commit d33023f4351ef7cdc5c6036c12b4156c3b321628
Author: Moritz Gmeiner <moritz@gmeiner.net>
Date:   Fri Aug 2 00:10:48 2024 +0200

    init commit

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..f527c17
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,6 @@
+# Dune artifacts
+_build/
+dune.lock
+
+# Local OPAM switch
+_opam/
diff --git a/.ocamlformat b/.ocamlformat
new file mode 100644
index 0000000..a32dae4
--- /dev/null
+++ b/.ocamlformat
@@ -0,0 +1,4 @@
+profile = default
+# profile = ocamlformat
+# profile = janestreet
+margin = 100
diff --git a/MlLox.opam b/MlLox.opam
new file mode 100644
index 0000000..b296c11
--- /dev/null
+++ b/MlLox.opam
@@ -0,0 +1,31 @@
+# This file is generated by dune, edit dune-project instead
+opam-version: "2.0"
+synopsis: "A short synopsis"
+description: "A longer description"
+maintainer: ["Maintainer Name"]
+authors: ["Author Name"]
+license: "LICENSE"
+tags: ["topics" "to describe" "your" "project"]
+homepage: "https://github.com/username/reponame"
+doc: "https://url/to/documentation"
+bug-reports: "https://github.com/username/reponame/issues"
+depends: [
+  "ocaml"
+  "dune" {>= "3.16"}
+  "odoc" {with-doc}
+]
+build: [
+  ["dune" "subst"] {dev}
+  [
+    "dune"
+    "build"
+    "-p"
+    name
+    "-j"
+    jobs
+    "@install"
+    "@runtest" {with-test}
+    "@doc" {with-doc}
+  ]
+]
+dev-repo: "git+https://github.com/username/reponame.git"
diff --git a/bin/dune b/bin/dune
new file mode 100644
index 0000000..cb59a81
--- /dev/null
+++ b/bin/dune
@@ -0,0 +1,4 @@
+(executable
+ (public_name MlLox)
+ (libraries Lox)
+ (name main))
diff --git a/bin/main.ml b/bin/main.ml
new file mode 100644
index 0000000..2dca889
--- /dev/null
+++ b/bin/main.ml
@@ -0,0 +1,16 @@
+let printUsage () =
+  print_endline "Usage: jlox [script]";
+  exit 64
+
+let () =
+  let argc = Array.length Sys.argv in
+  match argc - 1 with
+  | 0 -> Lox.runRepl ()
+  | 1 ->
+      let path = Sys.argv.(1) in
+      Printf.printf "Running script %s\n" path;
+      let ic = open_in path in
+      let source = In_channel.input_all ic in
+      let result = Lox.run source in
+      Result.iter_error Lox.Error.print_error result
+  | _ -> printUsage ()
diff --git a/dune-project b/dune-project
new file mode 100644
index 0000000..a0e4244
--- /dev/null
+++ b/dune-project
@@ -0,0 +1,26 @@
+(lang dune 3.16)
+
+(name MlLox)
+
+(generate_opam_files true)
+
+(source
+ (github username/reponame))
+
+(authors "Author Name")
+
+(maintainers "Maintainer Name")
+
+(license LICENSE)
+
+(documentation https://url/to/documentation)
+
+(package
+ (name MlLox)
+ (synopsis "A short synopsis")
+ (description "A longer description")
+ (depends ocaml dune)
+ (tags
+  (topics "to describe" your project)))
+
+; See the complete stanza docs at https://dune.readthedocs.io/en/stable/reference/dune-project/index.html
diff --git a/lib/dune b/lib/dune
new file mode 100644
index 0000000..52c8dad
--- /dev/null
+++ b/lib/dune
@@ -0,0 +1,4 @@
+(library
+ (name Lox)
+ (preprocess
+  (pps ppx_deriving.show)))
diff --git a/lib/error.ml b/lib/error.ml
new file mode 100644
index 0000000..d7a4471
--- /dev/null
+++ b/lib/error.ml
@@ -0,0 +1,26 @@
+type code_pos = { line : int; col : int }
+type lexer_error = { pos : code_pos; msg : string }
+
+module LexerError = struct
+  type t = lexer_error
+
+  let make (pos : code_pos) (msg : string) : lexer_error =
+    (* let pos = { line; col } in *)
+    { pos; msg }
+
+  let print (e : lexer_error) =
+    Printf.printf "LexerError at line %d, column %d: %s\n" e.pos.line e.pos.col e.msg
+end
+
+type lox_error = LexerError of lexer_error list
+
+let print_error (e : lox_error) =
+  match e with
+  | LexerError es ->
+      let num_errors = List.length es in
+      assert (num_errors != 0);
+      Printf.printf "found %d %s:\n" num_errors
+        (if num_errors = 1 then "LexerError" else "LexerErrors");
+      List.iter LexerError.print es
+
+let of_lexer_error e = Result.map_error (fun e -> LexerError e) e
diff --git a/lib/lexer.ml b/lib/lexer.ml
new file mode 100644
index 0000000..a15fb4a
--- /dev/null
+++ b/lib/lexer.ml
@@ -0,0 +1,177 @@
+open Error
+
+[@@@ocamlformat "disable"]
+  type token_type =
+  | LeftParen | RightParen | LeftBrace | RightBrace
+
+  | Plus | Minus | Star | Slash | Bang
+
+  | Dot | Comma | Semicolon | Equal
+
+  | EqualEqual | BangEqual | Greater | Less | GreaterEqual | LessEqual
+  
+  | Identifier of string
+  | String of string
+  | Number of float
+
+  | And | Class | Else | False | Fun | For | If | Nil | Or | Print | Return | Super | This | True
+  | Var | While
+
+  | Comment of string
+  
+  | Eof
+[@@deriving show { with_path = false }]
+[@@@ocamlformat "enable"]
+
+type token = { token_type : token_type; pos : code_pos }
+
+let show_token (token : token) =
+  let { line; col } = token.pos in
+  Printf.sprintf "<%s at %d:%d>" (show_token_type token.token_type) line col
+
+type lexer_result = (token list, lexer_error list) result
+
+type state = {
+  (* source code *)
+  source : string;
+  start_pos : int;
+  cur_pos : int;
+  (* store tokens and errors in reverse to make building the list more efficient *)
+  tokens_rev : token list;
+  errors_rev : lexer_error list;
+  (* position of current char in source *)
+  line : int;
+  col : int;
+}
+
+module State = struct
+  type t = state
+
+  let is_digit c = match c with '0' .. '9' -> true | _ -> false
+  let is_alpha c = match c with 'a' .. 'z' | 'A' .. 'Z' -> true | _ -> false
+  let is_alphanum c = is_digit c || is_alpha c
+  let is_identifier c = is_alphanum c || c = '_'
+  let is_at_end (state : state) : bool = state.cur_pos = String.length state.source
+
+  let get_lexeme (state : state) (first : int) (last : int) =
+    String.sub state.source first (last - first)
+
+  let advance (state : state) : char * state =
+    let c = state.source.[state.cur_pos] in
+    let state = { state with cur_pos = state.cur_pos + 1 } in
+    let state =
+      match c with
+      | '\t' -> { state with col = state.col + 4 }
+      | '\n' -> { state with line = state.line + 1; col = 0 }
+      | _ -> { state with col = state.col + 1 }
+    in
+    (c, state)
+
+  let peek (state : state) : char option =
+    if not (is_at_end state) then Some state.source.[state.cur_pos] else None
+
+  let advance_if (c : char) (state : state) : bool * state =
+    if peek state = Some c then (true, snd (advance state)) else (false, state)
+
+  let rec advance_until (c : char) (state : state) : bool * state =
+    if is_at_end state then (false, state)
+    else
+      let c', state = advance state in
+      if c' = c then (true, state) else advance_until c state
+
+  let rec advance_while (f : char -> bool) (state : state) : state =
+    match peek state with
+    | Some c when f c -> advance_while f (snd (advance state))
+    | _ -> state (* EOF or no match *)
+
+  let last_char (state : state) =
+    assert (state.cur_pos > 0);
+    state.source.[state.cur_pos - 1]
+
+  let append_token pos state token_type =
+    (* let pos = { line = state.line; col = state.col } in *)
+    { state with tokens_rev = { token_type; pos } :: state.tokens_rev }
+
+  let append_error pos state msg =
+    (* let pos = { line = state.line; col = state.col } in *)
+    { state with errors_rev = LexerError.make pos msg :: state.errors_rev }
+
+  let parse_number (state : state) =
+    let skip c state = snd @@ advance_if c state in
+    let code_pos = { line = state.line; col = state.col } in
+    let state =
+      state |> advance_while is_digit |> skip '.' |> advance_while is_digit |> skip 'e'
+      |> advance_while is_digit
+    in
+    let lexeme = get_lexeme state state.start_pos state.cur_pos in
+    let f = Float.of_string_opt lexeme in
+    match f with
+    | None -> append_error code_pos state (Printf.sprintf "Invalid float literal %s" lexeme)
+    | Some f -> append_token code_pos state (Number f)
+
+  let rec tokenize_rec (state : state) : state =
+    let pos = { line = state.line; col = state.col } in
+    let append_token = append_token pos in
+    let append_error = append_error pos in
+    if is_at_end state then append_token state Eof
+    else
+      let state = { state with start_pos = state.cur_pos } in
+      let c, state = advance state in
+      let state =
+        match c with
+        | '(' -> append_token state LeftParen
+        | ')' -> append_token state RightParen
+        | '{' -> append_token state LeftBrace
+        | '}' -> append_token state RightBrace
+        | ',' -> append_token state Comma
+        | ';' -> append_token state Semicolon
+        | '.' -> append_token state Dot
+        | '+' -> append_token state Plus
+        | '-' -> append_token state Minus
+        | '*' -> append_token state Star
+        | '!' ->
+            let b, state = advance_if '=' state in
+            append_token state (if b then BangEqual else Bang)
+        | '=' ->
+            let b, state = advance_if '=' state in
+            append_token state (if b then EqualEqual else Equal)
+        | '<' ->
+            let b, state = advance_if '=' state in
+            append_token state (if b then LessEqual else Less)
+        | '>' ->
+            let b, state = advance_if '=' state in
+            append_token state (if b then GreaterEqual else Greater)
+        | '/' ->
+            let found, state = advance_if '/' state in
+            if not found then append_token state Slash
+            else
+              let start_pos = state.cur_pos in
+              let _, state = advance_until '\n' state in
+              let lexeme = String.trim @@ get_lexeme state start_pos state.cur_pos in
+              append_token state (Comment lexeme)
+        | '"' ->
+            let found, state = advance_until '"' state in
+            if not found then append_error state "Unterminated string literal"
+            else
+              let lexeme = get_lexeme state (state.start_pos + 1) (state.cur_pos - 1) in
+              append_token state (String lexeme)
+        | '0' .. '9' -> parse_number state
+        | ' ' | '\t' | '\n' -> parse_number state
+        | c -> append_error state (String.escaped @@ Printf.sprintf "Unexpected character '%c'" c)
+      in
+      tokenize_rec state
+end
+
+let tokenize (source : string) : lexer_result =
+  print_endline "Scanning source";
+  print_endline "---";
+  print_endline source;
+  print_endline "---";
+  (* Ok [] *)
+  let state =
+    State.tokenize_rec
+      { source; start_pos = 0; cur_pos = 0; tokens_rev = []; errors_rev = []; line = 1; col = 0 }
+  in
+  (* reverse the reversed tokens/errors *)
+  if List.length state.errors_rev = 0 then Ok (List.rev state.tokens_rev)
+  else Error (List.rev state.errors_rev)
diff --git a/lib/lox.ml b/lib/lox.ml
new file mode 100644
index 0000000..5195f48
--- /dev/null
+++ b/lib/lox.ml
@@ -0,0 +1,26 @@
+let ( let* ) = Result.bind
+
+module Lexer = Lexer
+module Error = Error
+
+type token = Lexer.token
+type lox_error = Error.lox_error
+type lox_value = Nil
+
+let run (source : string) : (unit, lox_error) result =
+  let* tokens = Error.of_lexer_error (Lexer.tokenize source) in
+  let f token = Printf.printf "%s " (Lexer.show_token token) in
+  Printf.printf "Got %d tokens\n" (List.length tokens);
+  List.iter f tokens;
+  print_endline "";
+  Ok ()
+
+let runRepl () : unit =
+  try
+    while true do
+      print_string "> ";
+      let line = read_line () in
+      let result = run line in
+      Result.iter_error Error.print_error result
+    done
+  with End_of_file -> ()
diff --git a/lox/test.lox b/lox/test.lox
new file mode 100644
index 0000000..44ddf16
--- /dev/null
+++ b/lox/test.lox
@@ -0,0 +1,3 @@
+// test comment
+"string"
+{}(
diff --git a/test/dune b/test/dune
new file mode 100644
index 0000000..c3a90fb
--- /dev/null
+++ b/test/dune
@@ -0,0 +1,2 @@
+(test
+ (name test_MlLox))
diff --git a/test/test_MlLox.ml b/test/test_MlLox.ml
new file mode 100644
index 0000000..e69de29