Expanded keywords to have their own TokenType enum

include/sync/lexer.h

@@ -23,6 +23,55 @@ typedef enum {
     TOKEN_RBRACKET,
     TOKEN_CHARACTER,
     TOKEN_STRING,
+    // Definitions and Declarations
+    TOKEN_KW_CONST,
+    TOKEN_KW_ENUM,
+    TOKEN_KW_FN,
+    TOKEN_KW_LET,
+    TOKEN_KW_MUT,
+    TOKEN_KW_PUBLIC,
+    TOKEN_KW_STATIC,
+    TOKEN_KW_STRUCT,
+    TOKEN_KW_UNION,
+    // Control Flow
+    TOKEN_KW_BREAK,
+    TOKEN_KW_CASE,
+    TOKEN_KW_CONTINUE,
+    TOKEN_KW_DEFAULT,
+    TOKEN_KW_DO,
+    TOKEN_KW_ELSE,
+    TOKEN_KW_FOR,
+    TOKEN_KW_IF,
+    TOKEN_KW_MATCH,
+    TOKEN_KW_RETURN,
+    TOKEN_KW_SWITCH,
+    TOKEN_KW_WHILE,
+    // Values
+    TOKEN_KW_FALSE,
+    TOKEN_KW_TRUE,
+    // Types
+    TOKEN_KW_BOOL,
+    TOKEN_KW_F32,
+    TOKEN_KW_F64,
+    TOKEN_KW_I8,
+    TOKEN_KW_I16,
+    TOKEN_KW_I32,
+    TOKEN_KW_I64,
+    TOKEN_KW_U8,
+    TOKEN_KW_U16,
+    TOKEN_KW_U32,
+    TOKEN_KW_U64,
+    TOKEN_KW_VOID,
+    // Modules
+    TOKEN_KW_AS,
+    TOKEN_KW_IMPORT,
+    // Operators
+    TOKEN_KW_AND,
+    TOKEN_KW_IS,
+    TOKEN_KW_NOT,
+    TOKEN_KW_OR,
+    TOKEN_KW_SIZEOF,
+    TOKEN_KW_XOR,
 } TokenType;
 
 typedef struct {

src/lexer.c

@@ -9,6 +9,73 @@
 #include "sync/types.h"
 #include "sync/lexer.h"
 
+typedef struct {
+    const char* value;
+    TokenType type;
+} KeywordPair;
+
+typedef struct {
+    char is_keyword;
+    TokenType token;
+} KeywordResult;
+
+const KeywordPair KEYWORDS[] = {
+    // Definitions and Declarations
+    {"const", TOKEN_KW_CONST},
+    {"enum", TOKEN_KW_ENUM},
+    {"fn", TOKEN_KW_FN},
+    {"let", TOKEN_KW_LET},
+    {"mut", TOKEN_KW_MUT},
+    {"public", TOKEN_KW_PUBLIC},
+    {"static", TOKEN_KW_STATIC},
+    {"struct", TOKEN_KW_STRUCT},
+    {"union", TOKEN_KW_UNION},
+    // Control Flow
+    {"break", TOKEN_KW_BREAK},
+    {"case", TOKEN_KW_CASE},
+    {"continue", TOKEN_KW_CONTINUE},
+    {"default", TOKEN_KW_DEFAULT},
+    {"do", TOKEN_KW_DO},
+    {"else", TOKEN_KW_ELSE},
+    {"for", TOKEN_KW_FOR},
+    {"if", TOKEN_KW_IF},
+    {"match", TOKEN_KW_MATCH},
+    {"return", TOKEN_KW_RETURN},
+    {"switch", TOKEN_KW_SWITCH},
+    {"while", TOKEN_KW_WHILE},
+    // Values
+    {"False", TOKEN_KW_FALSE},
+    {"True", TOKEN_KW_TRUE},
+    // Types
+    {"bool", TOKEN_KW_BOOL},
+    {"f32", TOKEN_KW_F32},
+    {"f64", TOKEN_KW_F64},
+    {"i8", TOKEN_KW_I8},
+    {"i16", TOKEN_KW_I16},
+    {"i32", TOKEN_KW_I32},
+    {"i64", TOKEN_KW_I64},
+    {"u8", TOKEN_KW_U8},
+    {"u16", TOKEN_KW_U16},
+    {"u32", TOKEN_KW_U32},
+    {"u64", TOKEN_KW_U64},
+    {"void", TOKEN_KW_VOID},
+    // Modules
+    {"as", TOKEN_KW_AS},
+    {"import", TOKEN_KW_IMPORT},
+    // Operators
+    {"and", TOKEN_KW_AND},
+    {"is", TOKEN_KW_IS},
+    {"not", TOKEN_KW_NOT},
+    {"or", TOKEN_KW_OR},
+    {"sizeof", TOKEN_KW_SIZEOF},
+    {"xor", TOKEN_KW_XOR},
+};
+
+const char* RESERVED_KEYWORDS[] = {
+    "extern", "f16", "f128", "i128", "impl", "in", "move", "new", "self",
+    "super", "trait", "tuple", "type", "u128", "use", "where", "yeet",
+};
+
 void lexer_init(Lexer* lexer, const char* filename, const char* source) {
     lexer->filename = filename;
     lexer->source = source;
@@ -84,6 +151,26 @@ static char is_identifier_char(char c) {
     return isalnum(c) || c == '_';
 }
 
+static KeywordResult is_keyword(Lexer* lexer, size_t start, size_t length) {
+    size_t num_keywords = sizeof(KEYWORDS) / sizeof(KEYWORDS[0]);
+    for (size_t i = 0; i < num_keywords; i++)
+        if (
+            strncmp(&lexer->source[start], KEYWORDS[i].value, length) == 0 &&
+            strlen(KEYWORDS[i].value) == length
+        ) return (KeywordResult){1, KEYWORDS[i].type};
+    return (KeywordResult){0, TOKEN_EOF};
+}
+
+static char is_future_keyword(Lexer* lexer, size_t start, size_t length) {
+    size_t num_keywords = sizeof(RESERVED_KEYWORDS) / sizeof(RESERVED_KEYWORDS[0]);
+    for (size_t i = 0; i < num_keywords; i++)
+        if (
+            strncmp(&lexer->source[start], RESERVED_KEYWORDS[i], length) == 0 &&
+            strlen(RESERVED_KEYWORDS[i]) == length
+        ) return 1;
+    return 0;
+}
+
 static LexerResult lexer_result(Lexer* lexer, TokenType type, size_t start, size_t start_line) {
     TokenResult* result = (TokenResult*)malloc(sizeof(TokenResult));
     if (result == NULL)
@@ -181,6 +268,12 @@ static LexerResult lexer_next(Lexer* lexer) {
     // Identifiers
     if (is_identifier_start(c)) {
         while (is_identifier_char(peek(lexer))) advance(lexer);
+        size_t length = lexer->pos - start;
+        if (is_future_keyword(lexer, start, length))
+            return lexer_error(lexer, "Reserved Keyword for future use", start, start_line);
+        KeywordResult result = is_keyword(lexer, start, length);
+        if (result.is_keyword)
+            return lexer_result(lexer, result.token, start, start_line);
         return lexer_result(lexer, TOKEN_IDENTIFIER, start, start_line);
     }
 

src/main.c

@@ -10,10 +10,23 @@
 #include "sync/syntax.h"
 
 const char* TOKEN_TYPES[] = {
-    "EOF", "IDENTIFIER", "NUMBER", "OPERATOR",
-    "LPAREN", "RPAREN", "SEMICOLON", "LBRACE",
-    "RBRACE", "LBRACKET", "RBRACKET", "CHARACTER",
-    "STRING"
+    "EOF", "IDENTIFIER", "NUMBER", "OPERATOR", "LPAREN", "RPAREN", "SEMICOLON",
+    "LBRACE", "RBRACE", "LBRACKET", "RBRACKET", "CHARACTER", "STRING",
+    // Definitions and Declarations
+    "KW_CONST", "KW_ENUM", "KW_FN", "KW_LET", "KW_MUT", "KW_PUBLIC",
+    "KW_STATIC", "KW_STRUCT", "KW_UNION",
+    // Control Flow
+    "KW_BREAK", "KW_CASE", "KW_CONTINUE", "KW_DEFAULT", "KW_DO", "KW_ELSE",
+    "KW_FOR", "KW_IF", "KW_MATCH", "KW_RETURN", "KW_SWITCH", "KW_WHILE",
+    // Values
+    "KW_FALSE", "KW_TRUE",
+    // Types
+    "KW_BOOL", "KW_F32", "KW_F64", "KW_I8", "KW_I16", "KW_I32", "KW_I64",
+    "KW_U8", "KW_U16", "KW_U32", "KW_U64", "KW_VOID",
+    // Modules
+    "KW_AS", "KW_IMPORT",
+    // Operators
+    "KW_AND", "KW_IS", "KW_NOT", "KW_OR", "KW_SIZEOF", "KW_XOR",
 };
 
 static void print_token(Token token) {

src/syntax.c

@@ -11,7 +11,6 @@
 #include "sync/syntax.h"
 
 typedef struct {
-    const char* filename;
     TokenArray tokens;
     size_t pos;
 } Parser;
@@ -106,7 +105,45 @@ static SyntaxResult syntax_error(const char* message, Token token) {
     }};
 }
 
+static SyntaxResult parse_expression(Parser* parser) {
+}
+
 static SyntaxResult parse_statement(Parser* parser) {
+    Child* children = NULL;
+
+    if (
+        parser->tokens.tokens[parser->pos].type == TOKEN_EOF ||
+        parser->tokens.tokens[parser->pos].type == TOKEN_RBRACE
+    ) return syntax_error("Expected statement", parser->tokens.tokens[parser->pos]);
+    SyntaxResult result = parse_expression(parser);
+    if (result.type == SYNC_ERROR) return result;
+    if (result.result.type == SYNC_ERROR) return result;
+    if (children == NULL) {
+        children = (Child*)malloc(sizeof(Child));
+        if (children == NULL)
+            return (SyntaxResult){SYNC_ERROR, .error = (GeneralError){"Failed to allocate memory.", 1}};
+        children->next = NULL;
+        children->node = result.result.result;
+    } else {
+        GeneralError error = add_child(children, result.result.result);
+        if (error.message != NULL) return (SyntaxResult){SYNC_ERROR, .error = error};
+    }
+
+    if (parser->tokens.tokens[parser->pos].type != TOKEN_SEMICOLON)
+        return syntax_error("Expected statement", parser->tokens.tokens[parser->pos]);
+
+    SyntaxNode node = {
+        .type = NODE_SEMICOLON,
+        .start = parser->tokens.tokens[parser->pos].start,
+        .length = parser->tokens.tokens[parser->pos].length,
+        .file_info = parser->tokens.tokens[parser->pos].file_info,
+        .children = (Children){0, NULL}
+    };
+    GeneralError error = add_child(children, result.result.result);
+    if (error.message != NULL) return (SyntaxResult){SYNC_ERROR, .error = error};
+    parser->pos++;
+
+    return syntax_result(NODE_STATEMENT, children);
 }
 
 static SyntaxResult parse_block(Parser* parser) {
@@ -120,10 +157,14 @@ static SyntaxResult parse_block(Parser* parser) {
             if (result.result.type == SYNC_ERROR) return result;
             if (children == NULL) {
                 children = (Child*)malloc(sizeof(Child));
-                if (children == NULL) return (SyntaxResult){SYNC_ERROR, .error = (GeneralError){"Failed to allocate memory.", 1}};
+                if (children == NULL)
+                    return (SyntaxResult){SYNC_ERROR, .error = (GeneralError){"Failed to allocate memory.", 1}};
                 children->next = NULL;
                 children->node = result.result.result;
-            } else add_child(children, result.result.result);
+            } else {
+                GeneralError error = add_child(children, result.result.result);
+                if (error.message != NULL) return (SyntaxResult){SYNC_ERROR, .error = error};
+            }
             if (parser->tokens.tokens[parser->pos].type == TOKEN_EOF)
                 return syntax_error("Expected '}'", parser->tokens.tokens[parser->pos]);
         }
@@ -141,26 +182,25 @@ static SyntaxResult parse_block(Parser* parser) {
     return syntax_result(NODE_BLOCK, children);
 }
 
-SyntaxResult syntactical_analysis(TokenArray tokens) {
-    Parser* parser = (Parser*)malloc(sizeof(Parser));
-    if (parser == NULL) return (SyntaxResult){SYNC_ERROR, .error = (GeneralError){"Failed to allocate memory.", 1}};
+SyntaxResult syntactical_analysis(const char* filename, TokenArray tokens) {
+    Parser parser = { tokens, 0 };
     Child* children = NULL;
 
-    while (parser->tokens.tokens[parser->pos].type != TOKEN_EOF) {
-        SyntaxResult result = parse_block(parser);
-        if (result.type == SYNC_ERROR) { free(parser); return result; }
-        if (result.result.type == SYNC_ERROR) { free(parser); return result; }
+    while (parser.tokens.tokens[parser.pos].type != TOKEN_EOF) {
+        SyntaxResult result = parse_block(&parser);
+        if (result.type == SYNC_ERROR) return result;
+        if (result.result.type == SYNC_ERROR) return result;
         if (children == NULL) {
             children = (Child*)malloc(sizeof(Child));
-            if (children == NULL) {
-                free(parser);
+            if (children == NULL)
                 return (SyntaxResult){SYNC_ERROR, .error = (GeneralError){"Failed to allocate memory.", 1}};
-            }
             children->next = NULL;
             children->node = result.result.result;
-        } else add_child(children, result.result.result);
+        } else {
+            GeneralError error = add_child(children, result.result.result);
+            if (error.message != NULL) return (SyntaxResult){SYNC_ERROR, .error = error};
+        }
     }
 
-    free(parser);
     return syntax_result(NODE_FILE, children);
 }