Ported unicode conversions

ast.cpp

@@ -26,6 +26,7 @@ enum Ast_Kind: U32{
   AST_LENGTH_OF,
   AST_ALIGN_OF,
 
+  AST_BREAK,
   AST_COMPOUND,
   AST_TYPE,
   AST_VAR,
@@ -146,6 +147,7 @@ struct Ast_If: Ast{
 };
 
 struct Ast_Pass: Ast{};
+struct Ast_Break: Ast{};
 
 struct Ast_For: Ast{
   Ast_Expr *init;
@@ -357,6 +359,12 @@ ast_pass(Token *pos){
   return result;
 }
 
+function Ast_Break *
+ast_break(Token *pos){
+  AST_NEW(Break, BREAK, pos, AST_STMT);
+  return result;
+}
+
 function Ast_Return *
 ast_return(Token *pos, Ast_Expr *expr){
   AST_NEW(Return, RETURN, pos, AST_STMT);

ccodegen.cpp

@@ -415,6 +415,14 @@ gen_ast(Ast *ast){
       BREAK();
     }
 
+
+    CASE(BREAK, Break){
+      unused(node);
+      gen("break;");
+      BREAK();
+    }
+
+
     CASE(PASS, Pass){
       unused(node);
       gen("//pass");
@@ -695,6 +703,8 @@ typedef struct String{
   S64 len;
 }String;
 #define LIT(x) (String){.str=(U8 *)x, .len=sizeof(x)-1}
+#define assert(x) do{if(!(x))__debugbreak();}while(0)
+#define assert_msg(x,...) assert(x)
 
   )==");
 #endif

compiler.h

@@ -163,6 +163,7 @@ Intern_String keyword_true;
 Intern_String keyword_false;
 Intern_String keyword_for;
 Intern_String keyword_pass;
+Intern_String keyword_break;
 Intern_String keyword_elif;
 Intern_String keyword_assert;
 Intern_String keyword_sizeof;
@@ -210,6 +211,7 @@ lex_init(Allocator *token_string_arena, Allocator *map_allocator, Lexer *l){
   keyword_lengthof = l->intern("length_of"_s);
   keyword_alignof = l->intern("align_of"_s);
   keyword_true = l->intern("true"_s);
+  keyword_break = l->intern("break"_s);
   keyword_false = l->intern("false"_s);
   keyword_return = l->intern("return"_s);
   keyword_assert = l->intern("assert"_s);

parsing.cpp

@@ -220,6 +220,10 @@ parse_stmt_scope(Ast_Scope *scope_defined_outside = 0){
         scope->stmts.add(ast_return(token, expr));
       }
 
+      else if(token_match_keyword(keyword_break)){
+        scope->stmts.add(ast_break(token));
+      }
+
       else if(token_match_keyword(keyword_pass)){
         scope->stmts.add(ast_pass(token));
       }

programs/Windows.kl

@@ -27,7 +27,6 @@ MEM_COMMIT :: 0x00001000
 MEM_RESERVE :: 0x00002000
 MEM_RESET :: 0x00080000
 MEM_RESET_UNDO :: 0x1000000
-
 MEM_DECOMMIT :: 0x00004000
 MEM_RELEASE :: 0x00008000
 
@@ -35,14 +34,15 @@ PAGE_NOACCESS :: 1
 PAGE_READONLY :: 2
 PAGE_READWRITE :: 4
 PAGE_WRITECOPY :: 8
-PAGE_EXECUTE :: 0x10
+PAGE_EXECUTE :: 0x10; PAGE_EXECUTE_READ :: 0x20; PAGE_EXECUTE_READWRITE :: 0x40; PAGE_EXECUTE_WRITECOPY :: 0x80
-PAGE_EXECUTE_READ :: 0x20
-PAGE_EXECUTE_READWRITE :: 0x40
-PAGE_EXECUTE_WRITECOPY :: 0x80
-
 VirtualAlloc :: #foreign (lpAddress: LPVOID, dwSize: SIZE_T, flAllocationType: DWORD, flProtect: DWORD): LPVOID
 VirtualFree  :: #foreign (lpAddress: LPVOID, dwSize: SIZE_T, dwFreeType: DWORD): BOOL
 
+HEAP_ZERO_MEMORY :: 0x8; HEAP_NO_SERIALIZE :: 0x1; HEAP_GENERATE_EXCEPTIONS :: 0x4
+GetProcessHeap   :: #foreign (): HANDLE
+HeapAlloc        :: #foreign (hHeap: HANDLE, dwFlags: DWORD, dwByte: SIZE_T): LPVOID
+HeapFree         :: #foreign (hHeap: HANDLE, dwFlags: DWORD, lpMe: LPVOID): BOOL
+
 STD_INPUT_HANDLE  :: 4294967286//(-10)->DWORD
 STD_OUTPUT_HANDLE :: 4294967285//(-11)->DWORD
 //STD_ERROR_HANDLE  :: (-12)->DWORD

programs/main.kl

@@ -7,11 +7,13 @@ String32 :: struct;; str: *U32; len: S64
 UTF32_Result :: struct
   out_str: U32
   advance: S64
-  error  : S32
+  error  : S64
 UTF16_Result :: struct
   out_str: [2]U16
-  len    : S32
+  len    : S64
-  error  : S32
+  error  : S64
+question_mark16 :: 0x003f
+
 
 utf8_to_utf32 :: (c: *U8, max_advance: S64): UTF32_Result
   result: UTF32_Result
@@ -68,26 +70,34 @@ utf32_to_utf16 :: (codepoint: U32): UTF16_Result
 
   return result
 
-// string_to_string16 :: (allocator: *Allocator, int: String): String16
+process_heap: HANDLE
-//   String16 result = {exp_alloc_array(allocator, U16, (in.len*2)+1)}; // @Note(Krzosa): Should be more then enough space
+allocate :: (size: U64): *void
-//   for(S64 i = 0; i < in.len;){
+  if process_heap == 0
-//     UTF32_Result decode = utf8_to_utf32(in.str + i, in.len - i);
+    process_heap = GetProcessHeap()
-//     if(!decode.error){
+  return HeapAlloc(process_heap, 0, size)
-//       i += decode.advance;
-//       UTF16_Result encode = utf32_to_utf16(decode.out_str);
-//       if(!encode.error){
-//         for(S32 j = 0; j < encode.len; j++){
-//           result.str[result.len++] = encode.out_str[j];
-//         }
-//       }
-//       else unicode_error(question_mark16);
-//     }
-//     else unicode_error(question_mark16);
-//   }
 
-//   result.str[result.len] = 0;
+string_to_string16 :: (in: String): String16
-//   return result;
+  in_str := &in[0]
-// }
+  // @Note(Krzosa): Should be more then enough space
+  alloc_size := (length_of(in)*2)+1
+  result := String16{str = allocate(alloc_size->U64)->*U16}
+  for i := 0, i < length_of(in)
+    decode := utf8_to_utf32(in_str + i, length_of(in) - i)
+    if !decode.error
+      i += decode.advance
+      encode := utf32_to_utf16(decode.out_str)
+      if !encode.error
+        for j := 0, j < encode.len, j++
+          result.str[result.len++] = encode.out_str[j]
+      else
+        result.str[result.len++] = question_mark16
+        break
+    else
+      result.str[result.len++] = question_mark16
+      break
+
+  result.str[result.len] = 0
+  return result
 
 Vec2I :: struct;; x: S32; y: S32
 Vec2  :: struct;; x: F32; y: F32
@@ -129,8 +139,16 @@ window_procedure :: (hwnd: HWND, msg: UINT, wparam: WPARAM, lparam: LPARAM): LRE
 
 main :: (argc: int, argv: **char): int
   bitmap := create_bitmap({1280, 720})
+
+  string := "    豈 更 車 賈 滑 串 句 龜 龜 契 金 喇 奈 懶 癩 羅 蘿 螺 裸 邏 樂 洛 烙 珞 落 酪 駱 亂 卵 欄 爛 蘭 鸞 嵐 濫 藍 襤 拉 臘 蠟 廊 朗 浪 狼 郎 來 冷 勞 擄 櫓 爐 盧 老 蘆 虜 路 露 魯 鷺 碌 祿 綠 菉 錄 鹿 論 壟 弄 籠 聾 牢 磊 賂 雷 壘 屢 樓 淚 漏 累 縷 陋 勒 肋 凜 凌 稜 綾 菱 陵 讀 拏 樂 諾 丹 寧 怒 率 異 北 磻 便 復 不 泌 數 索 參 塞 省 葉 說 殺 辰 沈 拾 若 掠 略 亮 兩 凉 梁 糧 良 諒 量 勵 ..."
+  string_result := string_to_string16(string)
+
   result := utf8_to_utf32(&"A"[0], 1)
   assert(result.out_str == 'A, "Invalid decode")
-  result  = utf8_to_utf32(&"ć"[0], 2)
+
-  result  = utf8_to_utf32(&"ó"[0], 2)
+  result  = utf8_to_utf32(&"ć"[0], 2)
+  assert(result.out_str == 0x107, "Invalid decode")
+
+  result  = utf8_to_utf32(&"ó"[0], 2)
+  assert(result.out_str == 0xF3, "Invalid decode")
 

typechecking.cpp

@@ -438,6 +438,7 @@ resolve_stmt(Ast *ast, Ast_Type *ret){
       BREAK();
     }
 
+    case AST_BREAK:
     CASE(PASS, Pass){
       unused(node);
       BREAK();