wip compiler

src/tools/compiler/compiler.h

@@ -32,20 +32,16 @@ typedef struct plex_tab_s PlexTable;
 typedef struct scope_s Scope;
 typedef struct scope_tab_s ScopeTable;
 
+#define MAX_SYMBOL_NAME_LENGTH 64
+
 struct value_type_s {
   SymbolType type;
-  u32 name;
+  char name[MAX_SYMBOL_NAME_LENGTH];
+  u8 name_length;
   u32 size;
   u32 table_ref; // if it is a heap object
 };
 
-struct plex_def_s {
-  u32 name;
-  u32 size;
-  u32 field_ref_start;
-  u32 field_count;
-};
-
 struct plex_fields_tab_s {
   u32 *plex_refs;
   ValueType *fields;
@@ -53,13 +49,20 @@ struct plex_fields_tab_s {
   u32 capacity;
 };
 
+struct plex_def_s {
+  char name[MAX_SYMBOL_NAME_LENGTH];
+  u8 name_length;
+  u32 size;
+  u32 field_ref_start;
+  u32 field_count;
+};
+
 struct plex_tab_s {
   PlexDef *symbols;
   u32 count;
   u32 capacity;
 };
 
-#define MAX_SYMBOL_NAME_LENGTH 64
 struct symbol_s {
   char name[MAX_SYMBOL_NAME_LENGTH];
   u8 name_length;
@@ -80,6 +83,7 @@ struct scope_tab_s {
   u32 count;
   u32 capacity;
   i32 scope_ref;
+  u32 depth;
 };
 
 bool compile(VM *vm, ScopeTable *st, char *source);

src/tools/compiler/parser.c

@@ -7,14 +7,14 @@ typedef struct {
   const char *start;
   const char *current;
   int line;
-} Lexer;
+} Parser;
 
-Lexer lexer;
+Parser parser;
 
-void initLexer(const char *source) {
+void init_parser(const char *source) {
-  lexer.start = source;
+  parser.start = source;
-  lexer.current = source;
+  parser.current = source;
-  lexer.line = 1;
+  parser.line = 1;
 }
 
 static bool isAlpha(char c) {
@@ -23,36 +23,36 @@ static bool isAlpha(char c) {
 
 static bool isDigit(char c) { return c >= '0' && c <= '9'; }
 
-static bool isAtEnd() { return *lexer.current == '\0'; }
+static bool isAtEnd() { return *parser.current == '\0'; }
 
 static char advance() {
-  lexer.current++;
+  parser.current++;
-  return lexer.current[-1];
+  return parser.current[-1];
 }
 
-static char peek() { return *lexer.current; }
+static char peek() { return *parser.current; }
 
 static char peekNext() {
   if (isAtEnd())
     return '\0';
-  return lexer.current[1];
+  return parser.current[1];
 }
 
 static bool match(char expected) {
   if (isAtEnd())
     return false;
-  if (*lexer.current != expected)
+  if (*parser.current != expected)
     return false;
-  lexer.current++;
+  parser.current++;
   return true;
 }
 
 static Token makeToken(TokenType type) {
   Token token;
   token.type = type;
-  token.start = lexer.start;
+  token.start = parser.start;
-  token.length = (int)(lexer.current - lexer.start);
+  token.length = (int)(parser.current - parser.start);
-  token.line = lexer.line;
+  token.line = parser.line;
   return token;
 }
 
@@ -61,7 +61,7 @@ static Token errorToken(const char *message) {
   token.type = TOKEN_ERROR;
   token.start = message;
   token.length = (int)strlen(message);
-  token.line = lexer.line;
+  token.line = parser.line;
   return token;
 }
 
@@ -75,7 +75,7 @@ static void skipWhitespace() {
       advance();
       break;
     case '\n':
-      lexer.line++;
+      parser.line++;
       advance();
       break;
     case '/':
@@ -90,7 +90,7 @@ static void skipWhitespace() {
         advance();
         while (!isAtEnd()) {
           if (peek() == '\n')
-            lexer.line++;
+            parser.line++;
           if (peek() == '*' && peekNext() == '/') {
             advance();
             advance();
@@ -110,8 +110,8 @@ static void skipWhitespace() {
 
 static TokenType checkKeyword(int start, int length, const char *rest,
                               TokenType type) {
-  if (lexer.current - lexer.start == start + length &&
+  if (parser.current - parser.start == start + length &&
-      memcmp(lexer.start + start, rest, length) == 0) {
+      memcmp(parser.start + start, rest, length) == 0) {
     return type;
   }
 
@@ -217,8 +217,6 @@ static TokenType identifierType() {
           switch (lexer.start[2]) {
           case 'a':
             return checkKeyword(3, 1, "d", TOKEN_KEYWORD_READ);
-          case 'f':
-            return checkKeyword(3, 4, "resh", TOKEN_KEYWORD_REFRESH);
           case 't':
             return checkKeyword(3, 3, "urn", TOKEN_KEYWORD_RETURN);
           }
@@ -231,7 +229,14 @@ static TokenType identifierType() {
     if (lexer.current - lexer.start > 1) {
       switch (lexer.start[1]) {
       case 't':
-        return checkKeyword(2, 1, "r", TOKEN_TYPE_STR);
+        if (lexer.current - lexer.start > 2) {
+          switch (lexer.start[2]) {
+          case 'r':
+            return checkKeyword(2, 0, "", TOKEN_TYPE_STR);
+          case 'a':
+            return checkKeyword(2, 1, "t", TOKEN_KEYWORD_STAT);
+          }
+        }
       }
     }
     break;
@@ -269,8 +274,6 @@ static TokenType identifierType() {
       }
     }
     break;
-  case 'g':
-    return checkKeyword(1, 5, "lobal", TOKEN_KEYWORD_GLOBAL);
   }
 
   return TOKEN_IDENTIFIER;
@@ -303,7 +306,7 @@ static Token number() {
 static Token string() {
   while (peek() != '"' && !isAtEnd()) {
     if (peek() == '\n')
-      lexer.line++;
+      parser.line++;
     advance();
   }
 
@@ -317,7 +320,7 @@ static Token string() {
 
 Token next_token() {
   skipWhitespace();
-  lexer.start = lexer.current;
+  parser.start = parser.current;
 
   if (isAtEnd())
     return makeToken(TOKEN_EOF);
@@ -440,8 +443,6 @@ const char *token_type_to_string(TokenType type) {
     return "KEYWORD_TRUE";
   case TOKEN_KEYWORD_FALSE:
     return "KEYWORD_FALSE";
-  case TOKEN_KEYWORD_GLOBAL:
-    return "KEYWORD_GLOBAL";
   case TOKEN_OPERATOR_NOT:
     return "OPERATOR_NOT";
   case TOKEN_OPERATOR_AND:

src/tools/compiler/parser.h

@@ -31,7 +31,6 @@ typedef enum {
   TOKEN_KEYWORD_USE,
   TOKEN_KEYWORD_INIT,
   TOKEN_KEYWORD_THIS,
-  TOKEN_KEYWORD_GLOBAL,
   TOKEN_KEYWORD_OPEN,
   TOKEN_KEYWORD_READ,
   TOKEN_KEYWORD_WRITE,
@@ -82,7 +81,7 @@ typedef struct {
   int line;
 } Token;
 
-void initLexer(const char *source);
+void init_parser(const char *source);
 Token next_token();
 const char* token_type_to_string(TokenType type);
 

src/vm/vm.c

@@ -137,7 +137,7 @@ bool step_vm(VM *vm) {
       child->locals[i] = frame->locals[src_reg];
       mask = 1 << (src_reg % 32);
       if (frame->heap_mask[src_reg / 32] & mask) {
-          child->heap_mask[i / 32] |= 1 << (i % 32);
+        child->heap_mask[i / 32] |= 1 << (i % 32);
       }
     }
 
@@ -174,7 +174,8 @@ bool step_vm(VM *vm) {
         parent->heap_mask[parent->return_reg / 32] |= (1 << parent->return_reg);
       } else {
         parent->locals[parent->return_reg] = value;
-        parent->heap_mask[parent->return_reg / 32] &= ~(1 << parent->return_reg);
+        parent->heap_mask[parent->return_reg / 32] &=
+            ~(1 << parent->return_reg);
       }
     }
 
@@ -473,7 +474,7 @@ bool step_vm(VM *vm) {
     frame->locals[0] = dest;
     vm->flag = 1;
     return true;
-  }  
+  }
   case OP_MEMSET_16: {
     u32 i, start, end;
     u8 value_reg = read_u8(vm, code, vm->pc++);
@@ -535,7 +536,7 @@ bool step_vm(VM *vm) {
     frame->locals[0] = dest;
     vm->flag = 1;
     return true;
-  }  
+  }
   case OP_REG_MOV: {
     u8 dest, src1;
     src1 = read_u8(vm, code, vm->pc);