Porting unicode code

ccodegen.cpp

@@ -262,7 +262,9 @@ gen_expr(Ast_Expr *ast, Ast_Type *type_of_var){
     CASE(INDEX, Index){
       gen("(");
       gen_expr(node->expr);
-      if(node->expr->resolved_type == type_string) gen(".str");
+      if(is_string(node->resolved_type))
+        if(!(type_of_var && type_of_var == type_pointer_to_char))
+          gen(".str");
       gen("[");
       gen_expr(node->index);
       gen("]");

programs/main.kl

@@ -7,21 +7,91 @@ String32 :: struct;; str: *U32; len: S64
 UTF32_Result :: struct
   out_str: U32
   advance: S64
-  error  : Bool
+  error  : S32
 UTF16_Result :: struct
   out_str: [2]U16
   len    : S32
-  error  : Bool
+  error  : S32
 
 BINARY :: 0b1001
 
 utf8_to_utf32 :: (c: *U8, max_advance: S64): UTF32_Result
   result: UTF32_Result
+  if (c[0] & 0b10000000) == 0
+    if max_advance >= 1
+      c0 := c[0]->U32
+      result.out_str = c0
+      result.advance = 1
+    else;; result.error = 1
+
+  elif (c[0] & 0b11100000) == 0b11000000
+    if (c[1] & 0b11000000) == 0b10000000 // Continuation byte required
+      if max_advance >= 2
+        c0 := c[0]->U32; c1 := c[1]->U32
+        result.out_str = (c0 & 0b00011111) << 6 | (c1 & 0b00111111)
+        result.advance = 2
+      else;; result.error = 2
+    else;; result.error = 2
+
+  elif (c[0] & 0b11110000) == 0b11100000
+    if (c[1] & 0b11000000) == 0b10000000 && (c[2] & 0b11000000) == 0b10000000 // Two continuation bytes required
+      if max_advance >= 3
+        c0 := c[0]->U32; c1 := c[1]->U32; c2 := c[2]->U32
+        result.out_str = (c0 & 0b00001111) << 12 | (c1 & 0b00111111) << 6 | (c2 & 0b00111111)
+        result.advance = 3
+      else;; result.error = 3
+    else;; result.error = 3
+
+  elif (c[0] & 0b11111000) == 0b11110000
+    if (c[1] & 0b11000000) == 0b10000000 && (c[2] & 0b11000000) == 0b10000000 && (c[3] & 0b11000000) == 0b10000000 // Three continuation bytes required
+      if max_advance >= 4
+        c0 := c[0]->U32; c1 := c[1]->U32; c2 := c[2]->U32; c3 := c[3]->U32
+        result.out_str = (c0 & 0b00001111) << 18 | (c1 & 0b00111111) << 12 | (c2 & 0b00111111) << 6 | (c3 & 0b00111111)
+        result.advance = 4
+      else;; result.error = 4
+    else;; result.error = 4
+  else;; result.error = 4
+
   return result
 
+utf32_to_utf16 :: (codepoint: U32): UTF16_Result
+  result: UTF16_Result
+  if codepoint < 0x10000
+    result.out_str[0] = codepoint->U16
+    result.out_str[1] = 0
+    result.len = 1
+  elif codepoint <= 0x10FFFF
+    code: U32 = (codepoint - 0x10000)
+    result.out_str[0] = (0xD800 | (code >> 10))->U16
+    result.out_str[1] = (0xDC00 | (code & 0x3FF))->U16
+    result.len = 2
+  else
+    result.error = 1
+
+  return result
+
+// string_to_string16 :: (allocator: *Allocator, int: String): String16
+//   String16 result = {exp_alloc_array(allocator, U16, (in.len*2)+1)}; // @Note(Krzosa): Should be more then enough space
+//   for(S64 i = 0; i < in.len;){
+//     UTF32_Result decode = utf8_to_utf32(in.str + i, in.len - i);
+//     if(!decode.error){
+//       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;
+//   return result;
+// }
+
 
-Vec2I :: struct;; x: S32; y: S32
-Vec2  :: struct;; x: F32; y: F32
 
 Windows_Bitmap :: struct
   size: Vec2I
@@ -29,6 +99,10 @@ Windows_Bitmap :: struct
   hdc: HDC
   dib: HBITMAP
 
+Vec2I :: struct;; x: S32; y: S32
+Vec2  :: struct;; x: F32; y: F32
+
+
 create_bitmap :: (size: Vec2I, bottom_up: Bool = true): Windows_Bitmap
   result: Windows_Bitmap = {size = size}
   if bottom_up == false
@@ -57,9 +131,11 @@ window_procedure :: (hwnd: HWND, msg: UINT, wparam: WPARAM, lparam: LPARAM): LRE
   if msg == WM_DESTROY
     PostQuitMessage(0)
     return 0
-  else;; return DefWindowProc(hwnd, msg, wparam, lparam)
+  else;; return DefWindowProcW(hwnd, msg, wparam, lparam)
 
 main :: (argc: int, argv: **char): int
   bitmap := create_bitmap({1280, 720})
-
+  result := utf8_to_utf32(&"A"[0], 1)
+  result  = utf8_to_utf32(&"ć"[0], 2)
+  result  = utf8_to_utf32(&"ó"[0], 2)
 

programs/user32.kl

@@ -3,7 +3,7 @@ WNDPROC :: (hwnd: HWND, uMsg: UINT, wParam: WPARAM, lParam: LPARAM): LRESULT
 WNDCLASSW :: struct;; style: UINT; lpfnWndProc: WNDPROC; cbClsExtra: int; cbWndExtra: int; hInstance: HINSTANCE; hIcon: HICON; hCursor: HCURSOR; hbrBackground: HBRUSH; lpszMenuName: LPCWSTR; lpszClassName: LPCWSTR
 
 PostQuitMessage :: #foreign(nExitCode: int)
-DefWindowProc :: #foreign (hwnd: HWND, uMsg: UINT, wParam: WPARAM, lParam: LPARAM): LRESULT
+DefWindowProcW :: #foreign (hwnd: HWND, uMsg: UINT, wParam: WPARAM, lParam: LPARAM): LRESULT
 GetDC :: #foreign (hWnd: HWND): HDC
 CreateWindowA :: #foreign (dwExStyle: DWORD, lpClassName: *char, lpWindowName: *char, dwStyle: DWORD, X: int, Y: int, nWidth: int, nHeight: int, hWndParent: HWND, hMenu: HMENU, hInstance: HINSTANCE, lpParam: *void): HWND
 CreateWindowExW :: #foreign (dwExStyle: DWORD, lpClassName: LPCWSTR, lpWindowName: LPCWSTR, dwStyle: DWORD, X: int, Y: int, nWidth: int, nHeight: int, hWndParent: HWND, hMenu: HMENU, hInstance: HINSTANCE, lpPara: LPVOID): HWND

typechecking.cpp

@@ -776,6 +776,7 @@ resolve_expr(Ast_Expr *ast, Resolve_Flag flags, Ast_Type *compound_context){
       }
 
       node->index_original_type = left.type;
+      assert(left.type);
       if(is_string(left.type)){
         return operand_lvalue(type_u8);
       }