<Option<IntValue<'ctx>>, String> {
-    const SYMBOL: &str = "__nac3_slice_index_bound";
-    let func = ctx.module.get_function(SYMBOL).unwrap_or_else(|| {
-        let i32_t = ctx.ctx.i32_type();
-        let fn_t = i32_t.fn_type(&[i32_t.into(), i32_t.into()], false);
-        ctx.module.add_function(SYMBOL, fn_t, None)
-    });
-
-    let i = if let Some(v) = generator.gen_expr(ctx, i)? {
-        v.to_basic_value_enum(ctx, generator, i.custom.unwrap())?
-    } else {
-        return Ok(None);
-    };
-    Ok(Some(
-        ctx.builder
-            .build_call(func, &[i.into(), length.into()], "bounded_ind")
-            .map(CallSiteValue::try_as_basic_value)
-            .map(|v| v.map_left(BasicValueEnum::into_int_value))
-            .map(Either::unwrap_left)
-            .unwrap(),
-    ))
-}

nac3core/src/codegen/irrt/test.rs

@@ -0,0 +1,26 @@
+#[cfg(test)]
+mod tests {
+    use std::{path::Path, process::Command};
+
+    #[test]
+    fn run_irrt_test() {
+        assert!(
+            cfg!(feature = "test"),
+            "Please do `cargo test -F test` to compile `irrt_test.out` and run test"
+        );
+
+        let irrt_test_out_path = Path::new(concat!(env!("OUT_DIR"), "/irrt_test.out"));
+        let output = Command::new(irrt_test_out_path.to_str().unwrap()).output().unwrap();
+
+        if !output.status.success() {
+            eprintln!("irrt_test failed with status {}:", output.status);
+            eprintln!("====== stdout ======");
+            eprintln!("{}", String::from_utf8(output.stdout).unwrap());
+            eprintln!("====== stderr ======");
+            eprintln!("{}", String::from_utf8(output.stderr).unwrap());
+            eprintln!("====================");
+
+            panic!("irrt_test failed");
+        }
+    }
+}

nac3core/src/codegen/irrt/util.rs

@@ -0,0 +1,109 @@
+use crate::codegen::{CodeGenContext, CodeGenerator};
+
+// When [`TypeContext::size_type`] is 32-bits, the function name is "{fn_name}".
+// When [`TypeContext::size_type`] is 64-bits, the function name is "{fn_name}64".
+#[must_use]
+pub fn get_sizet_dependent_function_name<G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &CodeGenContext<'_, '_>,
+    name: &str,
+) -> String {
+    let mut name = name.to_owned();
+    match generator.get_size_type(ctx.ctx).get_bit_width() {
+        32 => {}
+        64 => name.push_str("64"),
+        bit_width => {
+            panic!("Unsupported int type bit width {bit_width}, must be either 32-bits or 64-bits")
+        }
+    }
+    name
+}
+
+pub mod function {
+    use crate::codegen::{model::*, CodeGenContext, CodeGenerator};
+    use inkwell::{
+        types::{BasicMetadataTypeEnum, BasicType, FunctionType},
+        values::{AnyValue, BasicMetadataValueEnum, BasicValue, BasicValueEnum, CallSiteValue},
+    };
+    use itertools::Itertools;
+
+    #[derive(Debug, Clone, Copy)]
+    struct Arg<'ctx> {
+        ty: BasicMetadataTypeEnum<'ctx>,
+        val: BasicMetadataValueEnum<'ctx>,
+    }
+
+    /// Helper structure to reduce IRRT Inkwell function call boilerplate
+    pub struct CallFunction<'ctx, 'a, 'b, 'c, 'd, G: CodeGenerator + ?Sized> {
+        generator: &'d mut G,
+        ctx: &'b CodeGenContext<'ctx, 'a>,
+        /// Function name
+        name: &'c str,
+        /// Call arguments
+        args: Vec<Arg<'ctx>>,
+    }
+
+    impl<'ctx, 'a, 'b, 'c, 'd, G: CodeGenerator + ?Sized> CallFunction<'ctx, 'a, 'b, 'c, 'd, G> {
+        pub fn begin(
+            generator: &'d mut G,
+            ctx: &'b CodeGenContext<'ctx, 'a>,
+            name: &'c str,
+        ) -> Self {
+            CallFunction { generator, ctx, name, args: Vec::new() }
+        }
+
+        /// Push a call argument to the function call.
+        ///
+        /// The `_name` parameter is there for self-documentation purposes.
+        #[allow(clippy::needless_pass_by_value)]
+        #[must_use]
+        pub fn arg<M: Model<'ctx>>(mut self, _name: &str, arg: Instance<'ctx, M>) -> Self {
+            let arg = Arg {
+                ty: arg.model.get_type(self.generator, self.ctx.ctx).as_basic_type_enum().into(),
+                val: arg.value.as_basic_value_enum().into(),
+            };
+            self.args.push(arg);
+            self
+        }
+
+        /// Call the function and expect the function to return a value of type of `return_model`.
+        #[must_use]
+        pub fn returning<M: Model<'ctx>>(self, name: &str, return_model: M) -> Instance<'ctx, M> {
+            let ret_ty = return_model.get_type(self.generator, self.ctx.ctx);
+
+            let ret = self.get_function(|tys| ret_ty.fn_type(tys, false), name);
+            let ret = BasicValueEnum::try_from(ret.as_any_value_enum()).unwrap(); // Must work
+            let ret = return_model.check_value(self.generator, self.ctx.ctx, ret).unwrap(); // Must work
+            ret
+        }
+
+        /// Like [`CallFunction::returning_`] but `return_model` is automatically inferred.
+        #[must_use]
+        pub fn returning_auto<M: Model<'ctx> + Default>(self, name: &str) -> Instance<'ctx, M> {
+            self.returning(name, M::default())
+        }
+
+        /// Call the function and expect the function to return a void-type.
+        pub fn returning_void(self) {
+            let ret_ty = self.ctx.ctx.void_type();
+
+            let _ = self.get_function(|tys| ret_ty.fn_type(tys, false), "");
+        }
+
+        fn get_function<F>(&self, make_fn_type: F, return_value_name: &str) -> CallSiteValue<'ctx>
+        where
+            F: FnOnce(&[BasicMetadataTypeEnum<'ctx>]) -> FunctionType<'ctx>,
+        {
+            // Get the LLVM function, declare the function if it doesn't exist - it will be defined by other
+            // components of NAC3.
+            let func = self.ctx.module.get_function(self.name).unwrap_or_else(|| {
+                let tys = self.args.iter().map(|arg| arg.ty).collect_vec();
+                let fn_type = make_fn_type(&tys);
+                self.ctx.module.add_function(self.name, fn_type, None)
+            });
+
+            let vals = self.args.iter().map(|arg| arg.val).collect_vec();
+            self.ctx.builder.build_call(func, &vals, return_value_name).unwrap()
+        }
+    }
+}

nac3core/src/codegen/llvm_intrinsics.rs

@@ -1,14 +1,12 @@
-use inkwell::{
-    context::Context,
-    intrinsics::Intrinsic,
-    types::{AnyTypeEnum::IntType, FloatType},
-    values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue, PointerValue},
-    AddressSpace,
-};
+use crate::codegen::CodeGenContext;
+use inkwell::context::Context;
+use inkwell::intrinsics::Intrinsic;
+use inkwell::types::AnyTypeEnum::IntType;
+use inkwell::types::FloatType;
+use inkwell::values::{BasicValueEnum, CallSiteValue, FloatValue, IntValue, PointerValue};
+use inkwell::AddressSpace;
 use itertools::Either;
 
-use super::CodeGenContext;
-
 /// Returns the string representation for the floating-point type `ft` when used in intrinsic
 /// functions.
 fn get_float_intrinsic_repr(ctx: &Context, ft: FloatType) -> &'static str {
@@ -185,7 +183,7 @@ pub fn call_memcpy_generic<'ctx>(
         dest
     } else {
         ctx.builder
-            .build_bit_cast(dest, llvm_p0i8, "")
+            .build_bitcast(dest, llvm_p0i8, "")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap()
     };
@@ -193,7 +191,7 @@ pub fn call_memcpy_generic<'ctx>(
         src
     } else {
         ctx.builder
-            .build_bit_cast(src, llvm_p0i8, "")
+            .build_bitcast(src, llvm_p0i8, "")
             .map(BasicValueEnum::into_pointer_value)
             .unwrap()
     };
@@ -201,58 +199,14 @@ pub fn call_memcpy_generic<'ctx>(
     call_memcpy(ctx, dest, src, len, is_volatile);
 }
 
-/// Invokes the `llvm.memcpy` intrinsic.
-///
-/// Unlike [`call_memcpy`], this function accepts any type of pointer value. If `dest` or `src` is
-/// not a pointer to an integer, the pointer(s) will be cast to `i8*` before invoking `memcpy`.
-/// Moreover, `len` now refers to the number of elements to copy (rather than number of bytes to
-/// copy).
-pub fn call_memcpy_generic_array<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    dest: PointerValue<'ctx>,
-    src: PointerValue<'ctx>,
-    len: IntValue<'ctx>,
-    is_volatile: IntValue<'ctx>,
-) {
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_p0i8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_sizeof_expr_t = llvm_i8.size_of().get_type();
-
-    let dest_elem_t = dest.get_type().get_element_type();
-    let src_elem_t = src.get_type().get_element_type();
-
-    let dest = if matches!(dest_elem_t, IntType(t) if t.get_bit_width() == 8) {
-        dest
-    } else {
-        ctx.builder
-            .build_bit_cast(dest, llvm_p0i8, "")
-            .map(BasicValueEnum::into_pointer_value)
-            .unwrap()
-    };
-    let src = if matches!(src_elem_t, IntType(t) if t.get_bit_width() == 8) {
-        src
-    } else {
-        ctx.builder
-            .build_bit_cast(src, llvm_p0i8, "")
-            .map(BasicValueEnum::into_pointer_value)
-            .unwrap()
-    };
-
-    let len = ctx.builder.build_int_z_extend_or_bit_cast(len, llvm_sizeof_expr_t, "").unwrap();
-    let len = ctx.builder.build_int_mul(len, src_elem_t.size_of().unwrap(), "").unwrap();
-
-    call_memcpy(ctx, dest, src, len, is_volatile);
-}
-
 /// Macro to find and generate build call for llvm intrinsic (body of llvm intrinsic function)
 ///
 /// Arguments:
 /// * `$ctx:ident`: Reference to the current Code Generation Context
 /// * `$name:ident`: Optional name to be assigned to the llvm build call (Option<&str>)
 /// * `$llvm_name:literal`: Name of underlying llvm intrinsic function
-/// * `$map_fn:ident`: Mapping function to be applied on `BasicValue` (`BasicValue` -> Function Return Type).
-///   Use `BasicValueEnum::into_int_value` for Integer return type and
-///   `BasicValueEnum::into_float_value` for Float return type
+/// * `$map_fn:ident`: Mapping function to be applied on `BasicValue` (`BasicValue` -> Function Return Type)
+/// Use `BasicValueEnum::into_int_value` for Integer return type and `BasicValueEnum::into_float_value` for Float return type
 /// * `$llvm_ty:ident`: Type of first operand
 /// * `,($val:ident)*`: Comma separated list of operands
 macro_rules! generate_llvm_intrinsic_fn_body {
@@ -268,8 +222,8 @@ macro_rules! generate_llvm_intrinsic_fn_body {
 /// Arguments:
 /// * `float/int`: Indicates the return and argument type of the function
 /// * `$fn_name:ident`: The identifier of the rust function to be generated
-/// * `$llvm_name:literal`: Name of underlying llvm intrinsic function.
-///   Omit "llvm." prefix from the function name i.e. use "ceil" instead of "llvm.ceil"
+/// * `$llvm_name:literal`: Name of underlying llvm intrinsic function
+/// Omit "llvm." prefix from the function name i.e. use "ceil" instead of "llvm.ceil"
 /// * `$val:ident`: The operand for unary operations
 /// * `$val1:ident`, `$val2:ident`: The operands for binary operations
 macro_rules! generate_llvm_intrinsic_fn {
@@ -386,25 +340,3 @@ pub fn call_float_powi<'ctx>(
         .map(Either::unwrap_left)
         .unwrap()
 }
-
-/// Invokes the [`llvm.ctpop`](https://llvm.org/docs/LangRef.html#llvm-ctpop-intrinsic) intrinsic.
-pub fn call_int_ctpop<'ctx>(
-    ctx: &CodeGenContext<'ctx, '_>,
-    src: IntValue<'ctx>,
-    name: Option<&str>,
-) -> IntValue<'ctx> {
-    const FN_NAME: &str = "llvm.ctpop";
-
-    let llvm_src_t = src.get_type();
-
-    let intrinsic_fn = Intrinsic::find(FN_NAME)
-        .and_then(|intrinsic| intrinsic.get_declaration(&ctx.module, &[llvm_src_t.into()]))
-        .unwrap();
-
-    ctx.builder
-        .build_call(intrinsic_fn, &[src.into()], name.unwrap_or_default())
-        .map(CallSiteValue::try_as_basic_value)
-        .map(|v| v.map_left(BasicValueEnum::into_int_value))
-        .map(Either::unwrap_left)
-        .unwrap()
-}

nac3core/src/codegen/mod.rs

@@ -1,12 +1,12 @@
-use std::{
-    collections::{HashMap, HashSet},
-    sync::{
-        atomic::{AtomicBool, Ordering},
-        Arc,
+use crate::{
+    codegen::classes::{ListType, ProxyType, RangeType},
+    symbol_resolver::{StaticValue, SymbolResolver},
+    toplevel::{helper::PrimDef, TopLevelContext, TopLevelDef},
+    typecheck::{
+        type_inferencer::{CodeLocation, PrimitiveStore},
+        typedef::{CallId, FuncArg, Type, TypeEnum, Unifier},
     },
-    thread,
 };
-
 use crossbeam::channel::{unbounded, Receiver, Sender};
 use inkwell::{
     attributes::{Attribute, AttributeLoc},
@@ -24,56 +24,36 @@ use inkwell::{
     AddressSpace, IntPredicate, OptimizationLevel,
 };
 use itertools::Itertools;
-use parking_lot::{Condvar, Mutex};
-
+use model::*;
 use nac3parser::ast::{Location, Stmt, StrRef};
-
-use crate::{
-    symbol_resolver::{StaticValue, SymbolResolver},
-    toplevel::{
-        helper::{extract_ndims, PrimDef},
-        numpy::unpack_ndarray_var_tys,
-        TopLevelContext, TopLevelDef,
-    },
-    typecheck::{
-        type_inferencer::{CodeLocation, PrimitiveStore},
-        typedef::{CallId, FuncArg, Type, TypeEnum, Unifier},
-    },
+use parking_lot::{Condvar, Mutex};
+use std::collections::{HashMap, HashSet};
+use std::sync::{
+    atomic::{AtomicBool, Ordering},
+    Arc,
 };
-use concrete_type::{ConcreteType, ConcreteTypeEnum, ConcreteTypeStore};
-pub use generator::{CodeGenerator, DefaultCodeGenerator};
-use types::{ndarray::NDArrayType, ListType, ProxyType, RangeType};
+use std::thread;
+use structure::{cslice::CSlice, exception::Exception, ndarray::NpArray};
 
 pub mod builtin_fns;
+pub mod classes;
 pub mod concrete_type;
 pub mod expr;
 pub mod extern_fns;
 mod generator;
 pub mod irrt;
 pub mod llvm_intrinsics;
+pub mod model;
 pub mod numpy;
+pub mod numpy_new;
 pub mod stmt;
-pub mod types;
-pub mod values;
+pub mod structure;
 
 #[cfg(test)]
 mod test;
 
-mod macros {
-    /// Codegen-variant of [`std::unreachable`] which accepts an instance of [`CodeGenContext`] as
-    /// its first argument to provide Python source information to indicate the codegen location
-    /// causing the assertion.
-    macro_rules! codegen_unreachable {
-        ($ctx:expr $(,)?) => {
-            std::unreachable!("unreachable code while processing {}", &$ctx.current_loc)
-        };
-        ($ctx:expr, $($arg:tt)*) => {
-            std::unreachable!("unreachable code while processing {}: {}", &$ctx.current_loc, std::format!("{}", std::format_args!($($arg)+)))
-        };
-    }
-
-    pub(crate) use codegen_unreachable;
-}
+use concrete_type::{ConcreteType, ConcreteTypeEnum, ConcreteTypeStore};
+pub use generator::{CodeGenerator, DefaultCodeGenerator};
 
 #[derive(Default)]
 pub struct StaticValueStore {
@@ -193,11 +173,11 @@ pub struct CodeGenContext<'ctx, 'a> {
     pub registry: &'a WorkerRegistry,
 
     /// Cache for constant strings.
-    pub const_strings: HashMap<String, BasicValueEnum<'ctx>>,
+    pub const_strings: HashMap<String, Struct<'ctx, CSlice>>,
 
     /// [`BasicBlock`] containing all `alloca` statements for the current function.
     pub init_bb: BasicBlock<'ctx>,
-    pub exception_val: Option<PointerValue<'ctx>>,
+    pub exception_val: Option<Ptr<'ctx, StructModel<Exception>>>,
 
     /// The header and exit basic blocks of a loop in this context. See
     /// <https://llvm.org/docs/LoopTerminology.html> for explanation of these terminology.
@@ -514,13 +494,8 @@ fn get_llvm_type<'ctx, G: CodeGenerator + ?Sized>(
                         }
 
                         TObj { obj_id, .. } if *obj_id == PrimDef::NDArray.id() => {
-                            let (dtype, ndims) = unpack_ndarray_var_tys(unifier, ty);
-                            let ndims = extract_ndims(unifier, ndims);
-                            let element_type = get_llvm_type(
-                                ctx, module, generator, unifier, top_level, type_cache, dtype,
-                            );
-
-                            NDArrayType::new(generator, ctx, element_type, Some(ndims)).as_base_type().into()
+                            let pndarray_model = PtrModel(StructModel(NpArray));
+                            pndarray_model.get_type(generator, ctx).as_basic_type_enum()
                         }
 
                         _ => unreachable!(
@@ -606,11 +581,11 @@ fn get_llvm_abi_type<'ctx, G: CodeGenerator + ?Sized>(
 ) -> BasicTypeEnum<'ctx> {
     // If the type is used in the definition of a function, return `i1` instead of `i8` for ABI
     // consistency.
-    if unifier.unioned(ty, primitives.bool) {
+    return if unifier.unioned(ty, primitives.bool) {
         ctx.bool_type().into()
     } else {
         get_llvm_type(ctx, module, generator, unifier, top_level, type_cache, ty)
-    }
+    };
 }
 
 /// Whether `sret` is needed for a return value with type `ty`.
@@ -726,43 +701,19 @@ pub fn gen_func_impl<
         ..primitives
     };
 
-    let mut type_cache: HashMap<_, _> = [
+    let cslice_model = StructModel(CSlice);
+    let pexn_model = PtrModel(StructModel(Exception));
+
+    let mut type_cache: HashMap<_, BasicTypeEnum<'ctx>> = [
         (primitives.int32, context.i32_type().into()),
         (primitives.int64, context.i64_type().into()),
         (primitives.uint32, context.i32_type().into()),
         (primitives.uint64, context.i64_type().into()),
         (primitives.float, context.f64_type().into()),
         (primitives.bool, context.i8_type().into()),
-        (primitives.str, {
-            let name = "str";
-            match module.get_struct_type(name) {
-                None => {
-                    let str_type = context.opaque_struct_type("str");
-                    let fields = [
-                        context.i8_type().ptr_type(AddressSpace::default()).into(),
-                        generator.get_size_type(context).into(),
-                    ];
-                    str_type.set_body(&fields, false);
-                    str_type.into()
-                }
-                Some(t) => t.as_basic_type_enum(),
-            }
-        }),
+        (primitives.str, cslice_model.get_type(generator, context).into()),
         (primitives.range, RangeType::new(context).as_base_type().into()),
-        (primitives.exception, {
-            let name = "Exception";
-            if let Some(t) = module.get_struct_type(name) {
-                t.ptr_type(AddressSpace::default()).as_basic_type_enum()
-            } else {
-                let exception = context.opaque_struct_type("Exception");
-                let int32 = context.i32_type().into();
-                let int64 = context.i64_type().into();
-                let str_ty = module.get_struct_type("str").unwrap().as_basic_type_enum();
-                let fields = [int32, str_ty, int32, int32, str_ty, str_ty, int64, int64, int64];
-                exception.set_body(&fields, false);
-                exception.ptr_type(AddressSpace::default()).as_basic_type_enum()
-            }
-        }),
+        (primitives.exception, pexn_model.get_type(generator, context).into()),
     ]
     .iter()
     .copied()
@@ -858,9 +809,10 @@ pub fn gen_func_impl<
     builder.position_at_end(init_bb);
     let body_bb = context.append_basic_block(fn_val, "body");
 
-    // Store non-vararg argument values into local variables
     let mut var_assignment = HashMap::new();
     let offset = u32::from(has_sret);
+
+    // Store non-vararg argument values into local variables
     for (n, arg) in args.iter().enumerate().filter(|(_, arg)| !arg.is_vararg) {
         let param = fn_val.get_nth_param((n as u32) + offset).unwrap();
         let local_type = get_llvm_type(
@@ -1124,106 +1076,3 @@ fn gen_in_range_check<'ctx>(
 fn get_va_count_arg_name(arg_name: StrRef) -> StrRef {
     format!("__{}_va_count", &arg_name).into()
 }
-
-/// Returns the alignment of the type.
-///
-/// This is necessary as `get_alignment` is not implemented as part of [`BasicType`].
-pub fn get_type_alignment<'ctx>(ty: impl Into<BasicTypeEnum<'ctx>>) -> IntValue<'ctx> {
-    match ty.into() {
-        BasicTypeEnum::ArrayType(ty) => ty.get_alignment(),
-        BasicTypeEnum::FloatType(ty) => ty.get_alignment(),
-        BasicTypeEnum::IntType(ty) => ty.get_alignment(),
-        BasicTypeEnum::PointerType(ty) => ty.get_alignment(),
-        BasicTypeEnum::StructType(ty) => ty.get_alignment(),
-        BasicTypeEnum::VectorType(ty) => ty.get_alignment(),
-    }
-}
-
-/// Inserts an `alloca` instruction with allocation `size` given in bytes and the alignment of the
-/// given type.
-///
-/// The returned [`PointerValue`] will have a type of `i8*`, a size of at least `size`, and will be
-/// aligned with the alignment of `align_ty`.
-pub fn type_aligned_alloca<'ctx, G: CodeGenerator + ?Sized>(
-    generator: &mut G,
-    ctx: &mut CodeGenContext<'ctx, '_>,
-    align_ty: impl Into<BasicTypeEnum<'ctx>>,
-    size: IntValue<'ctx>,
-    name: Option<&str>,
-) -> PointerValue<'ctx> {
-    /// Round `val` up to its modulo `power_of_two`.
-    fn round_up<'ctx>(
-        ctx: &CodeGenContext<'ctx, '_>,
-        val: IntValue<'ctx>,
-        power_of_two: IntValue<'ctx>,
-    ) -> IntValue<'ctx> {
-        debug_assert_eq!(
-            val.get_type().get_bit_width(),
-            power_of_two.get_type().get_bit_width(),
-            "`val` ({}) and `power_of_two` ({}) must be the same type",
-            val.get_type(),
-            power_of_two.get_type(),
-        );
-
-        let llvm_val_t = val.get_type();
-
-        let max_rem =
-            ctx.builder.build_int_sub(power_of_two, llvm_val_t.const_int(1, false), "").unwrap();
-        ctx.builder
-            .build_and(
-                ctx.builder.build_int_add(val, max_rem, "").unwrap(),
-                ctx.builder.build_not(max_rem, "").unwrap(),
-                "",
-            )
-            .unwrap()
-    }
-
-    let llvm_i8 = ctx.ctx.i8_type();
-    let llvm_pi8 = llvm_i8.ptr_type(AddressSpace::default());
-    let llvm_usize = generator.get_size_type(ctx.ctx);
-    let align_ty = align_ty.into();
-
-    let size = ctx.builder.build_int_truncate_or_bit_cast(size, llvm_usize, "").unwrap();
-
-    debug_assert_eq!(
-        size.get_type().get_bit_width(),
-        llvm_usize.get_bit_width(),
-        "Expected size_t ({}) for parameter `size` of `aligned_alloca`, got {}",
-        llvm_usize,
-        size.get_type(),
-    );
-
-    let alignment = get_type_alignment(align_ty);
-    let alignment = ctx.builder.build_int_truncate_or_bit_cast(alignment, llvm_usize, "").unwrap();
-
-    if ctx.registry.llvm_options.opt_level == OptimizationLevel::None {
-        let alignment_bitcount = llvm_intrinsics::call_int_ctpop(ctx, alignment, None);
-
-        ctx.make_assert(
-            generator,
-            ctx.builder
-                .build_int_compare(
-                    IntPredicate::EQ,
-                    alignment_bitcount,
-                    alignment_bitcount.get_type().const_int(1, false),
-                    "",
-                )
-                .unwrap(),
-            "0:AssertionError",
-            "Expected power-of-two alignment for aligned_alloca, got {0}",
-            [Some(alignment), None, None],
-            ctx.current_loc,
-        );
-    }
-
-    let buffer_size = round_up(ctx, size, alignment);
-    let aligned_slices = ctx.builder.build_int_unsigned_div(buffer_size, alignment, "").unwrap();
-
-    // Just to be absolutely sure, alloca in [i8 x alignment] slices
-    let buffer = ctx.builder.build_array_alloca(align_ty, aligned_slices, "").unwrap();
-
-    ctx.builder
-        .build_bit_cast(buffer, llvm_pi8, name.unwrap_or_default())
-        .map(BasicValueEnum::into_pointer_value)
-        .unwrap()
-}

nac3core/src/codegen/model/any.rs

@@ -0,0 +1,40 @@
+use inkwell::{
+    context::Context,
+    types::{BasicType, BasicTypeEnum},
+    values::BasicValueEnum,
+};
+
+use crate::codegen::CodeGenerator;
+
+use super::*;
+
+#[derive(Debug, Clone, Copy)]
+pub struct AnyModel<'ctx>(pub BasicTypeEnum<'ctx>);
+pub type Anything<'ctx> = Instance<'ctx, AnyModel<'ctx>>;
+
+impl<'ctx> Model<'ctx> for AnyModel<'ctx> {
+    type Value = BasicValueEnum<'ctx>;
+    type Type = BasicTypeEnum<'ctx>;
+
+    fn get_type<G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &G,
+        _ctx: &'ctx Context,
+    ) -> Self::Type {
+        self.0
+    }
+
+    fn check_type<T: BasicType<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &mut G,
+        _ctx: &'ctx Context,
+        ty: T,
+    ) -> Result<(), ModelError> {
+        let ty = ty.as_basic_type_enum();
+        if ty == self.0 {
+            Ok(())
+        } else {
+            Err(ModelError(format!("Expecting {}, but got {}", self.0, ty)))
+        }
+    }
+}

nac3core/src/codegen/model/core.rs

@@ -0,0 +1,121 @@
+use std::fmt;
+
+use inkwell::{context::Context, types::*, values::*};
+
+use super::*;
+use crate::codegen::{CodeGenContext, CodeGenerator};
+
+#[derive(Debug, Clone)]
+pub struct ModelError(pub String);
+
+impl ModelError {
+    pub(super) fn under_context(mut self, context: &str) -> Self {
+        self.0.push_str(" ... in ");
+        self.0.push_str(context);
+        self
+    }
+}
+
+pub trait Model<'ctx>: fmt::Debug + Clone + Copy {
+    type Value: BasicValue<'ctx> + TryFrom<BasicValueEnum<'ctx>>;
+    type Type: BasicType<'ctx>;
+
+    /// Return the [`BasicType`] of this model.
+    fn get_type<G: CodeGenerator + ?Sized>(&self, generator: &G, ctx: &'ctx Context) -> Self::Type;
+
+    /// Check if a [`BasicType`] is the same type of this model.
+    fn check_type<T: BasicType<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        ty: T,
+    ) -> Result<(), ModelError>;
+
+    /// Create an instance from a value with [`Instance::model`] being this model.
+    ///
+    /// Caller must make sure the type of `value` and the type of this `model` are equivalent.
+    fn believe_value(&self, value: Self::Value) -> Instance<'ctx, Self> {
+        Instance { model: *self, value }
+    }
+
+    /// Check if a [`BasicValue`]'s type is equivalent to the type of this model.
+    /// Wrap it into an [`Instance`] if it is.
+    fn check_value<V: BasicValue<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        value: V,
+    ) -> Result<Instance<'ctx, Self>, ModelError> {
+        let value = value.as_basic_value_enum();
+        self.check_type(generator, ctx, value.get_type())
+            .map_err(|err| err.under_context(format!("the value {value:?}").as_str()))?;
+
+        let Ok(value) = Self::Value::try_from(value) else {
+            unreachable!("check_type() has bad implementation")
+        };
+        Ok(self.believe_value(value))
+    }
+
+    // Allocate a value on the stack and return its pointer.
+    fn alloca<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        name: &str,
+    ) -> Ptr<'ctx, Self> {
+        let pmodel = PtrModel(*self);
+        let p = ctx.builder.build_alloca(self.get_type(generator, ctx.ctx), name).unwrap();
+        pmodel.believe_value(p)
+    }
+
+    // Allocate an array on the stack and return its pointer.
+    fn array_alloca<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        len: IntValue<'ctx>,
+        name: &str,
+    ) -> Ptr<'ctx, Self> {
+        let pmodel = PtrModel(*self);
+        let p =
+            ctx.builder.build_array_alloca(self.get_type(generator, ctx.ctx), len, name).unwrap();
+        pmodel.believe_value(p)
+    }
+
+    fn var_alloca<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        name: Option<&str>,
+    ) -> Result<Ptr<'ctx, Self>, String> {
+        let pmodel = PtrModel(*self);
+        let ty = self.get_type(generator, ctx.ctx).as_basic_type_enum();
+        let p = generator.gen_var_alloc(ctx, ty, name)?;
+        Ok(pmodel.believe_value(p))
+    }
+
+    fn array_var_alloca<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        len: IntValue<'ctx>,
+        name: Option<&'ctx str>,
+    ) -> Result<Ptr<'ctx, Self>, String> {
+        // TODO: Remove ArraySliceValue
+        let pmodel = PtrModel(*self);
+        let ty = self.get_type(generator, ctx.ctx).as_basic_type_enum();
+        let p = generator.gen_array_var_alloc(ctx, ty, len, name)?;
+        Ok(pmodel.believe_value(PointerValue::from(p)))
+    }
+}
+
+#[derive(Debug, Clone, Copy)]
+pub struct Instance<'ctx, M: Model<'ctx>> {
+    /// The model of this instance.
+    pub model: M,
+    /// The value of this instance.
+    ///
+    /// Caller must make sure the type of `value` and the type of this `model` are equivalent,
+    /// down to having the same [`IntType::get_bit_width`] in case of [`IntType`] for example.
+    pub value: M::Value,
+}

nac3core/src/codegen/model/int.rs

@@ -0,0 +1,276 @@
+use std::fmt;
+
+use inkwell::{context::Context, types::IntType, values::IntValue, IntPredicate};
+
+use crate::codegen::{CodeGenContext, CodeGenerator};
+
+use super::*;
+
+pub trait IntKind<'ctx>: fmt::Debug + Clone + Copy {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &G,
+        ctx: &'ctx Context,
+    ) -> IntType<'ctx>;
+}
+
+#[derive(Debug, Clone, Copy, Default)]
+pub struct Bool;
+#[derive(Debug, Clone, Copy, Default)]
+pub struct Byte;
+#[derive(Debug, Clone, Copy, Default)]
+pub struct Int32;
+#[derive(Debug, Clone, Copy, Default)]
+pub struct Int64;
+#[derive(Debug, Clone, Copy, Default)]
+pub struct SizeT;
+
+impl<'ctx> IntKind<'ctx> for Bool {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &G,
+        ctx: &'ctx Context,
+    ) -> IntType<'ctx> {
+        ctx.bool_type()
+    }
+}
+
+impl<'ctx> IntKind<'ctx> for Byte {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &G,
+        ctx: &'ctx Context,
+    ) -> IntType<'ctx> {
+        ctx.i8_type()
+    }
+}
+
+impl<'ctx> IntKind<'ctx> for Int32 {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &G,
+        ctx: &'ctx Context,
+    ) -> IntType<'ctx> {
+        ctx.i32_type()
+    }
+}
+
+impl<'ctx> IntKind<'ctx> for Int64 {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &G,
+        ctx: &'ctx Context,
+    ) -> IntType<'ctx> {
+        ctx.i64_type()
+    }
+}
+
+impl<'ctx> IntKind<'ctx> for SizeT {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &G,
+        ctx: &'ctx Context,
+    ) -> IntType<'ctx> {
+        generator.get_size_type(ctx)
+    }
+}
+
+#[derive(Debug, Clone, Copy)]
+pub struct AnyInt<'ctx>(pub IntType<'ctx>);
+
+impl<'ctx> IntKind<'ctx> for AnyInt<'ctx> {
+    fn get_int_type<G: CodeGenerator + ?Sized>(
+        &self,
+        _generator: &G,
+        _ctx: &'ctx Context,
+    ) -> IntType<'ctx> {
+        self.0
+    }
+}
+
+#[derive(Debug, Clone, Copy, Default)]
+pub struct IntModel<N>(pub N);
+pub type Int<'ctx, N> = Instance<'ctx, IntModel<N>>;
+
+impl<'ctx, N: IntKind<'ctx>> Model<'ctx> for IntModel<N> {
+    type Value = IntValue<'ctx>;
+    type Type = IntType<'ctx>;
+
+    #[must_use]
+    fn get_type<G: CodeGenerator + ?Sized>(&self, generator: &G, ctx: &'ctx Context) -> Self::Type {
+        self.0.get_int_type(generator, ctx)
+    }
+
+    fn check_type<T: inkwell::types::BasicType<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        ty: T,
+    ) -> Result<(), ModelError> {
+        let ty = ty.as_basic_type_enum();
+        let Ok(ty) = IntType::try_from(ty) else {
+            return Err(ModelError(format!("Expecting IntType, but got {ty:?}")));
+        };
+
+        let exp_ty = self.0.get_int_type(generator, ctx);
+        if ty.get_bit_width() != exp_ty.get_bit_width() {
+            return Err(ModelError(format!(
+                "Expecting IntType to have {} bit(s), but got {} bit(s)",
+                exp_ty.get_bit_width(),
+                ty.get_bit_width()
+            )));
+        }
+
+        Ok(())
+    }
+}
+
+impl<'ctx, N: IntKind<'ctx>> IntModel<N> {
+    pub fn constant<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        value: u64,
+    ) -> Int<'ctx, N> {
+        let value = self.get_type(generator, ctx).const_int(value, false);
+        self.believe_value(value)
+    }
+
+    pub fn const_0<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Int<'ctx, N> {
+        self.constant(generator, ctx, 0)
+    }
+
+    pub fn const_1<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Int<'ctx, N> {
+        self.constant(generator, ctx, 1)
+    }
+
+    pub fn const_all_1s<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Int<'ctx, N> {
+        let value = self.get_type(generator, ctx).const_all_ones();
+        self.believe_value(value)
+    }
+
+    pub fn s_extend_or_bit_cast<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        value: IntValue<'ctx>,
+        name: &str,
+    ) -> Int<'ctx, N> {
+        let value = ctx
+            .builder
+            .build_int_s_extend_or_bit_cast(value, self.get_type(generator, ctx.ctx), name)
+            .unwrap();
+        self.believe_value(value)
+    }
+
+    pub fn truncate<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        value: IntValue<'ctx>,
+        name: &str,
+    ) -> Int<'ctx, N> {
+        let value =
+            ctx.builder.build_int_truncate(value, self.get_type(generator, ctx.ctx), name).unwrap();
+        self.believe_value(value)
+    }
+}
+
+impl IntModel<Bool> {
+    #[must_use]
+    pub fn const_false<'ctx, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Int<'ctx, Bool> {
+        self.constant(generator, ctx, 0)
+    }
+
+    #[must_use]
+    pub fn const_true<'ctx, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Int<'ctx, Bool> {
+        self.constant(generator, ctx, 1)
+    }
+}
+
+impl<'ctx, N: IntKind<'ctx>> Int<'ctx, N> {
+    pub fn s_extend_or_bit_cast<NewN: IntKind<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        to_int_kind: NewN,
+        name: &str,
+    ) -> Int<'ctx, NewN> {
+        IntModel(to_int_kind).s_extend_or_bit_cast(generator, ctx, self.value, name)
+    }
+
+    pub fn truncate<NewN: IntKind<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        to_int_kind: NewN,
+        name: &str,
+    ) -> Int<'ctx, NewN> {
+        IntModel(to_int_kind).truncate(generator, ctx, self.value, name)
+    }
+
+    #[must_use]
+    pub fn add(
+        &self,
+        ctx: &CodeGenContext<'ctx, '_>,
+        other: Int<'ctx, N>,
+        name: &str,
+    ) -> Int<'ctx, N> {
+        let value = ctx.builder.build_int_add(self.value, other.value, name).unwrap();
+        self.model.believe_value(value)
+    }
+
+    #[must_use]
+    pub fn sub(
+        &self,
+        ctx: &CodeGenContext<'ctx, '_>,
+        other: Int<'ctx, N>,
+        name: &str,
+    ) -> Int<'ctx, N> {
+        let value = ctx.builder.build_int_sub(self.value, other.value, name).unwrap();
+        self.model.believe_value(value)
+    }
+
+    #[must_use]
+    pub fn mul(
+        &self,
+        ctx: &CodeGenContext<'ctx, '_>,
+        other: Int<'ctx, N>,
+        name: &str,
+    ) -> Int<'ctx, N> {
+        let value = ctx.builder.build_int_mul(self.value, other.value, name).unwrap();
+        self.model.believe_value(value)
+    }
+
+    pub fn compare(
+        &self,
+        ctx: &CodeGenContext<'ctx, '_>,
+        op: IntPredicate,
+        other: Int<'ctx, N>,
+        name: &str,
+    ) -> Int<'ctx, Bool> {
+        let bool_model = IntModel(Bool);
+        let value = ctx.builder.build_int_compare(op, self.value, other.value, name).unwrap();
+        bool_model.believe_value(value)
+    }
+}

nac3core/src/codegen/model/mod.rs

@@ -0,0 +1,12 @@
+mod any;
+mod core;
+mod int;
+mod ptr;
+mod structure;
+pub mod util;
+
+pub use any::*;
+pub use core::*;
+pub use int::*;
+pub use ptr::*;
+pub use structure::*;

nac3core/src/codegen/model/ptr.rs

@@ -0,0 +1,143 @@
+use inkwell::{
+    context::Context,
+    types::{BasicType, BasicTypeEnum, PointerType},
+    values::{IntValue, PointerValue},
+    AddressSpace,
+};
+
+use crate::codegen::{CodeGenContext, CodeGenerator};
+
+use super::*;
+
+#[derive(Debug, Clone, Copy, Default)]
+pub struct PtrModel<Element>(pub Element);
+pub type Ptr<'ctx, Element> = Instance<'ctx, PtrModel<Element>>;
+
+impl<'ctx, Element: Model<'ctx>> Model<'ctx> for PtrModel<Element> {
+    type Value = PointerValue<'ctx>;
+    type Type = PointerType<'ctx>;
+
+    fn get_type<G: CodeGenerator + ?Sized>(&self, generator: &G, ctx: &'ctx Context) -> Self::Type {
+        self.0.get_type(generator, ctx).ptr_type(AddressSpace::default())
+    }
+
+    fn check_type<T: BasicType<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        ty: T,
+    ) -> Result<(), ModelError> {
+        let ty = ty.as_basic_type_enum();
+        let Ok(ty) = PointerType::try_from(ty) else {
+            return Err(ModelError(format!("Expecting PointerType, but got {ty:?}")));
+        };
+
+        let elem_ty = ty.get_element_type();
+        let Ok(elem_ty) = BasicTypeEnum::try_from(elem_ty) else {
+            return Err(ModelError(format!(
+                "Expecting pointer element type to be a BasicTypeEnum, but got {elem_ty:?}"
+            )));
+        };
+
+        // TODO: inkwell `get_element_type()` will be deprecated.
+        // Remove the check for `get_element_type()` when the time comes.
+        self.0
+            .check_type(generator, ctx, elem_ty)
+            .map_err(|err| err.under_context("a PointerType"))?;
+
+        Ok(())
+    }
+}
+
+impl<'ctx, Element: Model<'ctx>> PtrModel<Element> {
+    /// Return a ***constant*** nullptr.
+    pub fn nullptr<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Ptr<'ctx, Element> {
+        let ptr = self.get_type(generator, ctx).const_null();
+        self.believe_value(ptr)
+    }
+
+    /// Cast a pointer into this model with [`inkwell::builder::Builder::build_pointer_cast`]
+    pub fn pointer_cast<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        ptr: PointerValue<'ctx>,
+        name: &str,
+    ) -> Ptr<'ctx, Element> {
+        let ptr =
+            ctx.builder.build_pointer_cast(ptr, self.get_type(generator, ctx.ctx), name).unwrap();
+        self.believe_value(ptr)
+    }
+}
+
+impl<'ctx, Element: Model<'ctx>> Ptr<'ctx, Element> {
+    /// Offset the pointer by [`inkwell::builder::Builder::build_in_bounds_gep`].
+    #[must_use]
+    pub fn offset<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        offset: IntValue<'ctx>,
+        name: &str,
+    ) -> Ptr<'ctx, Element> {
+        let new_ptr =
+            unsafe { ctx.builder.build_in_bounds_gep(self.value, &[offset], name).unwrap() };
+        self.model.check_value(generator, ctx.ctx, new_ptr).unwrap()
+    }
+
+    // Load the `i`-th element (0-based) on the array with [`inkwell::builder::Builder::build_in_bounds_gep`].
+    pub fn ix<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        i: IntValue<'ctx>,
+        name: &str,
+    ) -> Instance<'ctx, Element> {
+        self.offset(generator, ctx, i, name).load(generator, ctx, name)
+    }
+
+    /// Load the value with [`inkwell::builder::Builder::build_load`].
+    pub fn load<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        name: &str,
+    ) -> Instance<'ctx, Element> {
+        let value = ctx.builder.build_load(self.value, name).unwrap();
+        self.model.0.check_value(generator, ctx.ctx, value).unwrap() // If unwrap() panics, there is a logic error.
+    }
+
+    /// Store a value with [`inkwell::builder::Builder::build_store`].
+    pub fn store(&self, ctx: &CodeGenContext<'ctx, '_>, value: Instance<'ctx, Element>) {
+        ctx.builder.build_store(self.value, value.value).unwrap();
+    }
+
+    /// Return a casted pointer of element type `NewElement` with [`inkwell::builder::Builder::build_pointer_cast`].
+    pub fn transmute<NewElement: Model<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        new_model: NewElement,
+        name: &str,
+    ) -> Ptr<'ctx, NewElement> {
+        PtrModel(new_model).pointer_cast(generator, ctx, self.value, name)
+    }
+
+    /// Check if the pointer is null with [`inkwell::builder::Builder::build_is_null`].
+    pub fn is_null(&self, ctx: &CodeGenContext<'ctx, '_>, name: &str) -> Int<'ctx, Bool> {
+        let bool_model = IntModel(Bool);
+        let value = ctx.builder.build_is_null(self.value, name).unwrap();
+        bool_model.believe_value(value)
+    }
+
+    /// Check if the pointer is not null with [`inkwell::builder::Builder::build_is_not_null`].
+    pub fn is_not_null(&self, ctx: &CodeGenContext<'ctx, '_>, name: &str) -> Int<'ctx, Bool> {
+        let bool_model = IntModel(Bool);
+        let value = ctx.builder.build_is_not_null(self.value, name).unwrap();
+        bool_model.believe_value(value)
+    }
+}

nac3core/src/codegen/model/structure.rs

@@ -0,0 +1,203 @@
+use std::fmt;
+
+use inkwell::{
+    context::Context,
+    types::{BasicType, BasicTypeEnum, StructType},
+    values::StructValue,
+};
+
+use crate::codegen::{CodeGenContext, CodeGenerator};
+
+use super::*;
+
+#[derive(Debug, Clone, Copy)]
+pub struct GepField<M> {
+    pub gep_index: u64,
+    pub name: &'static str,
+    pub model: M,
+}
+
+pub trait FieldTraversal<'ctx> {
+    type Out<M>;
+
+    fn add<M: Model<'ctx>>(&mut self, name: &'static str, model: M) -> Self::Out<M>;
+
+    /// Like [`FieldTraversal::visit`] but [`Model`] is automatically inferred.
+    fn add_auto<M: Model<'ctx> + Default>(&mut self, name: &'static str) -> Self::Out<M> {
+        self.add(name, M::default())
+    }
+}
+
+pub struct GepFieldTraversal {
+    gep_index_counter: u64,
+}
+
+impl<'ctx> FieldTraversal<'ctx> for GepFieldTraversal {
+    type Out<M> = GepField<M>;
+
+    fn add<M: Model<'ctx>>(&mut self, name: &'static str, model: M) -> Self::Out<M> {
+        let gep_index = self.gep_index_counter;
+        self.gep_index_counter += 1;
+        Self::Out { gep_index, name, model }
+    }
+}
+
+struct TypeFieldTraversal<'ctx, 'a, G: CodeGenerator + ?Sized> {
+    generator: &'a G,
+    ctx: &'ctx Context,
+    field_types: Vec<BasicTypeEnum<'ctx>>,
+}
+
+impl<'ctx, 'a, G: CodeGenerator + ?Sized> FieldTraversal<'ctx> for TypeFieldTraversal<'ctx, 'a, G> {
+    type Out<M> = ();
+
+    fn add<M: Model<'ctx>>(&mut self, _name: &'static str, model: M) -> Self::Out<M> {
+        let t = model.get_type(self.generator, self.ctx).as_basic_type_enum();
+        self.field_types.push(t);
+    }
+}
+
+struct CheckTypeFieldTraversal<'ctx, 'a, G: CodeGenerator + ?Sized> {
+    generator: &'a mut G,
+    ctx: &'ctx Context,
+    index: u32,
+    scrutinee: StructType<'ctx>,
+    errors: Vec<ModelError>,
+}
+
+impl<'ctx, 'a, G: CodeGenerator + ?Sized> FieldTraversal<'ctx>
+    for CheckTypeFieldTraversal<'ctx, 'a, G>
+{
+    type Out<M> = ();
+
+    fn add<M: Model<'ctx>>(&mut self, name: &'static str, model: M) -> Self::Out<M> {
+        let i = self.index;
+        self.index += 1;
+
+        if let Some(t) = self.scrutinee.get_field_type_at_index(i) {
+            if let Err(err) = model.check_type(self.generator, self.ctx, t) {
+                self.errors.push(err.under_context(format!("field #{i} '{name}'").as_str()));
+            }
+        } // Otherwise, it will be caught
+    }
+}
+
+pub trait StructKind<'ctx>: fmt::Debug + Clone + Copy {
+    type Fields<F: FieldTraversal<'ctx>>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F>;
+
+    fn fields(&self) -> Self::Fields<GepFieldTraversal> {
+        self.traverse_fields(&mut GepFieldTraversal { gep_index_counter: 0 })
+    }
+
+    fn get_struct_type<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &G,
+        ctx: &'ctx Context,
+    ) -> StructType<'ctx> {
+        let mut traversal = TypeFieldTraversal { generator, ctx, field_types: Vec::new() };
+        self.traverse_fields(&mut traversal);
+
+        ctx.struct_type(&traversal.field_types, false)
+    }
+}
+
+#[derive(Debug, Clone, Copy, Default)]
+pub struct StructModel<S>(pub S);
+pub type Struct<'ctx, S> = Instance<'ctx, StructModel<S>>;
+
+impl<'ctx, S: StructKind<'ctx>> Model<'ctx> for StructModel<S> {
+    type Value = StructValue<'ctx>;
+    type Type = StructType<'ctx>;
+
+    fn get_type<G: CodeGenerator + ?Sized>(&self, generator: &G, ctx: &'ctx Context) -> Self::Type {
+        self.0.get_struct_type(generator, ctx)
+    }
+
+    fn check_type<T: BasicType<'ctx>, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        ty: T,
+    ) -> Result<(), ModelError> {
+        let ty = ty.as_basic_type_enum();
+        let Ok(ty) = StructType::try_from(ty) else {
+            return Err(ModelError(format!("Expecting StructType, but got {ty:?}")));
+        };
+
+        let mut traversal =
+            CheckTypeFieldTraversal { generator, ctx, index: 0, errors: Vec::new(), scrutinee: ty };
+        self.0.traverse_fields(&mut traversal);
+
+        let exp_num_fields = traversal.index;
+        let got_num_fields = u32::try_from(ty.get_field_types().len()).unwrap();
+        if exp_num_fields != got_num_fields {
+            return Err(ModelError(format!(
+                "Expecting StructType with {exp_num_fields} field(s), but got {got_num_fields}"
+            )));
+        }
+
+        if !traversal.errors.is_empty() {
+            return Err(traversal.errors[0].clone()); // TODO: Return other errors as well
+        }
+
+        Ok(())
+    }
+}
+
+impl<'ctx, S: StructKind<'ctx>> Ptr<'ctx, StructModel<S>> {
+    pub fn gep<M, GetField>(
+        &self,
+        ctx: &CodeGenContext<'ctx, '_>,
+        get_field: GetField,
+    ) -> Ptr<'ctx, M>
+    where
+        M: Model<'ctx>,
+        GetField: FnOnce(S::Fields<GepFieldTraversal>) -> GepField<M>,
+    {
+        let field = get_field(self.model.0 .0.fields());
+        let llvm_i32 = ctx.ctx.i32_type(); // i64 would segfault
+
+        let ptr = unsafe {
+            ctx.builder
+                .build_in_bounds_gep(
+                    self.value,
+                    &[llvm_i32.const_zero(), llvm_i32.const_int(field.gep_index, false)],
+                    field.name,
+                )
+                .unwrap()
+        };
+
+        let ptr_model = PtrModel(field.model);
+        ptr_model.believe_value(ptr)
+    }
+
+    /// Convenience function equivalent to `.gep(...).load(...)`.
+    pub fn get<M, GetField, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        get_field: GetField,
+        name: &str,
+    ) -> Instance<'ctx, M>
+    where
+        M: Model<'ctx>,
+        GetField: FnOnce(S::Fields<GepFieldTraversal>) -> GepField<M>,
+    {
+        self.gep(ctx, get_field).load(generator, ctx, name)
+    }
+
+    /// Convenience function equivalent to `.gep(...).store(...)`.
+    pub fn set<M, GetField>(
+        &self,
+        ctx: &CodeGenContext<'ctx, '_>,
+        get_field: GetField,
+        value: Instance<'ctx, M>,
+    ) where
+        M: Model<'ctx>,
+        GetField: FnOnce(S::Fields<GepFieldTraversal>) -> GepField<M>,
+    {
+        self.gep(ctx, get_field).store(ctx, value);
+    }
+}

nac3core/src/codegen/model/util.rs

@@ -0,0 +1,91 @@
+use inkwell::{types::BasicType, values::IntValue};
+
+/// `llvm.memcpy` but under the [`Model`] abstraction
+use crate::codegen::{
+    llvm_intrinsics::call_memcpy_generic,
+    stmt::{gen_for_callback_incrementing, BreakContinueHooks},
+    CodeGenContext, CodeGenerator,
+};
+
+use super::*;
+
+/// Convenience function.
+///
+/// Like [`call_memcpy_generic`] but with model abstractions and `is_volatile` set to `false`.
+pub fn call_memcpy_model<'ctx, Item: Model<'ctx> + Default, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &CodeGenContext<'ctx, '_>,
+    dst_array: Ptr<'ctx, Item>,
+    src_array: Ptr<'ctx, Item>,
+    num_items: IntValue<'ctx>,
+) {
+    let itemsize = Item::default().get_type(generator, ctx.ctx).size_of().unwrap();
+    let totalsize = ctx.builder.build_int_mul(itemsize, num_items, "totalsize").unwrap(); // TODO: Int types may not match.
+    let is_volatile = ctx.ctx.bool_type().const_zero();
+    call_memcpy_generic(ctx, dst_array.value, src_array.value, totalsize, is_volatile);
+}
+
+/// Like [`gen_for_callback_incrementing`] with [`Model`] abstractions.
+/// The [`IntKind`] is automatically inferred.
+pub fn gen_for_model_auto<'ctx, 'a, G, F, I>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, 'a>,
+    start: Int<'ctx, I>,
+    stop: Int<'ctx, I>,
+    step: Int<'ctx, I>,
+    body: F,
+) -> Result<(), String>
+where
+    G: CodeGenerator + ?Sized,
+    F: FnOnce(
+        &mut G,
+        &mut CodeGenContext<'ctx, 'a>,
+        BreakContinueHooks<'ctx>,
+        Int<'ctx, I>,
+    ) -> Result<(), String>,
+    I: IntKind<'ctx> + Default,
+{
+    let int_model = IntModel(I::default());
+
+    gen_for_callback_incrementing(
+        generator,
+        ctx,
+        None,
+        start.value,
+        (stop.value, false),
+        |g, ctx, hooks, i| {
+            let i = int_model.believe_value(i);
+            body(g, ctx, hooks, i)
+        },
+        step.value,
+    )
+}
+
+/// Like [`gen_if_callback`] with [`Model`] abstractions and without the `else` block.
+pub fn gen_if_model<'ctx, 'a, G, ThenFn>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, 'a>,
+    cond: Int<'ctx, Bool>,
+    then: ThenFn,
+) -> Result<(), String>
+where
+    G: CodeGenerator + ?Sized,
+    ThenFn: FnOnce(&mut G, &mut CodeGenContext<'ctx, 'a>) -> Result<(), String>,
+{
+    let current_bb = ctx.builder.get_insert_block().unwrap();
+    let then_bb = ctx.ctx.insert_basic_block_after(current_bb, "if.then");
+    let end_bb = ctx.ctx.insert_basic_block_after(then_bb, "if.end");
+
+    // Inserting into `current_bb`.
+    ctx.builder.build_conditional_branch(cond.value, then_bb, end_bb).unwrap();
+
+    // Inserting into `then_bb`
+    ctx.builder.position_at_end(then_bb);
+    then(generator, ctx)?;
+    ctx.builder.build_unconditional_branch(end_bb).unwrap();
+
+    // Reposition to `end_bb` for continuation.
+    ctx.builder.position_at_end(end_bb);
+
+    Ok(())
+}

nac3core/src/codegen/numpy_new.rs

@@ -0,0 +1,480 @@
+// TODO: Replace numpy.rs
+
+use inkwell::values::{BasicValue, BasicValueEnum};
+use nac3parser::ast::StrRef;
+
+use crate::{
+    codegen::structure::{
+        ndarray::{
+            scalar::split_scalar_or_ndarray, shape_util::parse_numpy_int_sequence, NDArrayObject,
+        },
+        tuple::TupleObject,
+    },
+    symbol_resolver::ValueEnum,
+    toplevel::{
+        numpy::{extract_ndims, unpack_ndarray_var_tys},
+        DefinitionId,
+    },
+    typecheck::typedef::{FunSignature, Type},
+};
+
+use super::{
+    irrt::call_nac3_ndarray_util_assert_shape_no_negative, model::*, CodeGenContext, CodeGenerator,
+};
+
+/// Get the zero value in `np.zeros()` of a `dtype`.
+fn ndarray_zero_value<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    dtype: Type,
+) -> BasicValueEnum<'ctx> {
+    if [ctx.primitives.int32, ctx.primitives.uint32]
+        .iter()
+        .any(|ty| ctx.unifier.unioned(dtype, *ty))
+    {
+        ctx.ctx.i32_type().const_zero().into()
+    } else if [ctx.primitives.int64, ctx.primitives.uint64]
+        .iter()
+        .any(|ty| ctx.unifier.unioned(dtype, *ty))
+    {
+        ctx.ctx.i64_type().const_zero().into()
+    } else if ctx.unifier.unioned(dtype, ctx.primitives.float) {
+        ctx.ctx.f64_type().const_zero().into()
+    } else if ctx.unifier.unioned(dtype, ctx.primitives.bool) {
+        ctx.ctx.bool_type().const_zero().into()
+    } else if ctx.unifier.unioned(dtype, ctx.primitives.str) {
+        ctx.gen_string(generator, "").value.into()
+    } else {
+        unreachable!()
+    }
+}
+
+/// Get the one value in `np.ones()` of a `dtype`.
+fn ndarray_one_value<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    elem_ty: Type,
+) -> BasicValueEnum<'ctx> {
+    if [ctx.primitives.int32, ctx.primitives.uint32]
+        .iter()
+        .any(|ty| ctx.unifier.unioned(elem_ty, *ty))
+    {
+        let is_signed = ctx.unifier.unioned(elem_ty, ctx.primitives.int32);
+        ctx.ctx.i32_type().const_int(1, is_signed).into()
+    } else if [ctx.primitives.int64, ctx.primitives.uint64]
+        .iter()
+        .any(|ty| ctx.unifier.unioned(elem_ty, *ty))
+    {
+        let is_signed = ctx.unifier.unioned(elem_ty, ctx.primitives.int64);
+        ctx.ctx.i64_type().const_int(1, is_signed).into()
+    } else if ctx.unifier.unioned(elem_ty, ctx.primitives.float) {
+        ctx.ctx.f64_type().const_float(1.0).into()
+    } else if ctx.unifier.unioned(elem_ty, ctx.primitives.bool) {
+        ctx.ctx.bool_type().const_int(1, false).into()
+    } else if ctx.unifier.unioned(elem_ty, ctx.primitives.str) {
+        ctx.gen_string(generator, "1").value.into()
+    } else {
+        unreachable!()
+    }
+}
+
+/// Helper function to create an ndarray with uninitialized values.
+///
+/// * `ndarray_ty` - The [`Type`] of the ndarray
+/// * `shape` - The user input shape argument
+/// * `shape_ty` - The [`Type`] of the shape argument
+///
+/// This function does data validation the `shape` input.
+fn create_empty_ndarray<'ctx, G>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    ndarray_ty: Type,
+    shape: BasicValueEnum<'ctx>,
+    shape_ty: Type,
+) -> NDArrayObject<'ctx>
+where
+    G: CodeGenerator + ?Sized,
+{
+    let (_, shape) = parse_numpy_int_sequence(generator, ctx, shape, shape_ty);
+
+    let ndarray =
+        NDArrayObject::alloca_uninitialized_of_type(generator, ctx, ndarray_ty, "ndarray");
+
+    // Validate `shape`
+    let ndims = ndarray.get_ndims(generator, ctx.ctx);
+    call_nac3_ndarray_util_assert_shape_no_negative(generator, ctx, ndims, shape);
+
+    // Setup `ndarray` with `shape`
+    ndarray.copy_shape_from_array(generator, ctx, shape);
+    ndarray.create_data(generator, ctx); // `shape` has to be set
+
+    ndarray
+}
+
+/// Generates LLVM IR for `np.empty`.
+pub fn gen_ndarray_empty<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse arguments
+    let shape_ty = fun.0.args[0].ty;
+    let shape = args[0].1.clone().to_basic_value_enum(ctx, generator, shape_ty)?;
+
+    // Implementation
+    let ndarray_ty = fun.0.ret;
+    let ndarray = create_empty_ndarray(generator, ctx, ndarray_ty, shape, shape_ty);
+
+    Ok(ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.zero`.
+pub fn gen_ndarray_zeros<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse arguments
+    let shape_ty = fun.0.args[0].ty;
+    let shape = args[0].1.clone().to_basic_value_enum(ctx, generator, shape_ty)?;
+
+    // Implementation
+    let ndarray_ty = fun.0.ret;
+    let ndarray = create_empty_ndarray(generator, ctx, ndarray_ty, shape, shape_ty);
+
+    let fill_value = ndarray_zero_value(generator, ctx, ndarray.dtype);
+    ndarray.fill(generator, ctx, fill_value);
+
+    Ok(ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.ones`.
+pub fn gen_ndarray_ones<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse arguments
+    let shape_ty = fun.0.args[0].ty;
+    let shape = args[0].1.clone().to_basic_value_enum(ctx, generator, shape_ty)?;
+
+    // Implementation
+    let ndarray_ty = fun.0.ret;
+    let ndarray = create_empty_ndarray(generator, ctx, ndarray_ty, shape, shape_ty);
+
+    let fill_value = ndarray_zero_value(generator, ctx, ndarray.dtype);
+    ndarray.fill(generator, ctx, fill_value);
+
+    Ok(ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.full`.
+pub fn gen_ndarray_full<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 2);
+
+    // Parse argument #1 shape
+    let shape_ty = fun.0.args[0].ty;
+    let shape = args[0].1.clone().to_basic_value_enum(ctx, generator, shape_ty)?;
+
+    // Parse argument #2 fill_value
+    let fill_value_ty = fun.0.args[1].ty;
+    let fill_value = args[1].1.clone().to_basic_value_enum(ctx, generator, fill_value_ty)?;
+
+    // Implementation
+    let ndarray_ty = fun.0.ret;
+    let ndarray = create_empty_ndarray(generator, ctx, ndarray_ty, shape, shape_ty);
+
+    ndarray.fill(generator, ctx, fill_value);
+
+    Ok(ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.broadcast_to`.
+pub fn gen_ndarray_broadcast_to<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 2);
+
+    // Parse argument #1 input
+    let input_ty = fun.0.args[0].ty;
+    let input = args[0].1.clone().to_basic_value_enum(ctx, generator, input_ty)?;
+
+    // Parse argument #2 shape
+    let shape_ty = fun.0.args[1].ty;
+    let shape = args[1].1.clone().to_basic_value_enum(ctx, generator, shape_ty)?;
+
+    // Define models
+    let sizet_model = IntModel(SizeT);
+
+    // Extract broadcast_ndims, this is the only way to get the
+    // ndims of the ndarray result statically.
+    let (_, broadcast_ndims_ty) = unpack_ndarray_var_tys(&mut ctx.unifier, fun.0.ret);
+    let broadcast_ndims = extract_ndims(&ctx.unifier, broadcast_ndims_ty);
+
+    // Process `input`
+    let in_ndarray =
+        split_scalar_or_ndarray(generator, ctx, input, input_ty).as_ndarray(generator, ctx);
+
+    // Process `shape`
+    let (_, broadcast_shape) = parse_numpy_int_sequence(generator, ctx, shape, shape_ty);
+    // NOTE: shape.size should equal to `broadcasted_ndims`.
+    let broadcast_ndims_llvm = sizet_model.constant(generator, ctx.ctx, broadcast_ndims);
+    call_nac3_ndarray_util_assert_shape_no_negative(
+        generator,
+        ctx,
+        broadcast_ndims_llvm,
+        broadcast_shape,
+    );
+
+    // Create broadcast view
+    let broadcast_ndarray =
+        in_ndarray.broadcast_to(generator, ctx, broadcast_ndims, broadcast_shape);
+
+    Ok(broadcast_ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.reshape`.
+pub fn gen_ndarray_reshape<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 2);
+
+    // Parse argument #1 input
+    let input_ty = fun.0.args[0].ty;
+    let input = args[0].1.clone().to_basic_value_enum(ctx, generator, input_ty)?;
+
+    // Parse argument #2 shape
+    let shape_ty = fun.0.args[1].ty;
+    let shape = args[1].1.clone().to_basic_value_enum(ctx, generator, shape_ty)?;
+
+    // Extract reshaped_ndims
+    let (_, reshaped_ndims_ty) = unpack_ndarray_var_tys(&mut ctx.unifier, fun.0.ret);
+    let reshaped_ndims = extract_ndims(&ctx.unifier, reshaped_ndims_ty);
+
+    // Process `input`
+    let in_ndarray =
+        split_scalar_or_ndarray(generator, ctx, input, input_ty).as_ndarray(generator, ctx);
+
+    // Process the shape input from user and resolve negative indices.
+    // The resulting `new_shape`'s size should be equal to reshaped_ndims.
+    // This is ensured by the typechecker.
+    let (_, new_shape) = parse_numpy_int_sequence(generator, ctx, shape, shape_ty);
+    let reshaped_ndarray = in_ndarray.reshape_or_copy(generator, ctx, reshaped_ndims, new_shape);
+
+    Ok(reshaped_ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.arange`.
+pub fn gen_ndarray_arange<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse argument #1 len
+    let input_ty = fun.0.args[0].ty;
+    let input = args[0].1.clone().to_basic_value_enum(ctx, generator, input_ty)?.into_int_value();
+
+    // Define models
+    let sizet_model = IntModel(SizeT);
+
+    // Process input
+    let input = sizet_model.s_extend_or_bit_cast(generator, ctx, input, "input_dim");
+
+    // Allocate the resulting ndarray
+    let ndarray = NDArrayObject::alloca_uninitialized(
+        generator,
+        ctx,
+        ctx.primitives.float,
+        1, // ndims = 1
+        "arange_ndarray",
+    );
+
+    // `ndarray.shape[0] = input`
+    let zero = sizet_model.const_0(generator, ctx.ctx);
+    ndarray
+        .value
+        .get(generator, ctx, |f| f.shape, "shape")
+        .offset(generator, ctx, zero.value, "dim")
+        .store(ctx, input);
+
+    // Create data and set elements
+    ndarray.create_data(generator, ctx);
+    ndarray.foreach_pointer(generator, ctx, |_generator, ctx, _hooks, i, pelement| {
+        let val =
+            ctx.builder.build_unsigned_int_to_float(i.value, ctx.ctx.f64_type(), "val").unwrap();
+        ctx.builder.build_store(pelement, val).unwrap();
+        Ok(())
+    })?;
+
+    Ok(ndarray.value.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.size`.
+pub fn gen_ndarray_size<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    let ndarray_ty = fun.0.args[0].ty;
+    let ndarray = args[0].1.clone().to_basic_value_enum(ctx, generator, ndarray_ty)?;
+
+    let ndarray = NDArrayObject::from_value_and_type(generator, ctx, ndarray, ndarray_ty);
+
+    let size = ndarray.size(generator, ctx).truncate(generator, ctx, Int32, "size");
+    Ok(size.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.shape`.
+pub fn gen_ndarray_shape<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse argument #1 ndarray
+    let ndarray_ty = fun.0.args[0].ty;
+    let ndarray = args[0].1.clone().to_basic_value_enum(ctx, generator, ndarray_ty)?;
+
+    // Define models
+    let sizet_model = IntModel(SizeT);
+
+    // Process ndarray
+    let ndarray = NDArrayObject::from_value_and_type(generator, ctx, ndarray, ndarray_ty);
+
+    let mut items = Vec::with_capacity(ndarray.ndims as usize);
+
+    for i in 0..ndarray.ndims {
+        let i = sizet_model.constant(generator, ctx.ctx, i);
+        let dim =
+            ndarray.value.get(generator, ctx, |f| f.shape, "").ix(generator, ctx, i.value, "dim");
+        let dim = dim.truncate(generator, ctx, Int32, "dim"); // TODO: keep using SizeT
+
+        items.push((dim.value.as_basic_value_enum(), ctx.primitives.int32));
+    }
+
+    let shape = TupleObject::create(generator, ctx, items, "shape");
+    Ok(shape.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `<ndarray>.strides`.
+pub fn gen_ndarray_strides<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    // TODO: This function looks exactly like `gen_ndarray_shapes`, code duplication?
+
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse argument #1 ndarray
+    let ndarray_ty = fun.0.args[0].ty;
+    let ndarray = args[0].1.clone().to_basic_value_enum(ctx, generator, ndarray_ty)?;
+
+    // Define models
+    let sizet_model = IntModel(SizeT);
+
+    // Process ndarray
+    let ndarray = NDArrayObject::from_value_and_type(generator, ctx, ndarray, ndarray_ty);
+
+    let mut items = Vec::with_capacity(ndarray.ndims as usize);
+
+    for i in 0..ndarray.ndims {
+        let i = sizet_model.constant(generator, ctx.ctx, i);
+        let dim =
+            ndarray.value.get(generator, ctx, |f| f.strides, "").ix(generator, ctx, i.value, "dim");
+        let dim = dim.truncate(generator, ctx, Int32, "dim"); // TODO: keep using SizeT
+
+        items.push((dim.value.as_basic_value_enum(), ctx.primitives.int32));
+    }
+
+    let strides = TupleObject::create(generator, ctx, items, "strides");
+    Ok(strides.value.as_basic_value_enum())
+}
+
+/// Generates LLVM IR for `np.transpose`.
+pub fn gen_ndarray_transpose<'ctx>(
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    obj: &Option<(Type, ValueEnum<'ctx>)>,
+    fun: (&FunSignature, DefinitionId),
+    args: &[(Option<StrRef>, ValueEnum<'ctx>)],
+    generator: &mut dyn CodeGenerator,
+) -> Result<BasicValueEnum<'ctx>, String> {
+    // TODO: The implementation will be changed once default values start working again.
+    // Read the comment on this function in BuiltinBuilder.
+
+    // TODO: Change axes values to `SizeT`
+
+    assert!(obj.is_none());
+    assert_eq!(args.len(), 1);
+
+    // Parse argument #1 ndarray
+    let ndarray_ty = fun.0.args[0].ty;
+    let ndarray = args[0].1.clone().to_basic_value_enum(ctx, generator, ndarray_ty)?;
+
+    // Implementation
+    let ndarray = NDArrayObject::from_value_and_type(generator, ctx, ndarray, ndarray_ty);
+
+    let has_axes = args.len() >= 2;
+    let transposed_ndarray = if has_axes {
+        // Parse argument #2 axes
+        let in_axes_ty = fun.0.args[1].ty;
+        let in_axes = args[1].1.clone().to_basic_value_enum(ctx, generator, in_axes_ty)?;
+
+        let (_, axes) = parse_numpy_int_sequence(generator, ctx, in_axes, in_axes_ty);
+
+        ndarray.transpose(generator, ctx, Some(axes))
+    } else {
+        ndarray.transpose(generator, ctx, None)
+    };
+
+    Ok(transposed_ndarray.value.value.as_basic_value_enum())
+}

nac3core/src/codegen/stmt.rs

@@ -1,3 +1,27 @@
+use super::model::*;
+use super::structure::cslice::CSlice;
+use super::{
+    super::symbol_resolver::ValueEnum,
+    expr::destructure_range,
+    irrt::{handle_slice_indices, list_slice_assignment},
+    structure::exception::Exception,
+    CodeGenContext, CodeGenerator, Int32, IntModel, Ptr, StructModel,
+};
+use crate::codegen::structure::ndarray::indexing::util::gen_ndarray_subscript_ndindexes;
+use crate::codegen::structure::ndarray::scalar::split_scalar_or_ndarray;
+use crate::codegen::structure::ndarray::NDArrayObject;
+use crate::{
+    codegen::{
+        classes::{ArrayLikeIndexer, ArraySliceValue, ListValue, RangeValue},
+        expr::gen_binop_expr,
+        gen_in_range_check,
+    },
+    toplevel::{DefinitionId, TopLevelDef},
+    typecheck::{
+        magic_methods::Binop,
+        typedef::{iter_type_vars, FunSignature, Type, TypeEnum},
+    },
+};
 use inkwell::{
     attributes::{Attribute, AttributeLoc},
     basic_block::BasicBlock,
@@ -6,28 +30,10 @@ use inkwell::{
     IntPredicate,
 };
 use itertools::{izip, Itertools};
-
 use nac3parser::ast::{
     Constant, ExcepthandlerKind, Expr, ExprKind, Location, Stmt, StmtKind, StrRef,
 };
 
-use super::{
-    expr::{destructure_range, gen_binop_expr},
-    gen_in_range_check,
-    irrt::{handle_slice_indices, list_slice_assignment},
-    macros::codegen_unreachable,
-    values::{ArrayLikeIndexer, ArraySliceValue, ListValue, RangeValue},
-    CodeGenContext, CodeGenerator,
-};
-use crate::{
-    symbol_resolver::ValueEnum,
-    toplevel::{DefinitionId, TopLevelDef},
-    typecheck::{
-        magic_methods::Binop,
-        typedef::{iter_type_vars, FunSignature, Type, TypeEnum},
-    },
-};
-
 /// See [`CodeGenerator::gen_var_alloc`].
 pub fn gen_var<'ctx>(
     ctx: &mut CodeGenContext<'ctx, '_>,
@@ -121,7 +127,7 @@ pub fn gen_store_target<'ctx, G: CodeGenerator>(
                 return Ok(None);
             };
             let BasicValueEnum::PointerValue(ptr) = val else {
-                codegen_unreachable!(ctx);
+                unreachable!();
             };
             unsafe {
                 ctx.builder.build_in_bounds_gep(
@@ -135,7 +141,7 @@ pub fn gen_store_target<'ctx, G: CodeGenerator>(
             }
             .unwrap()
         }
-        _ => codegen_unreachable!(ctx),
+        _ => unreachable!(),
     }))
 }
 
@@ -176,14 +182,6 @@ pub fn gen_assign<'ctx, G: CodeGenerator>(
                 }
             }
             let val = value.to_basic_value_enum(ctx, generator, target.custom.unwrap())?;
-
-            // Perform i1 <-> i8 conversion as needed
-            let val = if ctx.unifier.unioned(target.custom.unwrap(), ctx.primitives.bool) {
-                generator.bool_to_i8(ctx, val.into_int_value()).into()
-            } else {
-                val
-            };
-
             ctx.builder.build_store(ptr, val).unwrap();
         }
     };
@@ -201,12 +199,12 @@ pub fn gen_assign_target_list<'ctx, G: CodeGenerator>(
     // Deconstruct the tuple `value`
     let BasicValueEnum::StructValue(tuple) = value.to_basic_value_enum(ctx, generator, value_ty)?
     else {
-        codegen_unreachable!(ctx)
+        unreachable!()
     };
 
     // NOTE: Currently, RHS's type is forced to be a Tuple by the type inferencer.
     let TypeEnum::TTuple { ty: tuple_tys, .. } = &*ctx.unifier.get_ty(value_ty) else {
-        codegen_unreachable!(ctx);
+        unreachable!();
     };
 
     assert_eq!(tuple.get_type().count_fields() as usize, tuple_tys.len());
@@ -266,7 +264,7 @@ pub fn gen_assign_target_list<'ctx, G: CodeGenerator>(
             // Now assign with that sub-tuple to the starred target.
             generator.gen_assign(ctx, target, ValueEnum::Dynamic(sub_tuple_val), sub_tuple_ty)?;
         } else {
-            codegen_unreachable!(ctx) // The typechecker ensures this
+            unreachable!() // The typechecker ensures this
         }
 
         // Handle assignment after the starred target
@@ -310,13 +308,11 @@ pub fn gen_setitem<'ctx, G: CodeGenerator>(
                 .unwrap()
                 .to_basic_value_enum(ctx, generator, target_ty)?
                 .into_pointer_value();
-            let target = ListValue::from_pointer_value(target, llvm_usize, None);
+            let target = ListValue::from_ptr_val(target, llvm_usize, None);
 
             if let ExprKind::Slice { .. } = &key.node {
                 // Handle assigning to a slice
-                let ExprKind::Slice { lower, upper, step } = &key.node else {
-                    codegen_unreachable!(ctx)
-                };
+                let ExprKind::Slice { lower, upper, step } = &key.node else { unreachable!() };
                 let Some((start, end, step)) = handle_slice_indices(
                     lower,
                     upper,
@@ -331,7 +327,7 @@ pub fn gen_setitem<'ctx, G: CodeGenerator>(
 
                 let value =
                     value.to_basic_value_enum(ctx, generator, value_ty)?.into_pointer_value();
-                let value = ListValue::from_pointer_value(value, llvm_usize, None);
+                let value = ListValue::from_ptr_val(value, llvm_usize, None);
 
                 let target_item_ty = ctx.get_llvm_type(generator, target_item_ty);
                 let Some(src_ind) = handle_slice_indices(
@@ -411,7 +407,43 @@ pub fn gen_setitem<'ctx, G: CodeGenerator>(
             if *obj_id == ctx.primitives.ndarray.obj_id(&ctx.unifier).unwrap() =>
         {
             // Handle NDArray item assignment
-            todo!("ndarray subscript assignment is not yet implemented");
+            // Process target
+            let target = generator
+                .gen_expr(ctx, target)?
+                .unwrap()
+                .to_basic_value_enum(ctx, generator, target_ty)?;
+            let target = NDArrayObject::from_value_and_type(generator, ctx, target, target_ty);
+
+            // Process key
+            let key = gen_ndarray_subscript_ndindexes(generator, ctx, key)?;
+
+            // Process value
+            let value = value.to_basic_value_enum(ctx, generator, value_ty)?;
+
+            /*
+                Reference code:
+                ```python
+                target = target[key]
+                value = np.asarray(value)
+
+                shape = np.broadcast_shape((target, value))
+
+                target = np.broadcast_to(target, shape)
+                value = np.broadcast_to(value, shape)
+
+                ...and finally copy 1-1 from value to target.
+                ```
+            */
+            let target = target.index(generator, ctx, &key, "assign_target_ndarray");
+            let value =
+                split_scalar_or_ndarray(generator, ctx, value, value_ty).as_ndarray(generator, ctx);
+
+            let broadcast_result = NDArrayObject::broadcast_all(generator, ctx, &[target, value]);
+
+            let target = broadcast_result.ndarrays[0];
+            let value = broadcast_result.ndarrays[1];
+
+            target.copy_data_from(generator, ctx, value);
         }
         _ => {
             panic!("encountered unknown target type: {}", ctx.unifier.stringify(target_ty));
@@ -426,9 +458,7 @@ pub fn gen_for<G: CodeGenerator>(
     ctx: &mut CodeGenContext<'_, '_>,
     stmt: &Stmt<Option<Type>>,
 ) -> Result<(), String> {
-    let StmtKind::For { iter, target, body, orelse, .. } = &stmt.node else {
-        codegen_unreachable!(ctx)
-    };
+    let StmtKind::For { iter, target, body, orelse, .. } = &stmt.node else { unreachable!() };
 
     // var_assignment static values may be changed in another branch
     // if so, remove the static value as it may not be correct in this branch
@@ -463,15 +493,14 @@ pub fn gen_for<G: CodeGenerator>(
         TypeEnum::TObj { obj_id, .. }
             if *obj_id == ctx.primitives.range.obj_id(&ctx.unifier).unwrap() =>
         {
-            let iter_val =
-                RangeValue::from_pointer_value(iter_val.into_pointer_value(), Some("range"));
+            let iter_val = RangeValue::from_ptr_val(iter_val.into_pointer_value(), Some("range"));
             // Internal variable for loop; Cannot be assigned
             let i = generator.gen_var_alloc(ctx, int32.into(), Some("for.i.addr"))?;
             // Variable declared in "target" expression of the loop; Can be reassigned *or* shadowed
             let Some(target_i) =
                 generator.gen_store_target(ctx, target, Some("for.target.addr"))?
             else {
-                codegen_unreachable!(ctx)
+                unreachable!()
             };
             let (start, stop, step) = destructure_range(ctx, iter_val);
 
@@ -632,9 +661,9 @@ pub struct BreakContinueHooks<'ctx> {
 /// ```
 ///
 /// * `init` - A lambda containing IR statements declaring and initializing loop variables. The
-///   return value is a [Clone] value which will be passed to the other lambdas.
+/// return value is a [Clone] value which will be passed to the other lambdas.
 /// * `cond` - A lambda containing IR statements checking whether the loop should continue
-///   executing. The result value must be an `i1` indicating if the loop should continue.
+/// executing. The result value must be an `i1` indicating if the loop should continue.
 /// * `body` - A lambda containing IR statements within the loop body.
 /// * `update` - A lambda containing IR statements updating loop variables.
 pub fn gen_for_callback<'ctx, 'a, G, I, InitFn, CondFn, BodyFn, UpdateFn>(
@@ -717,9 +746,9 @@ where
 /// ```
 ///
 /// * `init_val` - The initial value of the loop variable. The type of this value will also be used
-///   as the type of the loop variable.
+/// as the type of the loop variable.
 /// * `max_val` - A tuple containing the maximum value of the loop variable, and whether the maximum
-///   value should be treated as inclusive (as opposed to exclusive).
+/// value should be treated as inclusive (as opposed to exclusive).
 /// * `body` - A lambda containing IR statements within the loop body.
 /// * `incr_val` - The value to increment the loop variable on each iteration.
 pub fn gen_for_callback_incrementing<'ctx, 'a, G, BodyFn>(
@@ -790,12 +819,12 @@ where
 ///
 /// - `is_unsigned`: Whether to treat the values of the `range` as unsigned.
 /// - `start_fn`: A lambda of IR statements that retrieves the `start` value of the `range`-like
-///   iterable.
+/// iterable.
 /// - `stop_fn`: A lambda of IR statements that retrieves the `stop` value of the `range`-like
-///   iterable. This value will be extended to the size of `start`.
+/// iterable. This value will be extended to the size of `start`.
 /// - `stop_inclusive`: Whether the stop value should be treated as inclusive.
 /// - `step_fn`: A lambda of IR statements that retrieves the `step` value of the  `range`-like
-///   iterable. This value will be extended to the size of `start`.
+/// iterable. This value will be extended to the size of `start`.
 /// - `body_fn`: A lambda of IR statements within the loop body.
 #[allow(clippy::too_many_arguments)]
 pub fn gen_for_range_callback<'ctx, 'a, G, StartFn, StopFn, StepFn, BodyFn>(
@@ -816,7 +845,7 @@ where
     BodyFn: FnOnce(
         &mut G,
         &mut CodeGenContext<'ctx, 'a>,
-        BreakContinueHooks<'ctx>,
+        BreakContinueHooks,
         IntValue<'ctx>,
     ) -> Result<(), String>,
 {
@@ -914,7 +943,7 @@ pub fn gen_while<G: CodeGenerator>(
     ctx: &mut CodeGenContext<'_, '_>,
     stmt: &Stmt<Option<Type>>,
 ) -> Result<(), String> {
-    let StmtKind::While { test, body, orelse, .. } = &stmt.node else { codegen_unreachable!(ctx) };
+    let StmtKind::While { test, body, orelse, .. } = &stmt.node else { unreachable!() };
 
     // var_assignment static values may be changed in another branch
     // if so, remove the static value as it may not be correct in this branch
@@ -944,7 +973,7 @@ pub fn gen_while<G: CodeGenerator>(
 
         return Ok(());
     };
-    let BasicValueEnum::IntValue(test) = test else { codegen_unreachable!(ctx) };
+    let BasicValueEnum::IntValue(test) = test else { unreachable!() };
 
     ctx.builder
         .build_conditional_branch(generator.bool_to_i1(ctx, test), body_bb, orelse_bb)
@@ -1092,7 +1121,7 @@ pub fn gen_if<G: CodeGenerator>(
     ctx: &mut CodeGenContext<'_, '_>,
     stmt: &Stmt<Option<Type>>,
 ) -> Result<(), String> {
-    let StmtKind::If { test, body, orelse, .. } = &stmt.node else { codegen_unreachable!(ctx) };
+    let StmtKind::If { test, body, orelse, .. } = &stmt.node else { unreachable!() };
 
     // var_assignment static values may be changed in another branch
     // if so, remove the static value as it may not be correct in this branch
@@ -1215,11 +1244,11 @@ pub fn exn_constructor<'ctx>(
     let zelf_id = if let TypeEnum::TObj { obj_id, .. } = &*ctx.unifier.get_ty(zelf_ty) {
         obj_id.0
     } else {
-        codegen_unreachable!(ctx)
+        unreachable!()
     };
     let defs = ctx.top_level.definitions.read();
     let def = defs[zelf_id].read();
-    let TopLevelDef::Class { name: zelf_name, .. } = &*def else { codegen_unreachable!(ctx) };
+    let TopLevelDef::Class { name: zelf_name, .. } = &*def else { unreachable!() };
     let exception_name = format!("{}:{}", ctx.resolver.get_exception_id(zelf_id), zelf_name);
     unsafe {
         let id_ptr = ctx.builder.build_in_bounds_gep(zelf, &[zero, zero], "exn.id").unwrap();
@@ -1276,47 +1305,36 @@ pub fn exn_constructor<'ctx>(
 pub fn gen_raise<'ctx, G: CodeGenerator + ?Sized>(
     generator: &mut G,
     ctx: &mut CodeGenContext<'ctx, '_>,
-    exception: Option<&BasicValueEnum<'ctx>>,
+    exception: Option<Ptr<'ctx, StructModel<Exception>>>,
     loc: Location,
 ) {
-    if let Some(exception) = exception {
-        unsafe {
-            let int32 = ctx.ctx.i32_type();
-            let zero = int32.const_zero();
-            let exception = exception.into_pointer_value();
-            let file_ptr = ctx
-                .builder
-                .build_in_bounds_gep(exception, &[zero, int32.const_int(1, false)], "file_ptr")
-                .unwrap();
-            let filename = ctx.gen_string(generator, loc.file.0);
-            ctx.builder.build_store(file_ptr, filename).unwrap();
-            let row_ptr = ctx
-                .builder
-                .build_in_bounds_gep(exception, &[zero, int32.const_int(2, false)], "row_ptr")
-                .unwrap();
-            ctx.builder.build_store(row_ptr, int32.const_int(loc.row as u64, false)).unwrap();
-            let col_ptr = ctx
-                .builder
-                .build_in_bounds_gep(exception, &[zero, int32.const_int(3, false)], "col_ptr")
-                .unwrap();
-            ctx.builder.build_store(col_ptr, int32.const_int(loc.column as u64, false)).unwrap();
+    if let Some(pexn) = exception {
+        let i32_model = IntModel(Int32);
+        let cslice_model = StructModel(CSlice);
 
-            let current_fun = ctx.builder.get_insert_block().unwrap().get_parent().unwrap();
-            let fun_name = ctx.gen_string(generator, current_fun.get_name().to_str().unwrap());
-            let name_ptr = ctx
-                .builder
-                .build_in_bounds_gep(exception, &[zero, int32.const_int(4, false)], "name_ptr")
-                .unwrap();
-            ctx.builder.build_store(name_ptr, fun_name).unwrap();
-        }
+        // Get and store filename
+        let filename = loc.file.0;
+        let filename = ctx.gen_string(generator, &String::from(filename)).value;
+        let filename = cslice_model.check_value(generator, ctx.ctx, filename).unwrap();
+        pexn.set(ctx, |f| f.filename, filename);
+
+        let row = i32_model.constant(generator, ctx.ctx, loc.row as u64);
+        pexn.set(ctx, |f| f.line, row);
+
+        let column = i32_model.constant(generator, ctx.ctx, loc.column as u64);
+        pexn.set(ctx, |f| f.column, column);
+
+        let current_fn = ctx.builder.get_insert_block().unwrap().get_parent().unwrap();
+        let fn_name = ctx.gen_string(generator, current_fn.get_name().to_str().unwrap());
+        pexn.set(ctx, |f| f.function, fn_name);
 
         let raise = get_builtins(generator, ctx, "__nac3_raise");
-        let exception = *exception;
-        ctx.build_call_or_invoke(raise, &[exception], "raise");
+        ctx.build_call_or_invoke(raise, &[pexn.value.into()], "raise");
     } else {
         let resume = get_builtins(generator, ctx, "__nac3_resume");
         ctx.build_call_or_invoke(resume, &[], "resume");
     }
+
     ctx.builder.build_unreachable().unwrap();
 }
 
@@ -1327,7 +1345,7 @@ pub fn gen_try<'ctx, 'a, G: CodeGenerator>(
     target: &Stmt<Option<Type>>,
 ) -> Result<(), String> {
     let StmtKind::Try { body, handlers, orelse, finalbody, .. } = &target.node else {
-        codegen_unreachable!(ctx)
+        unreachable!()
     };
 
     // if we need to generate anything related to exception, we must have personality defined
@@ -1404,7 +1422,7 @@ pub fn gen_try<'ctx, 'a, G: CodeGenerator>(
             if let TypeEnum::TObj { obj_id, .. } = &*ctx.unifier.get_ty(type_.custom.unwrap()) {
                 *obj_id
             } else {
-                codegen_unreachable!(ctx)
+                unreachable!()
             };
         let exception_name = format!("{}:{}", ctx.resolver.get_exception_id(obj_id.0), exn_name);
         let exn_id = ctx.resolver.get_string_id(&exception_name);
@@ -1676,23 +1694,6 @@ pub fn gen_return<G: CodeGenerator>(
     } else {
         None
     };
-
-    // Remap boolean return type into i1
-    let value = value.map(|ret_val| {
-        // The "return type" of a sret function is in the first parameter
-        let expected_ty = if ctx.need_sret {
-            func.get_type().get_param_types()[0]
-        } else {
-            func.get_type().get_return_type().unwrap()
-        };
-
-        if matches!(expected_ty, BasicTypeEnum::IntType(ty) if ty.get_bit_width() == 1) {
-            generator.bool_to_i1(ctx, ret_val.into_int_value()).into()
-        } else {
-            ret_val
-        }
-    });
-
     if let Some(return_target) = ctx.return_target {
         if let Some(value) = value {
             ctx.builder.build_store(ctx.return_buffer.unwrap(), value).unwrap();
@@ -1703,6 +1704,25 @@ pub fn gen_return<G: CodeGenerator>(
         ctx.builder.build_store(ctx.return_buffer.unwrap(), value.unwrap()).unwrap();
         ctx.builder.build_return(None).unwrap();
     } else {
+        // Remap boolean return type into i1
+        let value = value.map(|v| {
+            let expected_ty = func.get_type().get_return_type().unwrap();
+            let ret_val = v.as_basic_value_enum();
+
+            if expected_ty.is_int_type() && ret_val.is_int_value() {
+                let ret_type = expected_ty.into_int_type();
+                let ret_val = ret_val.into_int_value();
+
+                if ret_type.get_bit_width() == 1 && ret_val.get_type().get_bit_width() != 1 {
+                    generator.bool_to_i1(ctx, ret_val)
+                } else {
+                    ret_val
+                }
+                .into()
+            } else {
+                ret_val
+            }
+        });
         let value = value.as_ref().map(|v| v as &dyn BasicValue);
         ctx.builder.build_return(value).unwrap();
     }
@@ -1771,95 +1791,52 @@ pub fn gen_stmt<G: CodeGenerator>(
         StmtKind::Try { .. } => gen_try(generator, ctx, stmt)?,
         StmtKind::Raise { exc, .. } => {
             if let Some(exc) = exc {
-                let exn = if let ExprKind::Name { id, .. } = &exc.node {
-                    // Handle "raise Exception" short form
-                    let def_id = ctx.resolver.get_identifier_def(*id).map_err(|e| {
-                        format!("{} (at {})", e.iter().next().unwrap(), exc.location)
-                    })?;
-                    let def = ctx.top_level.definitions.read();
-                    let TopLevelDef::Class { constructor, .. } = *def[def_id.0].read() else {
-                        return Err(format!("Failed to resolve symbol {id} (at {})", exc.location));
-                    };
-
-                    let TypeEnum::TFunc(signature) =
-                        ctx.unifier.get_ty(constructor.unwrap()).as_ref().clone()
-                    else {
-                        return Err(format!("Failed to resolve symbol {id} (at {})", exc.location));
-                    };
-
-                    generator
-                        .gen_call(ctx, None, (&signature, def_id), Vec::default())?
-                        .map(Into::into)
-                } else {
-                    generator.gen_expr(ctx, exc)?
-                };
-
-                let exc = if let Some(v) = exn {
+                let exc = if let Some(v) = generator.gen_expr(ctx, exc)? {
                     v.to_basic_value_enum(ctx, generator, exc.custom.unwrap())?
                 } else {
                     return Ok(());
                 };
-                gen_raise(generator, ctx, Some(&exc), stmt.location);
+
+                let pexn_model = PtrModel(StructModel(Exception));
+                let exn = pexn_model.check_value(generator, ctx.ctx, exc).unwrap();
+
+                gen_raise(generator, ctx, Some(exn), stmt.location);
             } else {
                 gen_raise(generator, ctx, None, stmt.location);
             }
         }
         StmtKind::Assert { test, msg, .. } => {
-            let test = if let Some(v) = generator.gen_expr(ctx, test)? {
-                v.to_basic_value_enum(ctx, generator, test.custom.unwrap())?
-            } else {
+            let byte_model = IntModel(Byte);
+            let cslice_model = StructModel(CSlice);
+
+            let Some(test) = generator.gen_expr(ctx, test)? else {
                 return Ok(());
             };
+            let test = test.to_basic_value_enum(ctx, generator, ctx.primitives.bool)?;
+            let test = byte_model.check_value(generator, ctx.ctx, test).unwrap(); // Python `bool` is represented as `i8` in nac3core
+
+            // Check `msg`
             let err_msg = match msg {
                 Some(msg) => {
-                    if let Some(v) = generator.gen_expr(ctx, msg)? {
-                        v.to_basic_value_enum(ctx, generator, msg.custom.unwrap())?
-                    } else {
+                    let Some(msg) = generator.gen_expr(ctx, msg)? else {
                         return Ok(());
-                    }
+                    };
+
+                    let msg = msg.to_basic_value_enum(ctx, generator, ctx.primitives.str)?;
+                    cslice_model.check_value(generator, ctx.ctx, msg).unwrap()
                 }
-                None => ctx.gen_string(generator, "").into(),
+                None => ctx.gen_string(generator, ""),
             };
+
             ctx.make_assert_impl(
                 generator,
-                generator.bool_to_i1(ctx, test.into_int_value()),
+                test.value,
                 "0:AssertionError",
                 err_msg,
                 [None, None, None],
                 stmt.location,
             );
         }
-        StmtKind::Global { names, .. } => {
-            let registered_globals = ctx
-                .top_level
-                .definitions
-                .read()
-                .iter()
-                .filter_map(|def| {
-                    if let TopLevelDef::Variable { simple_name, ty, .. } = &*def.read() {
-                        Some((*simple_name, *ty))
-                    } else {
-                        None
-                    }
-                })
-                .collect_vec();
-
-            for id in names {
-                let Some((_, ty)) = registered_globals.iter().find(|(name, _)| name == id) else {
-                    return Err(format!("{id} is not a global at {}", stmt.location));
-                };
-
-                let resolver = ctx.resolver.clone();
-                let ptr = resolver
-                    .get_symbol_value(*id, ctx, generator)
-                    .map(|val| val.to_basic_value_enum(ctx, generator, *ty))
-                    .transpose()?
-                    .map(BasicValueEnum::into_pointer_value)
-                    .unwrap();
-
-                ctx.var_assignment.insert(*id, (ptr, None, 0));
-            }
-        }
         _ => unimplemented!(),
     };
     Ok(())

nac3core/src/codegen/structure/cslice.rs

@@ -0,0 +1,45 @@
+use inkwell::context::Context;
+
+use crate::codegen::{model::*, CodeGenerator};
+
+/// Fields of [`CSlice<'ctx>`].
+pub struct CSliceFields<'ctx, F: FieldTraversal<'ctx>> {
+    /// Pointer to data.
+    pub base: F::Out<PtrModel<IntModel<Byte>>>,
+    /// Number of bytes of data.
+    pub len: F::Out<IntModel<SizeT>>,
+}
+
+/// See <https://crates.io/crates/cslice>.
+///
+/// Additionally, see <https://github.com/m-labs/artiq/blob/b0d2705c385f64b6e6711c1726cd9178f40b598e/artiq/firmware/libeh/eh_artiq.rs>)
+/// for ARTIQ-specific notes.
+#[derive(Debug, Clone, Copy, Default)]
+pub struct CSlice;
+
+impl<'ctx> StructKind<'ctx> for CSlice {
+    type Fields<F: FieldTraversal<'ctx>> = CSliceFields<'ctx, F>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields { base: traversal.add_auto("base"), len: traversal.add_auto("len") }
+    }
+}
+
+impl StructModel<CSlice> {
+    /// Create a [`CSlice`].
+    ///
+    /// `base` and `len` must be LLVM global constants.
+    pub fn create_const<'ctx, G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+        base: Ptr<'ctx, IntModel<Byte>>,
+        len: Int<'ctx, SizeT>,
+    ) -> Struct<'ctx, CSlice> {
+        let value = self
+            .0
+            .get_struct_type(generator, ctx)
+            .const_named_struct(&[base.value.into(), len.value.into()]);
+        self.believe_value(value)
+    }
+}

nac3core/src/codegen/structure/exception.rs

@@ -0,0 +1,61 @@
+use crate::codegen::model::*;
+
+use super::cslice::CSlice;
+
+/// The LLVM int type of an Exception ID.
+pub type ExceptionId = Int32;
+
+/// Fields of [`Exception<'ctx>`]
+///
+/// The definition came from `pub struct Exception<'a>` in
+/// <https://github.com/m-labs/artiq/blob/master/artiq/firmware/libeh/eh_artiq.rs>.
+pub struct ExceptionFields<'ctx, F: FieldTraversal<'ctx>> {
+    /// nac3core's ID of the exception
+    pub id: F::Out<IntModel<ExceptionId>>,
+    /// The name of the file this `Exception` was raised in.
+    pub filename: F::Out<StructModel<CSlice>>,
+    /// The line number in the file this `Exception` was raised in.
+    pub line: F::Out<IntModel<Int32>>,
+    /// The column number in the file this `Exception` was raised in.
+    pub column: F::Out<IntModel<Int32>>,
+    /// The name of the Python function this `Exception` was raised in.
+    pub function: F::Out<StructModel<CSlice>>,
+    /// The message of this Exception.
+    ///
+    /// The message can optionally contain integer parameters `{0}`, `{1}`, and `{2}` in its string,
+    /// where they will be substituted by `params[0]`, `params[1]`, and `params[2]` respectively (as `int64_t`s).
+    /// Here is an example:
+    ///
+    /// ```ignore
+    /// "Index {0} is out of bounds! List only has {1} element(s)."
+    /// ```
+    ///
+    /// In this case, `params[0]` and `params[1]` must be specified, and `params[2]` is ***unused***.
+    /// Having only 3 parameters is a constraint in ARTIQ.
+    pub msg: F::Out<StructModel<CSlice>>,
+    pub params: [F::Out<IntModel<Int64>>; 3],
+}
+
+/// nac3core & ARTIQ's Exception
+#[derive(Debug, Clone, Copy, Default)]
+pub struct Exception;
+
+impl<'ctx> StructKind<'ctx> for Exception {
+    type Fields<F: FieldTraversal<'ctx>> = ExceptionFields<'ctx, F>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields {
+            id: traversal.add_auto("id"),
+            filename: traversal.add_auto("filename"),
+            line: traversal.add_auto("line"),
+            column: traversal.add_auto("column"),
+            function: traversal.add_auto("function"),
+            msg: traversal.add_auto("msg"),
+            params: [
+                traversal.add_auto("params[0]"),
+                traversal.add_auto("params[1]"),
+                traversal.add_auto("params[2]"),
+            ],
+        }
+    }
+}

nac3core/src/codegen/structure/list.rs

@@ -0,0 +1,70 @@
+use inkwell::values::BasicValue;
+
+use crate::{
+    codegen::{model::*, CodeGenContext, CodeGenerator},
+    typecheck::typedef::{iter_type_vars, Type, TypeEnum},
+};
+
+/// Fields of [`List`]
+pub struct ListFields<'ctx, F: FieldTraversal<'ctx>, Item: Model<'ctx>, Size: IntKind<'ctx>> {
+    /// Array pointer to content
+    pub items: F::Out<PtrModel<Item>>,
+    /// Number of items in the array
+    pub len: F::Out<IntModel<Size>>,
+}
+
+/// A list in NAC3.
+#[derive(Debug, Clone, Copy, Default)]
+pub struct List<Item, Len> {
+    /// Model of the list items
+    pub item: Item,
+    /// Model of type of integer storing the number of items on the list
+    pub len: Len,
+}
+
+impl<'ctx, Item: Model<'ctx>, Size: IntKind<'ctx>> StructKind<'ctx> for List<Item, Size> {
+    type Fields<F: FieldTraversal<'ctx>> = ListFields<'ctx, F, Item, Size>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields {
+            items: traversal.add("data", PtrModel(self.item)),
+            len: traversal.add("len", IntModel(self.len)),
+        }
+    }
+}
+
+/// A NAC3 Python List object.
+pub struct ListObject<'ctx> {
+    /// Typechecker type of the list items
+    pub item_type: Type,
+    pub value: Ptr<'ctx, StructModel<List<AnyModel<'ctx>, SizeT>>>,
+}
+
+impl<'ctx> ListObject<'ctx> {
+    /// Create a [`ListObject`] from an LLVM value and its typechecker [`Type`].
+    pub fn from_value_and_type<V: BasicValue<'ctx>, G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        list_val: V,
+        list_type: Type,
+    ) -> Self {
+        // Check typechecker type and extract `item_type`
+        let item_type = match &*ctx.unifier.get_ty(list_type) {
+            TypeEnum::TObj { obj_id, params, .. }
+                if *obj_id == ctx.primitives.list.obj_id(&ctx.unifier).unwrap() =>
+            {
+                iter_type_vars(params).next().unwrap().ty // Extract `item_type`
+            }
+            _ => {
+                panic!("Expecting type to be a list, but got {}", ctx.unifier.stringify(list_type))
+            }
+        };
+
+        let item_model = AnyModel(ctx.get_llvm_type(generator, item_type));
+        let plist_model = PtrModel(StructModel(List { item: item_model, len: SizeT }));
+
+        // Create object
+        let value = plist_model.check_value(generator, ctx.ctx, list_val).unwrap();
+        ListObject { item_type, value }
+    }
+}

nac3core/src/codegen/structure/mod.rs

@@ -0,0 +1,5 @@
+pub mod cslice;
+pub mod exception;
+pub mod list;
+pub mod ndarray;
+pub mod tuple;

nac3core/src/codegen/structure/ndarray/broadcast.rs

@@ -0,0 +1,134 @@
+use itertools::Itertools;
+
+use crate::{
+    codegen::{
+        irrt::{call_nac3_ndarray_broadcast_shapes, call_nac3_ndarray_broadcast_to},
+        model::*,
+        CodeGenContext, CodeGenerator,
+    },
+    toplevel::numpy::get_broadcast_all_ndims,
+};
+
+use super::NDArrayObject;
+
+/// Fields of [`ShapeEntry`]
+pub struct ShapeEntryFields<'ctx, F: FieldTraversal<'ctx>> {
+    pub ndims: F::Out<IntModel<SizeT>>,
+    pub shape: F::Out<PtrModel<IntModel<SizeT>>>,
+}
+
+/// An IRRT structure used in broadcasting.
+#[derive(Debug, Clone, Copy, Default)]
+pub struct ShapeEntry;
+
+impl<'ctx> StructKind<'ctx> for ShapeEntry {
+    type Fields<F: FieldTraversal<'ctx>> = ShapeEntryFields<'ctx, F>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields { ndims: traversal.add_auto("ndims"), shape: traversal.add_auto("shape") }
+    }
+}
+
+impl<'ctx> NDArrayObject<'ctx> {
+    /// Create a broadcast view on this ndarray with a target shape.
+    ///
+    /// * `target_ndims` - The ndims type after broadcasting to the given shape.
+    /// The caller has to figure this out for this function.
+    /// * `target_shape` - An array pointer pointing to the target shape.
+    #[must_use]
+    pub fn broadcast_to<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        target_ndims: u64,
+        target_shape: Ptr<'ctx, IntModel<SizeT>>,
+    ) -> Self {
+        let broadcast_ndarray = NDArrayObject::alloca_uninitialized(
+            generator,
+            ctx,
+            self.dtype,
+            target_ndims,
+            "broadcast_ndarray_to_dst",
+        );
+        broadcast_ndarray.copy_shape_from_array(generator, ctx, target_shape);
+
+        call_nac3_ndarray_broadcast_to(generator, ctx, self.value, broadcast_ndarray.value);
+        broadcast_ndarray
+    }
+}
+/// A result produced by [`broadcast_all_ndarrays`]
+#[derive(Debug, Clone)]
+pub struct BroadcastAllResult<'ctx> {
+    /// The statically known `ndims` of the broadcast result.
+    pub ndims: u64,
+    /// The broadcasting shape.
+    pub shape: Ptr<'ctx, IntModel<SizeT>>,
+    /// Broadcasted views on the inputs.
+    ///
+    /// All of them will have `shape` [`BroadcastAllResult::shape`] and
+    /// `ndims` [`BroadcastAllResult::ndims`]. The length of the vector
+    /// is the same as the input.
+    pub ndarrays: Vec<NDArrayObject<'ctx>>,
+}
+
+impl<'ctx> NDArrayObject<'ctx> {
+    // TODO: DOCUMENT: Behaves like `np.broadcast()`, except returns results differently.
+    pub fn broadcast_all<G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        ndarrays: &[Self],
+    ) -> BroadcastAllResult<'ctx> {
+        assert!(!ndarrays.is_empty());
+
+        let sizet_model = IntModel(SizeT);
+        let shape_model = StructModel(ShapeEntry);
+
+        let broadcast_ndims = get_broadcast_all_ndims(ndarrays.iter().map(|ndarray| ndarray.ndims));
+
+        // Prepare input shape entries
+        let num_shape_entries =
+            sizet_model.constant(generator, ctx.ctx, u64::try_from(ndarrays.len()).unwrap());
+        let shape_entries =
+            shape_model.array_alloca(generator, ctx, num_shape_entries.value, "shape_entries");
+        for (i, ndarray) in ndarrays.iter().enumerate() {
+            let i = sizet_model.constant(generator, ctx.ctx, i as u64).value;
+
+            let shape_entry = shape_entries.offset(generator, ctx, i, "shape_entry");
+
+            let this_ndims = ndarray.value.get(generator, ctx, |f| f.ndims, "this_ndims");
+            shape_entry.set(ctx, |f| f.ndims, this_ndims);
+
+            let this_shape = ndarray.value.get(generator, ctx, |f| f.shape, "this_shape");
+            shape_entry.set(ctx, |f| f.shape, this_shape);
+        }
+
+        // Prepare destination
+        let broadcast_ndims_llvm = sizet_model.constant(generator, ctx.ctx, broadcast_ndims);
+        let broadcast_shape =
+            sizet_model.array_alloca(generator, ctx, broadcast_ndims_llvm.value, "dst_shape");
+
+        // Compute the target broadcast shape `dst_shape` for all ndarrays.
+        call_nac3_ndarray_broadcast_shapes(
+            generator,
+            ctx,
+            num_shape_entries,
+            shape_entries,
+            broadcast_ndims_llvm,
+            broadcast_shape,
+        );
+
+        // Now that we know about the broadcasting shape, broadcast all the inputs.
+
+        // Broadcast all the inputs to shape `dst_shape`.
+        let broadcast_ndarrays: Vec<_> = ndarrays
+            .iter()
+            .map(|ndarray| ndarray.broadcast_to(generator, ctx, broadcast_ndims, broadcast_shape))
+            .collect_vec();
+
+        BroadcastAllResult {
+            ndims: broadcast_ndims,
+            shape: broadcast_shape,
+            ndarrays: broadcast_ndarrays,
+        }
+    }
+}

nac3core/src/codegen/structure/ndarray/functions.rs

@@ -0,0 +1,562 @@
+use inkwell::{
+    values::{BasicValue, FloatValue, IntValue},
+    FloatPredicate, IntPredicate,
+};
+use itertools::Itertools;
+
+use crate::{
+    codegen::{
+        llvm_intrinsics,
+        model::{
+            util::{gen_for_model_auto, gen_if_model},
+            *,
+        },
+        CodeGenContext, CodeGenerator,
+    },
+    typecheck::typedef::Type,
+};
+
+use super::{scalar::ScalarObject, NDArrayObject};
+
+/// Convenience function to crash the program when types of arguments are not supported.
+/// Used to be debugged with a stacktrace.
+fn unsupported_type<I>(ctx: &CodeGenContext<'_, '_>, tys: I) -> !
+where
+    I: IntoIterator<Item = Type>,
+{
+    unreachable!(
+        "unsupported types found '{}'",
+        tys.into_iter().map(|ty| format!("'{}'", ctx.unifier.stringify(ty))).join(", "),
+    )
+}
+
+#[derive(Debug, Clone, Copy)]
+pub enum FloorOrCeil {
+    Floor,
+    Ceil,
+}
+
+#[derive(Debug, Clone, Copy)]
+pub enum MinOrMax {
+    Min,
+    Max,
+}
+
+fn signed_ints(ctx: &CodeGenContext<'_, '_>) -> Vec<Type> {
+    vec![ctx.primitives.int32, ctx.primitives.int64]
+}
+
+fn unsigned_ints(ctx: &CodeGenContext<'_, '_>) -> Vec<Type> {
+    vec![ctx.primitives.uint32, ctx.primitives.uint64]
+}
+
+fn ints(ctx: &CodeGenContext<'_, '_>) -> Vec<Type> {
+    vec![ctx.primitives.int32, ctx.primitives.int64, ctx.primitives.uint32, ctx.primitives.uint64]
+}
+
+fn int_like(ctx: &CodeGenContext<'_, '_>) -> Vec<Type> {
+    vec![
+        ctx.primitives.bool,
+        ctx.primitives.int32,
+        ctx.primitives.int64,
+        ctx.primitives.uint32,
+        ctx.primitives.uint64,
+    ]
+}
+
+fn cast_to_int_conversion<'ctx, 'a, G, HandleFloatFn>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, 'a>,
+    scalar: ScalarObject<'ctx>,
+    ret_int_dtype: Type,
+    handle_float: HandleFloatFn,
+) -> ScalarObject<'ctx>
+where
+    G: CodeGenerator + ?Sized,
+    HandleFloatFn:
+        FnOnce(&mut G, &mut CodeGenContext<'ctx, 'a>, FloatValue<'ctx>) -> IntValue<'ctx>,
+{
+    let ret_int_dtype_llvm = ctx.get_llvm_type(generator, ret_int_dtype).into_int_type();
+
+    let result = if ctx.unifier.unioned(scalar.dtype, ctx.primitives.float) {
+        // Special handling for floats
+        let n = scalar.value.into_float_value();
+        handle_float(generator, ctx, n)
+    } else if ctx.unifier.unioned_any(scalar.dtype, int_like(ctx)) {
+        let n = scalar.value.into_int_value();
+
+        if n.get_type().get_bit_width() <= ret_int_dtype_llvm.get_bit_width() {
+            ctx.builder.build_int_z_extend(n, ret_int_dtype_llvm, "zext").unwrap()
+        } else {
+            ctx.builder.build_int_truncate(n, ret_int_dtype_llvm, "trunc").unwrap()
+        }
+    } else {
+        unsupported_type(ctx, [scalar.dtype]);
+    };
+
+    assert_eq!(ret_int_dtype_llvm.get_bit_width(), result.get_type().get_bit_width()); // Sanity check
+    ScalarObject { value: result.into(), dtype: ret_int_dtype }
+}
+
+impl<'ctx> ScalarObject<'ctx> {
+    /// Convenience function. Assume this scalar has typechecker type float64, get its underlying LLVM value.
+    ///
+    /// Panic if the type is wrong.
+    pub fn into_float64(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> FloatValue<'ctx> {
+        if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            self.value.into_float_value() // self.value must be a FloatValue
+        } else {
+            panic!("not a float type")
+        }
+    }
+
+    /// Convenience function. Assume this scalar has typechecker type int32, get its underlying LLVM value.
+    ///
+    /// Panic if the type is wrong.
+    pub fn into_int32(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> IntValue<'ctx> {
+        if ctx.unifier.unioned(self.dtype, ctx.primitives.int32) {
+            let value = self.value.into_int_value();
+            debug_assert_eq!(value.get_type().get_bit_width(), 32); // Sanity check
+            value
+        } else {
+            panic!("not a float type")
+        }
+    }
+
+    /// Compare two scalars. Only int-to-int and float-to-float comparisons are allowed.
+    /// Panic otherwise.
+    pub fn compare<G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        lhs: ScalarObject<'ctx>,
+        rhs: ScalarObject<'ctx>,
+        int_predicate: IntPredicate,
+        float_predicate: FloatPredicate,
+        name: &str,
+    ) -> Int<'ctx, Bool> {
+        if !ctx.unifier.unioned(lhs.dtype, rhs.dtype) {
+            unsupported_type(ctx, [lhs.dtype, rhs.dtype]);
+        }
+
+        let bool_model = IntModel(Bool);
+
+        let common_ty = lhs.dtype;
+        let result = if ctx.unifier.unioned(common_ty, ctx.primitives.float) {
+            let lhs = lhs.value.into_float_value();
+            let rhs = rhs.value.into_float_value();
+            ctx.builder.build_float_compare(float_predicate, lhs, rhs, name).unwrap()
+        } else if ctx.unifier.unioned_any(common_ty, int_like(ctx)) {
+            let lhs = lhs.value.into_int_value();
+            let rhs = rhs.value.into_int_value();
+            ctx.builder.build_int_compare(int_predicate, lhs, rhs, name).unwrap()
+        } else {
+            unsupported_type(ctx, [lhs.dtype, rhs.dtype]);
+        };
+
+        bool_model.check_value(generator, ctx.ctx, result).unwrap()
+    }
+
+    /// Invoke NAC3's builtin `int32()`.
+    #[must_use]
+    pub fn cast_to_int32<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Self {
+        cast_to_int_conversion(
+            generator,
+            ctx,
+            *self,
+            ctx.primitives.int32,
+            |_generator, ctx, input| {
+                let n =
+                    ctx.builder.build_float_to_signed_int(input, ctx.ctx.i64_type(), "").unwrap();
+                ctx.builder.build_int_truncate(n, ctx.ctx.i32_type(), "conv").unwrap()
+            },
+        )
+    }
+
+    /// Invoke NAC3's builtin `int64()`.
+    #[must_use]
+    pub fn cast_to_int64<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Self {
+        cast_to_int_conversion(
+            generator,
+            ctx,
+            *self,
+            ctx.primitives.int64,
+            |_generator, ctx, input| {
+                ctx.builder.build_float_to_signed_int(input, ctx.ctx.i64_type(), "").unwrap()
+            },
+        )
+    }
+
+    /// Invoke NAC3's builtin `uint32()`.
+    #[must_use]
+    pub fn cast_to_uint32<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Self {
+        cast_to_int_conversion(
+            generator,
+            ctx,
+            *self,
+            ctx.primitives.uint32,
+            |_generator, ctx, n| {
+                let n_gez = ctx
+                    .builder
+                    .build_float_compare(FloatPredicate::OGE, n, n.get_type().const_zero(), "")
+                    .unwrap();
+
+                let to_int32 =
+                    ctx.builder.build_float_to_signed_int(n, ctx.ctx.i32_type(), "").unwrap();
+                let to_uint64 =
+                    ctx.builder.build_float_to_unsigned_int(n, ctx.ctx.i64_type(), "").unwrap();
+
+                ctx.builder
+                    .build_select(
+                        n_gez,
+                        ctx.builder.build_int_truncate(to_uint64, ctx.ctx.i32_type(), "").unwrap(),
+                        to_int32,
+                        "conv",
+                    )
+                    .unwrap()
+                    .into_int_value()
+            },
+        )
+    }
+
+    /// Invoke NAC3's builtin `uint64()`.
+    #[must_use]
+    pub fn cast_to_uint64<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Self {
+        cast_to_int_conversion(
+            generator,
+            ctx,
+            *self,
+            ctx.primitives.uint64,
+            |_generator, ctx, n| {
+                let val_gez = ctx
+                    .builder
+                    .build_float_compare(FloatPredicate::OGE, n, n.get_type().const_zero(), "")
+                    .unwrap();
+
+                let to_int64 =
+                    ctx.builder.build_float_to_signed_int(n, ctx.ctx.i64_type(), "").unwrap();
+                let to_uint64 =
+                    ctx.builder.build_float_to_unsigned_int(n, ctx.ctx.i64_type(), "").unwrap();
+                ctx.builder
+                    .build_select(val_gez, to_uint64, to_int64, "conv")
+                    .unwrap()
+                    .into_int_value()
+            },
+        )
+    }
+
+    /// Invoke NAC3's builtin `bool()`.
+    #[must_use]
+    pub fn cast_to_bool(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> Self {
+        // TODO: Why is the original code being so lax about i1 and i8 for the returned int type?
+        let result = if ctx.unifier.unioned(self.dtype, ctx.primitives.bool) {
+            self.value.into_int_value()
+        } else if ctx.unifier.unioned_any(self.dtype, ints(ctx)) {
+            let n = self.value.into_int_value();
+            ctx.builder
+                .build_int_compare(inkwell::IntPredicate::NE, n, n.get_type().const_zero(), "bool")
+                .unwrap()
+        } else if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            let n = self.value.into_float_value();
+            ctx.builder
+                .build_float_compare(FloatPredicate::UNE, n, n.get_type().const_zero(), "bool")
+                .unwrap()
+        } else {
+            unsupported_type(ctx, [self.dtype])
+        };
+
+        ScalarObject { dtype: ctx.primitives.bool, value: result.as_basic_value_enum() }
+    }
+
+    /// Invoke NAC3's builtin `float()`.
+    #[must_use]
+    pub fn cast_to_float(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> Self {
+        let llvm_f64 = ctx.ctx.f64_type();
+
+        let result: FloatValue<'_> = if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            self.value.into_float_value()
+        } else if ctx
+            .unifier
+            .unioned_any(self.dtype, [signed_ints(ctx).as_slice(), &[ctx.primitives.bool]].concat())
+        {
+            let n = self.value.into_int_value();
+            ctx.builder.build_signed_int_to_float(n, llvm_f64, "sitofp").unwrap()
+        } else if ctx.unifier.unioned_any(self.dtype, unsigned_ints(ctx)) {
+            let n = self.value.into_int_value();
+            ctx.builder.build_unsigned_int_to_float(n, llvm_f64, "uitofp").unwrap()
+        } else {
+            unsupported_type(ctx, [self.dtype]);
+        };
+
+        ScalarObject { value: result.as_basic_value_enum(), dtype: ctx.primitives.float }
+    }
+
+    /// Invoke NAC3's builtin `round()`.
+    #[must_use]
+    pub fn round<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        ret_int_dtype: Type,
+    ) -> Self {
+        let ret_int_dtype_llvm = ctx.get_llvm_type(generator, ret_int_dtype).into_int_type();
+
+        let result = if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            let n = self.value.into_float_value();
+            let n = llvm_intrinsics::call_float_round(ctx, n, None);
+            ctx.builder.build_float_to_signed_int(n, ret_int_dtype_llvm, "round").unwrap()
+        } else {
+            unsupported_type(ctx, [self.dtype, ret_int_dtype])
+        };
+        ScalarObject { dtype: ret_int_dtype, value: result.as_basic_value_enum() }
+    }
+
+    /// Invoke NAC3's builtin `np_round()`.
+    ///
+    /// NOTE: `np.round()` has different behaviors than `round()` in terms of their result
+    /// on "tie" cases and return type.
+    #[must_use]
+    pub fn np_round(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> Self {
+        let result = if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            let n = self.value.into_float_value();
+            llvm_intrinsics::call_float_rint(ctx, n, None)
+        } else {
+            unsupported_type(ctx, [self.dtype])
+        };
+        ScalarObject { dtype: ctx.primitives.float, value: result.as_basic_value_enum() }
+    }
+
+    /// Invoke NAC3's builtin `min()` or `max()`.
+    pub fn min_or_max(
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        kind: MinOrMax,
+        a: Self,
+        b: Self,
+    ) -> Self {
+        if !ctx.unifier.unioned(a.dtype, b.dtype) {
+            unsupported_type(ctx, [a.dtype, b.dtype])
+        }
+
+        let common_dtype = a.dtype;
+
+        if ctx.unifier.unioned(common_dtype, ctx.primitives.float) {
+            let function = match kind {
+                MinOrMax::Min => llvm_intrinsics::call_float_minnum,
+                MinOrMax::Max => llvm_intrinsics::call_float_maxnum,
+            };
+            let result =
+                function(ctx, a.value.into_float_value(), b.value.into_float_value(), None);
+            ScalarObject { value: result.as_basic_value_enum(), dtype: ctx.primitives.float }
+        } else if ctx.unifier.unioned_any(
+            common_dtype,
+            [unsigned_ints(ctx).as_slice(), &[ctx.primitives.bool]].concat(),
+        ) {
+            // Treating bool has an unsigned int since that is convenient
+            let function = match kind {
+                MinOrMax::Min => llvm_intrinsics::call_int_umin,
+                MinOrMax::Max => llvm_intrinsics::call_int_umax,
+            };
+            let result = function(ctx, a.value.into_int_value(), b.value.into_int_value(), None);
+            ScalarObject { value: result.as_basic_value_enum(), dtype: common_dtype }
+        } else {
+            unsupported_type(ctx, [common_dtype])
+        }
+    }
+
+    /// Invoke NAC3's builtin `floor()` or `ceil()`.
+    ///
+    /// * `ret_int_dtype` - The type of int to return.
+    ///
+    /// Takes in a float/int and returns an int of type `ret_int_dtype`
+    #[must_use]
+    pub fn floor_or_ceil<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        kind: FloorOrCeil,
+        ret_int_dtype: Type,
+    ) -> Self {
+        let ret_int_dtype_llvm = ctx.get_llvm_type(generator, ret_int_dtype).into_int_type();
+
+        if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            let function = match kind {
+                FloorOrCeil::Floor => llvm_intrinsics::call_float_floor,
+                FloorOrCeil::Ceil => llvm_intrinsics::call_float_ceil,
+            };
+            let n = self.value.into_float_value();
+            let n = function(ctx, n, None);
+
+            let n = ctx.builder.build_float_to_signed_int(n, ret_int_dtype_llvm, "").unwrap();
+            ScalarObject { dtype: ret_int_dtype, value: n.as_basic_value_enum() }
+        } else {
+            unsupported_type(ctx, [self.dtype])
+        }
+    }
+
+    /// Invoke NAC3's builtin `np_floor()`/ `np_ceil()`.
+    ///
+    /// Takes in a float/int and returns a float64 result.
+    #[must_use]
+    pub fn np_floor_or_ceil(&self, ctx: &mut CodeGenContext<'ctx, '_>, kind: FloorOrCeil) -> Self {
+        if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            let function = match kind {
+                FloorOrCeil::Floor => llvm_intrinsics::call_float_floor,
+                FloorOrCeil::Ceil => llvm_intrinsics::call_float_ceil,
+            };
+            let n = self.value.into_float_value();
+            let n = function(ctx, n, None);
+            ScalarObject { dtype: ctx.primitives.float, value: n.as_basic_value_enum() }
+        } else {
+            unsupported_type(ctx, [self.dtype])
+        }
+    }
+
+    /// Invoke NAC3's builtin `abs()`.
+    pub fn abs(&self, ctx: &mut CodeGenContext<'ctx, '_>) -> Self {
+        if ctx.unifier.unioned(self.dtype, ctx.primitives.float) {
+            let n = self.value.into_float_value();
+            let n = llvm_intrinsics::call_float_fabs(ctx, n, Some("abs"));
+            ScalarObject { value: n.into(), dtype: ctx.primitives.float }
+        } else if ctx.unifier.unioned_any(self.dtype, ints(ctx)) {
+            let n = self.value.into_int_value();
+
+            let is_poisoned = ctx.ctx.bool_type().const_zero(); // is_poisoned = false
+            let n = llvm_intrinsics::call_int_abs(ctx, n, is_poisoned, Some("abs"));
+
+            ScalarObject { value: n.into(), dtype: self.dtype }
+        } else {
+            unsupported_type(ctx, [self.dtype])
+        }
+    }
+}
+
+impl<'ctx> NDArrayObject<'ctx> {
+    /// Helper function to implement NAC3's builtin `np_min()`, `np_max()`, `np_argmin()`, and `np_argmax()`.
+    ///
+    /// Generate LLVM IR to find the extremum and index of the **first** extremum value.
+    ///
+    /// Care has also been taken to make the error messages match that of NumPy.
+    fn min_max_argmin_argmax_helper<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        kind: MinOrMax,
+        on_empty_err_msg: &str,
+    ) -> (ScalarObject<'ctx>, Int<'ctx, SizeT>) {
+        let sizet_model = IntModel(SizeT);
+        let dtype_llvm = ctx.get_llvm_type(generator, self.dtype);
+
+        // If the ndarray is empty, throw an error.
+        let is_empty = self.is_empty(generator, ctx);
+        ctx.make_assert(
+            generator,
+            is_empty.value,
+            "0:ValueError",
+            on_empty_err_msg,
+            [None, None, None],
+            ctx.current_loc,
+        );
+
+        // Setup and initialize the extremum to be the first element in the ndarray
+        let pextremum_index = sizet_model.alloca(generator, ctx, "extremum_index");
+        let pextremum = ctx.builder.build_alloca(dtype_llvm, "extremum").unwrap();
+
+        let zero = sizet_model.const_0(generator, ctx.ctx);
+        pextremum_index.store(ctx, zero);
+
+        let first_scalar = self.get_nth(generator, ctx, zero);
+        ctx.builder.build_store(pextremum, first_scalar.value).unwrap();
+
+        // Find extremum
+        let start = sizet_model.const_1(generator, ctx.ctx); // Start on 1
+        let stop = self.size(generator, ctx);
+        let step = sizet_model.const_1(generator, ctx.ctx);
+        gen_for_model_auto(generator, ctx, start, stop, step, |generator, ctx, _hooks, i| {
+            // Worth reading on "Notes" in <https://numpy.org/doc/stable/reference/generated/numpy.min.html#numpy.min>
+            // on how `NaN` values have to be handled.
+
+            let scalar = self.get_nth(generator, ctx, i);
+
+            let old_extremum = ctx.builder.build_load(pextremum, "current_extremum").unwrap();
+            let old_extremum = ScalarObject { dtype: self.dtype, value: old_extremum };
+
+            let new_extremum = ScalarObject::min_or_max(ctx, kind, old_extremum, scalar);
+
+            // Check if new_extremum is more extreme than old_extremum.
+            let update_index = ScalarObject::compare(
+                generator,
+                ctx,
+                new_extremum,
+                old_extremum,
+                IntPredicate::NE,
+                FloatPredicate::ONE,
+                "",
+            );
+
+            gen_if_model(generator, ctx, update_index, |_generator, ctx| {
+                pextremum_index.store(ctx, i);
+                Ok(())
+            })
+            .unwrap();
+            Ok(())
+        })
+        .unwrap();
+
+        // Finally return the extremum and extremum index.
+        let extremum_index = pextremum_index.load(generator, ctx, "extremum_index");
+
+        let extremum = ctx.builder.build_load(pextremum, "extremum_value").unwrap();
+        let extremum = ScalarObject { dtype: self.dtype, value: extremum };
+
+        (extremum, extremum_index)
+    }
+
+    /// Invoke NAC3's builtin `np_min()` or `np_max()`.
+    pub fn min_or_max<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        kind: MinOrMax,
+    ) -> ScalarObject<'ctx> {
+        let on_empty_err_msg = format!(
+            "zero-size array to reduction operation {} which has no identity",
+            match kind {
+                MinOrMax::Min => "minimum",
+                MinOrMax::Max => "maximum",
+            }
+        );
+        self.min_max_argmin_argmax_helper(generator, ctx, kind, &on_empty_err_msg).0
+    }
+
+    /// Invoke NAC3's builtin `np_argmin()` or `np_argmax()`.
+    pub fn argmin_or_argmax<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        kind: MinOrMax,
+    ) -> Int<'ctx, SizeT> {
+        let on_empty_err_msg = format!(
+            "attempt to get {} of an empty sequence",
+            match kind {
+                MinOrMax::Min => "argmin",
+                MinOrMax::Max => "argmax",
+            }
+        );
+        self.min_max_argmin_argmax_helper(generator, ctx, kind, &on_empty_err_msg).1
+    }
+}

nac3core/src/codegen/structure/ndarray/indexing.rs

@@ -0,0 +1,346 @@
+use crate::codegen::{irrt::call_nac3_ndarray_index, model::*, CodeGenContext, CodeGenerator};
+
+use super::{scalar::ScalarOrNDArray, NDArrayObject};
+
+pub type NDIndexType = Byte;
+
+/// Fields of [`NDIndex`]
+#[derive(Debug, Clone, Copy)]
+pub struct NDIndexFields<'ctx, F: FieldTraversal<'ctx>> {
+    pub type_: F::Out<IntModel<NDIndexType>>, // Defined to be uint8_t in IRRT
+    pub data: F::Out<PtrModel<IntModel<Byte>>>,
+}
+
+/// An IRRT representation fo an ndarray subscript index.
+#[derive(Debug, Clone, Copy, Default, PartialEq, Eq)]
+pub struct NDIndex;
+
+impl<'ctx> StructKind<'ctx> for NDIndex {
+    type Fields<F: FieldTraversal<'ctx>> = NDIndexFields<'ctx, F>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields { type_: traversal.add_auto("type"), data: traversal.add_auto("data") }
+    }
+}
+
+/// Fields of [`UserSlice`]
+#[derive(Debug, Clone)]
+pub struct UserSliceFields<'ctx, F: FieldTraversal<'ctx>> {
+    pub start_defined: F::Out<IntModel<Bool>>,
+    pub start: F::Out<IntModel<Int32>>,
+    pub stop_defined: F::Out<IntModel<Bool>>,
+    pub stop: F::Out<IntModel<Int32>>,
+    pub step_defined: F::Out<IntModel<Bool>>,
+    pub step: F::Out<IntModel<Int32>>,
+}
+
+/// An IRRT representation of a user slice.
+#[derive(Debug, Clone, Copy, Default, PartialEq, Eq)]
+pub struct UserSlice;
+
+impl<'ctx> StructKind<'ctx> for UserSlice {
+    type Fields<F: FieldTraversal<'ctx>> = UserSliceFields<'ctx, F>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields {
+            start_defined: traversal.add_auto("start_defined"),
+            start: traversal.add_auto("start"),
+            stop_defined: traversal.add_auto("stop_defined"),
+            stop: traversal.add_auto("stop"),
+            step_defined: traversal.add_auto("step_defined"),
+            step: traversal.add_auto("step"),
+        }
+    }
+}
+
+/// A convenience structure to prepare a [`UserSlice`].
+#[derive(Debug, Clone)]
+pub struct RustUserSlice<'ctx> {
+    pub start: Option<Int<'ctx, Int32>>,
+    pub stop: Option<Int<'ctx, Int32>>,
+    pub step: Option<Int<'ctx, Int32>>,
+}
+
+impl<'ctx> RustUserSlice<'ctx> {
+    /// Write the contents to an LLVM [`UserSlice`].
+    pub fn write_to_user_slice<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        dst_slice_ptr: Ptr<'ctx, StructModel<UserSlice>>,
+    ) {
+        let bool_model = IntModel(Bool);
+
+        let false_ = bool_model.constant(generator, ctx.ctx, 0);
+        let true_ = bool_model.constant(generator, ctx.ctx, 1);
+
+        // TODO: Code duplication. Probably okay...?
+
+        match self.start {
+            Some(start) => {
+                dst_slice_ptr.gep(ctx, |f| f.start_defined).store(ctx, true_);
+                dst_slice_ptr.gep(ctx, |f| f.start).store(ctx, start);
+            }
+            None => dst_slice_ptr.gep(ctx, |f| f.start_defined).store(ctx, false_),
+        }
+
+        match self.stop {
+            Some(stop) => {
+                dst_slice_ptr.gep(ctx, |f| f.stop_defined).store(ctx, true_);
+                dst_slice_ptr.gep(ctx, |f| f.stop).store(ctx, stop);
+            }
+            None => dst_slice_ptr.gep(ctx, |f| f.stop_defined).store(ctx, false_),
+        }
+
+        match self.step {
+            Some(step) => {
+                dst_slice_ptr.gep(ctx, |f| f.step_defined).store(ctx, true_);
+                dst_slice_ptr.gep(ctx, |f| f.step).store(ctx, step);
+            }
+            None => dst_slice_ptr.gep(ctx, |f| f.step_defined).store(ctx, false_),
+        }
+    }
+}
+
+// A convenience enum variant to store the content and type of an NDIndex in high level.
+#[derive(Debug, Clone)]
+pub enum RustNDIndex<'ctx> {
+    SingleElement(Int<'ctx, Int32>), // TODO: To be SizeT
+    Slice(RustUserSlice<'ctx>),
+    NewAxis,
+    Ellipsis,
+}
+
+impl<'ctx> RustNDIndex<'ctx> {
+    /// Get the value to set `NDIndex::type` for this variant.
+    fn get_type_id(&self) -> u64 {
+        // Defined in IRRT, must be in sync
+        match self {
+            RustNDIndex::SingleElement(_) => 0,
+            RustNDIndex::Slice(_) => 1,
+            RustNDIndex::NewAxis => 2,
+            RustNDIndex::Ellipsis => 3,
+        }
+    }
+
+    /// Write the contents to an LLVM [`NDIndex`].
+    fn write_to_ndindex<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        dst_ndindex_ptr: Ptr<'ctx, StructModel<NDIndex>>,
+    ) {
+        let ndindex_type_model = IntModel(NDIndexType::default());
+        let i32_model = IntModel(Int32);
+        let user_slice_model = StructModel(UserSlice);
+
+        // Set `dst_ndindex_ptr->type`
+        dst_ndindex_ptr
+            .gep(ctx, |f| f.type_)
+            .store(ctx, ndindex_type_model.constant(generator, ctx.ctx, self.get_type_id()));
+
+        // Set `dst_ndindex_ptr->data`
+        match self {
+            RustNDIndex::SingleElement(in_index) => {
+                let index_ptr = i32_model.alloca(generator, ctx, "index");
+                index_ptr.store(ctx, *in_index);
+
+                dst_ndindex_ptr
+                    .gep(ctx, |f| f.data)
+                    .store(ctx, index_ptr.transmute(generator, ctx, IntModel(Byte), ""));
+            }
+            RustNDIndex::Slice(in_rust_slice) => {
+                let user_slice_ptr = user_slice_model.alloca(generator, ctx, "user_slice");
+                in_rust_slice.write_to_user_slice(generator, ctx, user_slice_ptr);
+
+                dst_ndindex_ptr
+                    .gep(ctx, |f| f.data)
+                    .store(ctx, user_slice_ptr.transmute(generator, ctx, IntModel(Byte), ""));
+            }
+            RustNDIndex::NewAxis => {}
+            RustNDIndex::Ellipsis => {}
+        }
+    }
+
+    /// Allocate an array of `NDIndex`es on the stack and return its stack pointer.
+    pub fn alloca_ndindexes<G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &CodeGenContext<'ctx, '_>,
+        in_ndindexes: &[RustNDIndex<'ctx>],
+    ) -> (Int<'ctx, SizeT>, Ptr<'ctx, StructModel<NDIndex>>) {
+        let sizet_model = IntModel(SizeT);
+        let ndindex_model = StructModel(NDIndex);
+
+        let num_ndindexes = sizet_model.constant(generator, ctx.ctx, in_ndindexes.len() as u64);
+        let ndindexes =
+            ndindex_model.array_alloca(generator, ctx, num_ndindexes.value, "ndindexes");
+        for (i, in_ndindex) in in_ndindexes.iter().enumerate() {
+            let i = sizet_model.constant(generator, ctx.ctx, i as u64);
+            let pndindex = ndindexes.offset(generator, ctx, i.value, "");
+            in_ndindex.write_to_ndindex(generator, ctx, pndindex);
+        }
+
+        (num_ndindexes, ndindexes)
+    }
+}
+
+impl<'ctx> NDArrayObject<'ctx> {
+    /// Get the ndims [`Type`] after indexing with a given slice.
+    #[must_use]
+    pub fn deduce_ndims_after_indexing_with(&self, indexes: &[RustNDIndex<'ctx>]) -> u64 {
+        let mut ndims = self.ndims;
+        for index in indexes {
+            match index {
+                RustNDIndex::SingleElement(_) => {
+                    ndims -= 1; // Single elements decrements ndims
+                }
+                RustNDIndex::Slice(_) => {}
+                RustNDIndex::NewAxis => {
+                    ndims += 1; // `np.newaxis` / `none` adds a new axis
+                }
+                RustNDIndex::Ellipsis => {}
+            }
+        }
+        ndims
+    }
+
+    /// Index into the ndarray, and return a newly-allocated view on this ndarray.
+    ///
+    /// This function behaves like NumPy's ndarray indexing, but if the indexes index
+    /// into a single element, an unsized ndarray is returned.
+    #[must_use]
+    pub fn index<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        indexes: &[RustNDIndex<'ctx>],
+        name: &str,
+    ) -> Self {
+        let dst_ndims = self.deduce_ndims_after_indexing_with(indexes);
+        let dst_ndarray =
+            NDArrayObject::alloca_uninitialized(generator, ctx, self.dtype, dst_ndims, name);
+
+        let (num_indexes, indexes) = RustNDIndex::alloca_ndindexes(generator, ctx, indexes);
+        call_nac3_ndarray_index(
+            generator,
+            ctx,
+            num_indexes,
+            indexes,
+            self.value,
+            dst_ndarray.value,
+        );
+
+        dst_ndarray
+    }
+
+    /// Like [`NDArrayObject::index`] but returns a scalar if the indexes index
+    /// into a single element.
+    #[must_use]
+    pub fn index_or_scalar<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        indexes: &[RustNDIndex<'ctx>],
+        name: &str,
+    ) -> ScalarOrNDArray<'ctx> {
+        let sizet_model = IntModel(SizeT);
+        let zero = sizet_model.const_0(generator, ctx.ctx);
+
+        let subndarray = self.index(generator, ctx, indexes, name);
+        if subndarray.is_unsized() {
+            // NOTE: `np.size(self) == 0` here is never possible.
+            ScalarOrNDArray::Scalar(subndarray.get_nth(generator, ctx, zero))
+        } else {
+            ScalarOrNDArray::NDArray(subndarray)
+        }
+    }
+}
+
+pub mod util {
+    use itertools::Itertools;
+    use nac3parser::ast::{Constant, Expr, ExprKind};
+
+    use crate::{
+        codegen::{expr::gen_slice, model::*, CodeGenContext, CodeGenerator},
+        typecheck::typedef::{Type, TypeEnum},
+    };
+
+    use super::{RustNDIndex, RustUserSlice};
+
+    /// Generate LLVM code to transform an ndarray subscript expression to
+    /// its list of [`RustNDIndex`]
+    ///
+    /// i.e.,
+    /// ```python
+    /// my_ndarray[::3, 1, :2:]
+    ///            ^^^^^^^^^^^ Then these into a three `RustNDIndex`es
+    /// ```
+    pub fn gen_ndarray_subscript_ndindexes<'ctx, G: CodeGenerator>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        subscript: &Expr<Option<Type>>,
+    ) -> Result<Vec<RustNDIndex<'ctx>>, String> {
+        // TODO: Support https://numpy.org/doc/stable/user/basics.indexing.html#dimensional-indexing-tools
+        let i32_model = IntModel(Int32);
+
+        // Annoying notes about `slice`
+        //  - `my_array[5]`
+        //    - slice is a `Constant`
+        //  - `my_array[:5]`
+        //    - slice is a `Slice`
+        //  - `my_array[:]`
+        //    - slice is a `Slice`, but lower upper step would all be `Option::None`
+        //  - `my_array[:, :]`
+        //    - slice is now a `Tuple` of two `Slice`-s
+        //
+        // In summary:
+        //  - when there is a comma "," within [], `slice` will be a `Tuple` of the entries.
+        //  - when there is not comma "," within [] (i.e., just a single entry), `slice` will be that entry itself.
+        //
+        // So we first "flatten" out the slice expression
+        let index_exprs = match &subscript.node {
+            ExprKind::Tuple { elts, .. } => elts.iter().collect_vec(),
+            _ => vec![subscript],
+        };
+
+        // Process all index expressions
+        let mut rust_ndindexes: Vec<RustNDIndex> = Vec::with_capacity(index_exprs.len()); // Not using iterators here because `?` is used here.
+        for index_expr in index_exprs {
+            // NOTE: Currently nac3core's slices do not have an object representation,
+            // so the code/implementation looks awkward - we have to do pattern matching on the expression
+            let ndindex = if let ExprKind::Slice { lower, upper, step } = &index_expr.node {
+                // Handle slices
+
+                // Helper function here to deduce code duplication
+                let (lower, upper, step) = gen_slice(generator, ctx, lower, upper, step)?;
+                RustNDIndex::Slice(RustUserSlice { start: lower, stop: upper, step })
+            } else if let ExprKind::Constant { value: Constant::Ellipsis, .. } = &index_expr.node {
+                // Handle '...'
+                RustNDIndex::Ellipsis
+            } else {
+                match &*ctx.unifier.get_ty(index_expr.custom.unwrap()) {
+                    TypeEnum::TObj { obj_id, .. }
+                        if *obj_id == ctx.primitives.option.obj_id(&ctx.unifier).unwrap() =>
+                    {
+                        // Handle `np.newaxis` / `None`
+                        RustNDIndex::NewAxis
+                    }
+                    _ => {
+                        // Treat and handle everything else as a single element index.
+                        let index =
+                            generator.gen_expr(ctx, index_expr)?.unwrap().to_basic_value_enum(
+                                ctx,
+                                generator,
+                                ctx.primitives.int32, // Must be int32, this checks for illegal values
+                            )?;
+                        let index = i32_model.check_value(generator, ctx.ctx, index).unwrap();
+
+                        RustNDIndex::SingleElement(index)
+                    }
+                }
+            };
+            rust_ndindexes.push(ndindex);
+        }
+        Ok(rust_ndindexes)
+    }
+}

nac3core/src/codegen/structure/ndarray/mapping.rs

@@ -0,0 +1,158 @@
+use inkwell::values::BasicValueEnum;
+use itertools::Itertools;
+use util::gen_for_model_auto;
+
+use crate::{
+    codegen::{
+        model::*,
+        structure::ndarray::{scalar::ScalarObject, NDArrayObject},
+        CodeGenContext, CodeGenerator,
+    },
+    typecheck::typedef::Type,
+};
+
+use super::scalar::ScalarOrNDArray;
+
+impl<'ctx> NDArrayObject<'ctx> {
+    /// TODO: Document me. Has complex behavior.
+    /// and explain why `ret_dtype` has to be specified beforehand.
+    pub fn broadcasting_starmap<'a, G, MappingFn>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, 'a>,
+        ndarrays: &[Self],
+        ret_dtype: Type,
+        mapping: MappingFn,
+    ) -> Result<Self, String>
+    where
+        G: CodeGenerator + ?Sized,
+        MappingFn: FnOnce(
+            &mut G,
+            &mut CodeGenContext<'ctx, 'a>,
+            Int<'ctx, SizeT>,
+            &[ScalarObject<'ctx>],
+        ) -> Result<BasicValueEnum<'ctx>, String>,
+    {
+        let sizet_model = IntModel(SizeT);
+
+        // Broadcast inputs
+        let broadcast_result = NDArrayObject::broadcast_all(generator, ctx, ndarrays);
+
+        // Allocate the resulting ndarray
+        let mapped_ndarray = NDArrayObject::alloca_uninitialized(
+            generator,
+            ctx,
+            ret_dtype,
+            broadcast_result.ndims,
+            "mapped_ndarray",
+        );
+        mapped_ndarray.copy_shape_from_array(generator, ctx, broadcast_result.shape);
+        mapped_ndarray.create_data(generator, ctx);
+
+        // Map element-wise and store results into `mapped_ndarray`.
+        let start = sizet_model.const_0(generator, ctx.ctx);
+        let stop = broadcast_result.ndarrays[0].size(generator, ctx); // They all should have the same `np.size`.
+        let step = sizet_model.const_1(generator, ctx.ctx);
+        gen_for_model_auto(generator, ctx, start, stop, step, move |generator, ctx, _hooks, i| {
+            let elements =
+                ndarrays.iter().map(|ndarray| ndarray.get_nth(generator, ctx, i)).collect_vec();
+
+            let ret = mapping(generator, ctx, i, &elements)?;
+
+            let pret = mapped_ndarray.get_nth_pointer(generator, ctx, i, "pret");
+            ctx.builder.build_store(pret, ret).unwrap();
+            Ok(())
+        })?;
+
+        Ok(mapped_ndarray)
+    }
+
+    pub fn map<'a, G, Mapping>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, 'a>,
+        ret_dtype: Type,
+        mapping: Mapping,
+    ) -> Result<Self, String>
+    where
+        G: CodeGenerator + ?Sized,
+        Mapping: FnOnce(
+            &mut G,
+            &mut CodeGenContext<'ctx, 'a>,
+            Int<'ctx, SizeT>,
+            ScalarObject<'ctx>,
+        ) -> Result<BasicValueEnum<'ctx>, String>,
+    {
+        NDArrayObject::broadcasting_starmap(
+            generator,
+            ctx,
+            &[*self],
+            ret_dtype,
+            |generator, ctx, i, scalars| mapping(generator, ctx, i, scalars[0]),
+        )
+    }
+}
+
+impl<'ctx> ScalarOrNDArray<'ctx> {
+    /// TODO: Document me. Has complex behavior.
+    /// and explain why `ret_dtype` has to be specified beforehand.
+    pub fn broadcasting_starmap<'a, G, MappingFn>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, 'a>,
+        inputs: &[Self],
+        ret_dtype: Type,
+        mapping: MappingFn,
+    ) -> Result<Self, String>
+    where
+        G: CodeGenerator + ?Sized,
+        MappingFn: FnOnce(
+            &mut G,
+            &mut CodeGenContext<'ctx, 'a>,
+            Int<'ctx, SizeT>,
+            &[ScalarObject<'ctx>],
+        ) -> Result<BasicValueEnum<'ctx>, String>,
+    {
+        let sizet_model = IntModel(SizeT);
+
+        // Check if all inputs are ScalarObjects
+        let all_scalars: Option<Vec<_>> =
+            inputs.iter().map(ScalarObject::try_from).try_collect().ok();
+
+        if let Some(scalars) = all_scalars {
+            let i = sizet_model.const_0(generator, ctx.ctx); // Pass 0 as the index
+            let scalar =
+                ScalarObject { value: mapping(generator, ctx, i, &scalars)?, dtype: ret_dtype };
+            Ok(ScalarOrNDArray::Scalar(scalar))
+        } else {
+            // Promote all input to ndarrays and map through them.
+            let inputs = inputs.iter().map(|input| input.as_ndarray(generator, ctx)).collect_vec();
+            let ndarray =
+                NDArrayObject::broadcasting_starmap(generator, ctx, &inputs, ret_dtype, mapping)?;
+            Ok(ScalarOrNDArray::NDArray(ndarray))
+        }
+    }
+
+    pub fn map<'a, G, Mapping>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, 'a>,
+        ret_dtype: Type,
+        mapping: Mapping,
+    ) -> Result<Self, String>
+    where
+        G: CodeGenerator + ?Sized,
+        Mapping: FnOnce(
+            &mut G,
+            &mut CodeGenContext<'ctx, 'a>,
+            Int<'ctx, SizeT>,
+            ScalarObject<'ctx>,
+        ) -> Result<BasicValueEnum<'ctx>, String>,
+    {
+        ScalarOrNDArray::broadcasting_starmap(
+            generator,
+            ctx,
+            &[*self],
+            ret_dtype,
+            |generator, ctx, i, scalars| mapping(generator, ctx, i, scalars[0]),
+        )
+    }
+}

nac3core/src/codegen/structure/ndarray/mod.rs

@@ -0,0 +1,543 @@
+pub mod broadcast;
+pub mod functions;
+pub mod indexing;
+pub mod mapping;
+pub mod product;
+pub mod scalar;
+pub mod shape_util;
+
+use crate::{
+    codegen::{
+        irrt::{
+            call_nac3_ndarray_copy_data, call_nac3_ndarray_get_nth_pelement,
+            call_nac3_ndarray_is_c_contiguous, call_nac3_ndarray_len, call_nac3_ndarray_nbytes,
+            call_nac3_ndarray_resolve_and_check_new_shape, call_nac3_ndarray_set_strides_by_shape,
+            call_nac3_ndarray_size, call_nac3_ndarray_transpose,
+        },
+        model::*,
+        stmt::BreakContinueHooks,
+        CodeGenContext, CodeGenerator,
+    },
+    toplevel::numpy::{extract_ndims, unpack_ndarray_var_tys},
+    typecheck::typedef::Type,
+};
+use inkwell::{
+    context::Context,
+    types::BasicType,
+    values::{BasicValue, BasicValueEnum, PointerValue},
+    AddressSpace, IntPredicate,
+};
+use scalar::ScalarObject;
+use util::{call_memcpy_model, gen_for_model_auto};
+
+pub struct NpArrayFields<'ctx, F: FieldTraversal<'ctx>> {
+    pub data: F::Out<PtrModel<IntModel<Byte>>>,
+    pub itemsize: F::Out<IntModel<SizeT>>,
+    pub ndims: F::Out<IntModel<SizeT>>,
+    pub shape: F::Out<PtrModel<IntModel<SizeT>>>,
+    pub strides: F::Out<PtrModel<IntModel<SizeT>>>,
+}
+
+// TODO: Rename to `NDArray` when the old NDArray is removed.
+#[derive(Debug, Clone, Copy, Default)]
+pub struct NpArray;
+
+impl<'ctx> StructKind<'ctx> for NpArray {
+    type Fields<F: FieldTraversal<'ctx>> = NpArrayFields<'ctx, F>;
+
+    fn traverse_fields<F: FieldTraversal<'ctx>>(&self, traversal: &mut F) -> Self::Fields<F> {
+        Self::Fields {
+            data: traversal.add_auto("data"),
+            itemsize: traversal.add_auto("itemsize"),
+            ndims: traversal.add_auto("ndims"),
+            shape: traversal.add_auto("shape"),
+            strides: traversal.add_auto("strides"),
+        }
+    }
+}
+
+/// A NAC3 Python ndarray object.
+#[derive(Debug, Clone, Copy)]
+pub struct NDArrayObject<'ctx> {
+    pub dtype: Type,
+    pub ndims: u64,
+    pub value: Ptr<'ctx, StructModel<NpArray>>,
+}
+
+impl<'ctx> NDArrayObject<'ctx> {
+    /// Create an [`NDArrayObject`] from an LLVM value and its typechecker [`Type`].
+    pub fn from_value_and_type<V: BasicValue<'ctx>, G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        value: V,
+        ty: Type,
+    ) -> Self {
+        let pndarray_model = PtrModel(StructModel(NpArray));
+
+        let (dtype, ndims) = unpack_ndarray_var_tys(&mut ctx.unifier, ty);
+        let ndims = extract_ndims(&ctx.unifier, ndims);
+        let value = pndarray_model.check_value(generator, ctx.ctx, value).unwrap();
+        NDArrayObject { dtype, ndims, value }
+    }
+
+    /// Get the `np.size()` of this ndarray.
+    pub fn size<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Int<'ctx, SizeT> {
+        call_nac3_ndarray_size(generator, ctx, self.value)
+    }
+
+    /// Get the `ndarray.nbytes` of this ndarray.
+    pub fn nbytes<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Int<'ctx, SizeT> {
+        call_nac3_ndarray_nbytes(generator, ctx, self.value)
+    }
+
+    /// Get the `len()` of this ndarray.
+    pub fn len<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Int<'ctx, SizeT> {
+        call_nac3_ndarray_len(generator, ctx, self.value)
+    }
+
+    /// Check if this ndarray is C-contiguous.
+    ///
+    /// See NumPy's `flags["C_CONTIGUOUS"]`: <https://numpy.org/doc/stable/reference/generated/numpy.ndarray.flags.html#numpy.ndarray.flags>
+    pub fn is_c_contiguous<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Int<'ctx, Bool> {
+        call_nac3_ndarray_is_c_contiguous(generator, ctx, self.value)
+    }
+
+    /// Get the pointer to the n-th (0-based) element.
+    ///
+    /// The returned pointer has the element type of the LLVM type of this ndarray's `dtype`.
+    pub fn get_nth_pointer<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        nth: Int<'ctx, SizeT>,
+        name: &str,
+    ) -> PointerValue<'ctx> {
+        let elem_ty = ctx.get_llvm_type(generator, self.dtype);
+
+        let p = call_nac3_ndarray_get_nth_pelement(generator, ctx, self.value, nth);
+        ctx.builder
+            .build_pointer_cast(p.value, elem_ty.ptr_type(AddressSpace::default()), name)
+            .unwrap()
+    }
+
+    /// Get the n-th (0-based) scalar.
+    pub fn get_nth<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        nth: Int<'ctx, SizeT>,
+    ) -> ScalarObject<'ctx> {
+        let p = self.get_nth_pointer(generator, ctx, nth, "value");
+        let value = ctx.builder.build_load(p, "value").unwrap();
+        ScalarObject { dtype: self.dtype, value }
+    }
+
+    /// Call [`call_nac3_ndarray_set_strides_by_shape`] on this ndarray to update `strides`.
+    ///
+    /// Please refer to the IRRT implementation to see its purpose.
+    pub fn update_strides_by_shape<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) {
+        call_nac3_ndarray_set_strides_by_shape(generator, ctx, self.value);
+    }
+
+    /// Copy data from another ndarray.
+    ///
+    /// This ndarray and `src` is that their `np.size()` should be the same. Their shapes
+    /// do not matter. The copying order is determined by how their flattened views look.
+    ///
+    /// Panics if the `dtype`s of ndarrays are different.
+    pub fn copy_data_from<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        src: NDArrayObject<'ctx>,
+    ) {
+        assert!(ctx.unifier.unioned(self.dtype, src.dtype), "self and src dtype should match");
+        call_nac3_ndarray_copy_data(generator, ctx, src.value, self.value);
+    }
+
+    /// Allocate an ndarray on the stack given its `ndims` and `dtype`.
+    ///
+    /// `shape` and `strides` will be automatically allocated on the stack.
+    ///
+    /// The returned ndarray's content will be:
+    /// - `data`: set to `nullptr`.
+    /// - `itemsize`: set to the `sizeof()` of `dtype`.
+    /// - `ndims`: set to the value of  `ndims`.
+    /// - `shape`: allocated with an array of length `ndims` with uninitialized values.
+    /// - `strides`: allocated with an array of length `ndims` with uninitialized values.
+    pub fn alloca_uninitialized<G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        dtype: Type,
+        ndims: u64,
+        name: &str,
+    ) -> Self {
+        let sizet_model = IntModel(SizeT);
+        let ndarray_model = StructModel(NpArray);
+        let ndarray_data_model = PtrModel(IntModel(Byte));
+
+        let pndarray = ndarray_model.alloca(generator, ctx, name);
+
+        let data = ndarray_data_model.nullptr(generator, ctx.ctx);
+        pndarray.set(ctx, |f| f.data, data);
+
+        let itemsize = ctx.get_llvm_type(generator, dtype).size_of().unwrap();
+        let itemsize =
+            sizet_model.s_extend_or_bit_cast(generator, ctx, itemsize, "alloca_itemsize");
+        pndarray.set(ctx, |f| f.itemsize, itemsize);
+
+        let ndims_val = sizet_model.constant(generator, ctx.ctx, ndims);
+        pndarray.set(ctx, |f| f.ndims, ndims_val);
+
+        let shape = sizet_model.array_alloca(generator, ctx, ndims_val.value, "alloca_shape");
+        pndarray.set(ctx, |f| f.shape, shape);
+
+        let strides = sizet_model.array_alloca(generator, ctx, ndims_val.value, "alloca_strides");
+        pndarray.set(ctx, |f| f.strides, strides);
+
+        NDArrayObject { dtype, ndims, value: pndarray }
+    }
+
+    /// Convenience function.
+    /// Like [`NDArrayObject::alloca_uninitialized`] but directly takes the typechecker type of the ndarray.
+    pub fn alloca_uninitialized_of_type<G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        ndarray_ty: Type,
+        name: &str,
+    ) -> Self {
+        let (dtype, ndims) = unpack_ndarray_var_tys(&mut ctx.unifier, ndarray_ty);
+        let ndims = extract_ndims(&ctx.unifier, ndims);
+        Self::alloca_uninitialized(generator, ctx, dtype, ndims, name)
+    }
+
+    /// Clone this ndaarray - Allocate a new ndarray with the same shape as this ndarray and copy the contents
+    /// over.
+    ///
+    /// The new ndarray will own its data and will be C-contiguous.
+    pub fn make_clone<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        name: &str,
+    ) -> Self {
+        let clone =
+            NDArrayObject::alloca_uninitialized(generator, ctx, self.dtype, self.ndims, name);
+
+        let shape = self.value.gep(ctx, |f| f.shape).load(generator, ctx, "shape");
+        clone.copy_shape_from_array(generator, ctx, shape);
+        clone.create_data(generator, ctx);
+        clone.copy_data_from(generator, ctx, *self);
+        clone
+    }
+
+    /// Get this ndarray's `ndims` as an LLVM constant.
+    pub fn get_ndims<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &'ctx Context,
+    ) -> Int<'ctx, SizeT> {
+        let sizet_model = IntModel(SizeT);
+        sizet_model.constant(generator, ctx, self.ndims)
+    }
+
+    /// Get if this ndarray's `np.size` is `0` - containing no content.
+    pub fn is_empty<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Int<'ctx, Bool> {
+        let sizet_model = IntModel(SizeT);
+
+        let size = self.size(generator, ctx);
+        size.compare(ctx, IntPredicate::EQ, sizet_model.const_0(generator, ctx.ctx), "is_empty")
+    }
+
+    /// Return true if this ndarray is unsized - `ndims == 0` and only contains a scalar.
+    ///
+    /// This is a staticially known property of ndarrays. This is why it is returning
+    /// a Rust boolean instead of a [`BasicValue`].
+    #[must_use]
+    pub fn is_unsized(&self) -> bool {
+        self.ndims == 0
+    }
+
+    /// Initialize an ndarray's `data` by allocating a buffer on the stack.
+    /// The allocated data buffer is considered to be *owned* by the ndarray.
+    ///
+    /// `strides` of the ndarray will also be updated with `set_strides_by_shape`.
+    ///
+    /// `shape` and `itemsize` of the ndarray ***must*** be initialized first.
+    pub fn create_data<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) {
+        let byte_model = IntModel(Byte);
+
+        let nbytes = self.nbytes(generator, ctx);
+
+        let data = byte_model.array_alloca(generator, ctx, nbytes.value, "data");
+        self.value.set(ctx, |f| f.data, data);
+
+        self.update_strides_by_shape(generator, ctx);
+    }
+
+    /// Copy shape dimensions from an array.
+    pub fn copy_shape_from_array<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        src_shape: Ptr<'ctx, IntModel<SizeT>>,
+    ) {
+        let dst_shape = self.value.get(generator, ctx, |f| f.shape, "dst_shape");
+        let num_items = self.get_ndims(generator, ctx.ctx).value;
+        call_memcpy_model(generator, ctx, dst_shape, src_shape, num_items);
+    }
+
+    /// Copy shape dimensions from an ndarray.
+    /// Panics if `ndims` mismatches.
+    pub fn copy_shape_from_ndarray<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        src_ndarray: NDArrayObject<'ctx>,
+    ) {
+        assert_eq!(self.ndims, src_ndarray.ndims);
+        let src_shape = src_ndarray.value.get(generator, ctx, |f| f.shape, "src_shape");
+        self.copy_shape_from_array(generator, ctx, src_shape);
+    }
+
+    /// Copy strides dimensions from an array.
+    pub fn copy_strides_from_array<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        src_strides: Ptr<'ctx, IntModel<SizeT>>,
+    ) {
+        let dst_strides = self.value.get(generator, ctx, |f| f.strides, "dst_strides");
+        let num_items = self.get_ndims(generator, ctx.ctx).value;
+        call_memcpy_model(generator, ctx, dst_strides, src_strides, num_items);
+    }
+
+    /// Copy strides dimensions from an ndarray.
+    /// Panics if `ndims` mismatches.
+    pub fn copy_strides_from_ndarray<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        src_ndarray: NDArrayObject<'ctx>,
+    ) {
+        assert_eq!(self.ndims, src_ndarray.ndims);
+        let src_strides = src_ndarray.value.get(generator, ctx, |f| f.strides, "src_strides");
+        self.copy_strides_from_array(generator, ctx, src_strides);
+    }
+
+    /// Iterate through every element pointer in the ndarray in its flatten view.
+    ///
+    /// `body` also access to [`BreakContinueHooks`] to short-circuit and an element's
+    /// index. The given element pointer also has been casted to the LLVM type of this ndarray's `dtype`.
+    pub fn foreach_pointer<'a, G, F>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, 'a>,
+        body: F,
+    ) -> Result<(), String>
+    where
+        G: CodeGenerator + ?Sized,
+        F: FnOnce(
+            &mut G,
+            &mut CodeGenContext<'ctx, 'a>,
+            BreakContinueHooks<'ctx>,
+            Int<'ctx, SizeT>,
+            PointerValue<'ctx>,
+        ) -> Result<(), String>,
+    {
+        let sizet_model = IntModel(SizeT);
+
+        let start = sizet_model.const_0(generator, ctx.ctx);
+        let stop = self.size(generator, ctx);
+        let step = sizet_model.const_1(generator, ctx.ctx);
+
+        gen_for_model_auto(generator, ctx, start, stop, step, |generator, ctx, hooks, i| {
+            let pelement = self.get_nth_pointer(generator, ctx, i, "element");
+            body(generator, ctx, hooks, i, pelement)
+        })
+    }
+
+    /// Iterate through every scalar in this ndarray.
+    pub fn foreach<'a, G, F>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, 'a>,
+        body: F,
+    ) -> Result<(), String>
+    where
+        G: CodeGenerator + ?Sized,
+        F: FnOnce(
+            &mut G,
+            &mut CodeGenContext<'ctx, 'a>,
+            BreakContinueHooks<'ctx>,
+            Int<'ctx, SizeT>,
+            ScalarObject<'ctx>,
+        ) -> Result<(), String>,
+    {
+        self.foreach_pointer(generator, ctx, |generator, ctx, hooks, i, p| {
+            let value = ctx.builder.build_load(p, "value").unwrap();
+            let scalar = ScalarObject { dtype: self.dtype, value };
+            body(generator, ctx, hooks, i, scalar)
+        })
+    }
+
+    /// Fill the ndarray with a value.
+    ///
+    /// `fill_value` must have the same LLVM type as the `dtype` of this ndarray.
+    pub fn fill<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        fill_value: BasicValueEnum<'ctx>,
+    ) {
+        self.foreach_pointer(generator, ctx, |_generator, ctx, _hooks, _i, pelement| {
+            ctx.builder.build_store(pelement, fill_value).unwrap();
+            Ok(())
+        })
+        .unwrap();
+    }
+
+    /// Create a reshaped view on this ndarray like `np.reshape()`.
+    ///
+    /// If there is a `-1` in `new_shape`, it will be resolved; `new_shape` would **NOT** be modified as a result.
+    ///
+    /// If reshape without copying is impossible, this function will allocate a new ndarray and copy contents.
+    ///
+    /// * `new_ndims` - The number of dimensions of `new_shape` as a [`Type`].
+    /// * `new_shape` - The target shape to do `np.reshape()`.
+    #[must_use]
+    pub fn reshape_or_copy<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        new_ndims: u64,
+        new_shape: Ptr<'ctx, IntModel<SizeT>>,
+    ) -> Self {
+        // TODO: The current criterion for whether to do a full copy or not is by checking `is_c_contiguous`,
+        // but this is not optimal. Look into how numpy does it.
+
+        let current_bb = ctx.builder.get_insert_block().unwrap();
+        let then_bb = ctx.ctx.insert_basic_block_after(current_bb, "then_bb");
+        let else_bb = ctx.ctx.insert_basic_block_after(then_bb, "else_bb");
+        let end_bb = ctx.ctx.insert_basic_block_after(else_bb, "end_bb");
+
+        let dst_ndarray = NDArrayObject::alloca_uninitialized(
+            generator,
+            ctx,
+            self.dtype,
+            new_ndims,
+            "reshaped_ndarray",
+        );
+        dst_ndarray.copy_shape_from_array(generator, ctx, new_shape);
+
+        let size = self.size(generator, ctx);
+        let new_ndims = dst_ndarray.get_ndims(generator, ctx.ctx);
+        call_nac3_ndarray_resolve_and_check_new_shape(generator, ctx, size, new_ndims, new_shape);
+
+        let is_c_contiguous = self.is_c_contiguous(generator, ctx);
+        ctx.builder.build_conditional_branch(is_c_contiguous.value, then_bb, else_bb).unwrap();
+
+        // Inserting into then_bb: reshape is possible without copying
+        ctx.builder.position_at_end(then_bb);
+        dst_ndarray.update_strides_by_shape(generator, ctx);
+        dst_ndarray.value.set(ctx, |f| f.data, self.value.get(generator, ctx, |f| f.data, "data"));
+        ctx.builder.build_unconditional_branch(end_bb).unwrap();
+
+        // Inserting into else_bb: reshape is impossible without copying
+        ctx.builder.position_at_end(else_bb);
+        dst_ndarray.create_data(generator, ctx);
+        dst_ndarray.copy_data_from(generator, ctx, *self);
+        ctx.builder.build_unconditional_branch(end_bb).unwrap();
+
+        // Reposition for continuation
+        ctx.builder.position_at_end(end_bb);
+
+        dst_ndarray
+    }
+
+    /// Create a flattened view of this ndarray, like `np.ravel()`.
+    ///
+    /// Uses [`NDArrayObject::reshape_or_copy`] under-the-hood so ndarray may or may not be copied.
+    pub fn ravel_or_copy<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> Self {
+        // Define models
+        let sizet_model = IntModel(SizeT);
+
+        let num0 = sizet_model.const_0(generator, ctx.ctx);
+        let num1 = sizet_model.const_1(generator, ctx.ctx);
+        let num_neg1 = sizet_model.const_all_1s(generator, ctx.ctx);
+
+        // Create `[-1]` and pass to `reshape_or_copy`.
+        let new_shape = sizet_model.array_alloca(generator, ctx, num1.value, "new_shape");
+        new_shape.offset(generator, ctx, num0.value, "").store(ctx, num_neg1);
+
+        self.reshape_or_copy(generator, ctx, 1, new_shape)
+    }
+
+    /// Create a transposed view on this ndarray like `np.transpose(<ndarray>, <axes> = None)`.
+    /// * `axes` - If specified, should be an array of the permutation (negative indices are **allowed**).
+    pub fn transpose<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        axes: Option<Ptr<'ctx, IntModel<SizeT>>>,
+    ) -> Self {
+        // Define models
+        let sizet_model = IntModel(SizeT);
+
+        let transposed_ndarray = NDArrayObject::alloca_uninitialized(
+            generator,
+            ctx,
+            self.dtype,
+            self.ndims,
+            "transposed_ndarray",
+        );
+
+        let num_axes = self.get_ndims(generator, ctx.ctx);
+
+        // `axes = nullptr` if `axes` is unspecified.
+        let axes = axes.unwrap_or_else(|| PtrModel(sizet_model).nullptr(generator, ctx.ctx));
+
+        call_nac3_ndarray_transpose(
+            generator,
+            ctx,
+            self.value,
+            transposed_ndarray.value,
+            num_axes,
+            axes,
+        );
+
+        transposed_ndarray
+    }
+}

nac3core/src/codegen/structure/ndarray/product.rs

@@ -0,0 +1,18 @@
+use crate::codegen::{CodeGenContext, CodeGenerator};
+
+use super::NDArrayObject;
+
+impl<'ctx> NDArrayObject<'ctx> {
+    /// TODO: Document me
+    pub fn matmul_at_least_2d<G: CodeGenerator + ?Sized>(
+        generator: &G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        a: Self,
+        b: Self,
+    ) -> Self {
+        assert!(a.ndims >= 2);
+        assert!(b.ndims >= 2);
+
+        todo!()
+    }
+}

nac3core/src/codegen/structure/ndarray/scalar.rs

@@ -0,0 +1,143 @@
+use inkwell::values::{BasicValue, BasicValueEnum};
+
+use crate::{
+    codegen::{model::*, CodeGenContext, CodeGenerator},
+    typecheck::typedef::{Type, TypeEnum},
+};
+
+use super::NDArrayObject;
+
+/// An LLVM numpy scalar with its [`Type`].
+///
+/// Intended to be used with [`ScalarOrNDArray`].
+///
+/// A scalar does not have to be an actual number. It could be arbitrary objects.
+#[derive(Debug, Clone, Copy)]
+pub struct ScalarObject<'ctx> {
+    pub dtype: Type,
+    pub value: BasicValueEnum<'ctx>,
+}
+
+impl<'ctx> ScalarObject<'ctx> {
+    /// Promote this scalar to an unsized ndarray (like doing `np.asarray`).
+    ///
+    /// The scalar value is allocated onto the stack, and the ndarray's `data` will point to that
+    /// allocated value.
+    pub fn as_ndarray<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> NDArrayObject<'ctx> {
+        let pbyte_model = PtrModel(IntModel(Byte));
+
+        // We have to put the value on the stack to get a data pointer.
+        let data = ctx.builder.build_alloca(self.value.get_type(), "as_ndarray_scalar").unwrap();
+        ctx.builder.build_store(data, self.value).unwrap();
+        let data = pbyte_model.pointer_cast(generator, ctx, data, "data");
+
+        let ndarray =
+            NDArrayObject::alloca_uninitialized(generator, ctx, self.dtype, 0, "scalar_ndarray");
+        ndarray.value.set(ctx, |f| f.data, data);
+        ndarray
+    }
+}
+
+/// A convenience enum for implementing scalar/ndarray agnostic utilities.
+#[derive(Debug, Clone, Copy)]
+pub enum ScalarOrNDArray<'ctx> {
+    Scalar(ScalarObject<'ctx>),
+    NDArray(NDArrayObject<'ctx>),
+}
+
+impl<'ctx> ScalarOrNDArray<'ctx> {
+    /// Get the underlying [`BasicValueEnum<'ctx>`] of this [`ScalarOrNDArray`].
+    #[must_use]
+    pub fn to_basic_value_enum(self) -> BasicValueEnum<'ctx> {
+        match self {
+            ScalarOrNDArray::Scalar(scalar) => scalar.value,
+            ScalarOrNDArray::NDArray(ndarray) => ndarray.value.value.as_basic_value_enum(),
+        }
+    }
+
+    #[must_use]
+    pub fn into_scalar(&self) -> ScalarObject<'ctx> {
+        match self {
+            ScalarOrNDArray::NDArray(_ndarray) => panic!("Got NDArray"),
+            ScalarOrNDArray::Scalar(scalar) => *scalar,
+        }
+    }
+
+    #[must_use]
+    pub fn into_ndarray(&self) -> NDArrayObject<'ctx> {
+        match self {
+            ScalarOrNDArray::NDArray(ndarray) => *ndarray,
+            ScalarOrNDArray::Scalar(_scalar) => panic!("Got Scalar"),
+        }
+    }
+
+    /// Convert this [`ScalarOrNDArray`] to an ndarray - behaves like `np.asarray`.
+    /// - If this is an ndarray, the ndarray is returned.
+    /// - If this is a scalar, an unsized ndarray view is created on it.
+    pub fn as_ndarray<G: CodeGenerator + ?Sized>(
+        &self,
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+    ) -> NDArrayObject<'ctx> {
+        match self {
+            ScalarOrNDArray::NDArray(ndarray) => *ndarray,
+            ScalarOrNDArray::Scalar(scalar) => scalar.as_ndarray(generator, ctx),
+        }
+    }
+
+    #[must_use]
+    pub fn dtype(&self) -> Type {
+        match self {
+            ScalarOrNDArray::Scalar(scalar) => scalar.dtype,
+            ScalarOrNDArray::NDArray(ndarray) => ndarray.dtype,
+        }
+    }
+}
+
+impl<'ctx> TryFrom<&ScalarOrNDArray<'ctx>> for ScalarObject<'ctx> {
+    type Error = ();
+
+    fn try_from(value: &ScalarOrNDArray<'ctx>) -> Result<Self, Self::Error> {
+        match value {
+            ScalarOrNDArray::Scalar(scalar) => Ok(*scalar),
+            ScalarOrNDArray::NDArray(_ndarray) => Err(()),
+        }
+    }
+}
+
+impl<'ctx> TryFrom<&ScalarOrNDArray<'ctx>> for NDArrayObject<'ctx> {
+    type Error = ();
+
+    fn try_from(value: &ScalarOrNDArray<'ctx>) -> Result<Self, Self::Error> {
+        match value {
+            ScalarOrNDArray::Scalar(_scalar) => Err(()),
+            ScalarOrNDArray::NDArray(ndarray) => Ok(*ndarray),
+        }
+    }
+}
+
+/// Split an [`BasicValueEnum<'ctx>`] into a [`ScalarOrNDArray`] depending
+/// on its [`Type`].
+pub fn split_scalar_or_ndarray<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    input: BasicValueEnum<'ctx>,
+    input_ty: Type,
+) -> ScalarOrNDArray<'ctx> {
+    match &*ctx.unifier.get_ty(input_ty) {
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.ndarray.obj_id(&ctx.unifier).unwrap() =>
+        {
+            let ndarray = NDArrayObject::from_value_and_type(generator, ctx, input, input_ty);
+            ScalarOrNDArray::NDArray(ndarray)
+        }
+        _ => {
+            let scalar = ScalarObject { dtype: input_ty, value: input };
+            ScalarOrNDArray::Scalar(scalar)
+        }
+    }
+}

nac3core/src/codegen/structure/ndarray/shape_util.rs

@@ -0,0 +1,112 @@
+use inkwell::values::BasicValueEnum;
+use util::gen_for_model_auto;
+
+use crate::{
+    codegen::{model::*, structure::list::ListObject, CodeGenContext, CodeGenerator},
+    typecheck::typedef::{Type, TypeEnum},
+};
+
+/// Parse a NumPy-like "int sequence" input and return the int sequence as an array and its length.
+///
+/// * `sequence` - The `sequence` parameter.
+/// * `sequence_ty` - The typechecker type of `sequence`
+///
+/// The `sequence` argument type may only be one of the following:
+///   1. A list of `int32`;   e.g., `np.empty([600, 800, 3])`
+///   2. A tuple of `int32`;  e.g., `np.empty((600, 800, 3))`
+///   3. A scalar `int32`;    e.g., `np.empty(3)`, this is functionally equivalent to `np.empty([3])`
+///
+/// All `int32` values will be sign-extended to `SizeT`.
+pub fn parse_numpy_int_sequence<'ctx, G: CodeGenerator + ?Sized>(
+    generator: &mut G,
+    ctx: &mut CodeGenContext<'ctx, '_>,
+    input_sequence: BasicValueEnum<'ctx>,
+    input_sequence_ty: Type,
+) -> (Int<'ctx, SizeT>, Ptr<'ctx, IntModel<SizeT>>) {
+    let sizet_model = IntModel(SizeT);
+
+    let zero = sizet_model.const_0(generator, ctx.ctx);
+    let one = sizet_model.const_1(generator, ctx.ctx);
+
+    // The result `list` to return.
+    match &*ctx.unifier.get_ty(input_sequence_ty) {
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.list.obj_id(&ctx.unifier).unwrap() =>
+        {
+            // 1. A list of `int32`; e.g., `np.empty([600, 800, 3])`
+
+            // Check `input_sequence`
+            let input_sequence =
+                ListObject::from_value_and_type(generator, ctx, input_sequence, input_sequence_ty);
+
+            let len = input_sequence.value.gep(ctx, |f| f.len).load(generator, ctx, "len");
+            let result = sizet_model.array_alloca(generator, ctx, len.value, "int_sequence");
+
+            // Load all the `int32`s from the input_sequence, cast them to `SizeT`, and store them into `result`
+            gen_for_model_auto(generator, ctx, zero, len, one, |generator, ctx, _hooks, i| {
+                // Load the i-th int32 in the input sequence
+                let int = input_sequence
+                    .value
+                    .get(generator, ctx, |f| f.items, "int")
+                    .ix(generator, ctx, i.value, "int")
+                    .value
+                    .into_int_value();
+
+                // Cast to SizeT
+                let int = sizet_model.s_extend_or_bit_cast(generator, ctx, int, "int");
+
+                // Store
+                result.offset(generator, ctx, i.value, "int").store(ctx, int);
+
+                Ok(())
+            })
+            .unwrap();
+
+            (len, result)
+        }
+        TypeEnum::TTuple { ty: tuple_types, .. } => {
+            // 2. A tuple of ints; e.g., `np.empty((600, 800, 3))`
+            let input_sequence = input_sequence.into_struct_value(); // A tuple is a struct
+
+            let len_int = tuple_types.len();
+
+            let len = sizet_model.constant(generator, ctx.ctx, len_int as u64);
+            let result = sizet_model.array_alloca(generator, ctx, len.value, "int_sequence");
+
+            for i in 0..len_int {
+                // Get the i-th element off of the tuple and load it into `result`.
+                let int = ctx
+                    .builder
+                    .build_extract_value(input_sequence, i as u32, "int")
+                    .unwrap()
+                    .into_int_value();
+                let int = sizet_model.s_extend_or_bit_cast(generator, ctx, int, "int");
+
+                let offset = sizet_model.constant(generator, ctx.ctx, i as u64);
+                result.offset(generator, ctx, offset.value, "int").store(ctx, int);
+            }
+
+            (len, result)
+        }
+        TypeEnum::TObj { obj_id, .. }
+            if *obj_id == ctx.primitives.int32.obj_id(&ctx.unifier).unwrap() =>
+        {
+            // 3. A scalar int; e.g., `np.empty(3)`, this is functionally equivalent to `np.empty([3])`
+            let input_int = input_sequence.into_int_value();
+
+            let len = sizet_model.const_1(generator, ctx.ctx);
+            let result = sizet_model.array_alloca(generator, ctx, len.value, "int_sequence");
+
+            let int = sizet_model.s_extend_or_bit_cast(generator, ctx, input_int, "int");
+
+            // Storing into result[0]
+            result.store(ctx, int);
+
+            (len, result)
+        }
+        _ => panic!(
+            "encountered unknown sequence type: {}",
+            ctx.unifier.stringify(input_sequence_ty)
+        ),
+    }
+}

nac3core/src/codegen/structure/tuple.rs

@@ -0,0 +1,39 @@
+use inkwell::values::{BasicValueEnum, StructValue};
+use itertools::Itertools;
+
+use crate::{
+    codegen::{CodeGenContext, CodeGenerator},
+    typecheck::typedef::Type,
+};
+
+pub struct TupleObject<'ctx> {
+    pub tys: Vec<Type>,
+    pub value: StructValue<'ctx>,
+}
+
+impl<'ctx> TupleObject<'ctx> {
+    pub fn create<I, G: CodeGenerator + ?Sized>(
+        generator: &mut G,
+        ctx: &mut CodeGenContext<'ctx, '_>,
+        items: I,
+        name: &str,
+    ) -> Self
+    where
+        I: IntoIterator<Item = (BasicValueEnum<'ctx>, Type)>,
+    {
+        let (vals, tys): (Vec<_>, Vec<_>) = items.into_iter().unzip();
+
+        // let tuple_ty = ctx.unifier.add_ty(TypeEnum::TTuple { ty: tys });
+        let llvm_tys = tys.iter().map(|ty| ctx.get_llvm_type(generator, *ty)).collect_vec();
+        let llvm_tuple_ty = ctx.ctx.struct_type(&llvm_tys, false);
+        let pllvm_tuple = ctx.builder.build_alloca(llvm_tuple_ty, "tuple").unwrap();
+        for (i, val) in vals.into_iter().enumerate() {
+            // Store the dim value into the tuple
+            let pval = ctx.builder.build_struct_gep(pllvm_tuple, i as u32, "value").unwrap();
+            ctx.builder.build_store(pval, val).unwrap();
+        }
+
+        let value = ctx.builder.build_load(pllvm_tuple, name).unwrap().into_struct_value();
+        TupleObject { tys, value }
+    }
+}