add constant buffer compression

Makefile

@@ -26,7 +26,7 @@ full-debug: $(CFILES) $(LEXER_C) $(LEXER_H)
 optimised: $(CFILES) $(LEXER_C) $(LEXER_H)
 	mkdir -p bin
 	gcc -O3 -fprofile-generate -o $(BINARY) $(CFILES) $(CFLAGS)
-	${BINARY}
+	${BINARY} test.ar
 	gcc -O3 -fprofile-use -o $(BINARY) $(CFILES) $(CFLAGS)
 	
 

src/main.c

@@ -6,6 +6,7 @@
 #include "translator/translator.h"
 
 #include <endian.h>
+#include <string.h>
 #include <locale.h>
 #include <stdbool.h>
 #include <stddef.h>
@@ -61,8 +62,7 @@ int main(int argc, char *argv[]) {
   constantsSize = htole64(constantsSize);
   bytecodeSize = htole64(bytecodeSize);
   
-
+  fwrite(&FILE_IDENTIFIER, sizeof(char), strlen(FILE_IDENTIFIER), file);
-  fwrite(&FILE_IDENTIFIER, sizeof(char), sizeof(FILE_IDENTIFIER)/sizeof(char), file);
   fwrite(&version_number_htole64ed, sizeof(uint64_t), 1, file);
   fwrite(&regCount, sizeof(uint64_t), 1, file);
   fwrite(&constantsSize, sizeof(uint64_t), 1, file);

src/parser/assignable/access/access.c

@@ -1,6 +1,7 @@
 #include "access.h"
 #include "../../../lexer/token.h"
 #include "../../../memory.h"
+#include "../../string/string.h"
 #include "../../parser.h"
 #include <stdio.h>
 #include <stdlib.h>
@@ -18,11 +19,9 @@ ParsedValue *parse_access(char *file, DArray *tokens, size_t *index,
   if (first_token->type == TOKEN_DOT) {
     error_if_finished(file, tokens, index);
     Token *token = darray_get(tokens, *index);
-    ParsedValue parsedString;
+    ParsedValue *parsedString = parse_string(token, false);
-    parsedString.type = AST_STRING;
+    darray_push(&parsedAccess->access, parsedString);
-    parsedString.data =
+    free(parsedString);
-        strcpy(checked_malloc(strlen(token->value) + 1), token->value);
-    darray_push(&parsedAccess->access, &parsedString);
   } else {
     while (true) {
       skip_newlines_and_indents(tokens, index);

src/parser/dictionary/dictionary.c

@@ -2,6 +2,7 @@
 #include "../../lexer/token.h"
 #include "../../memory.h"
 #include "../parser.h"
+#include "../string/string.h"
 #include <stddef.h>
 #include <stdio.h>
 #include <stdlib.h>
@@ -22,9 +23,12 @@ ParsedValue *parse_dictionary(char *file, DArray *tokens, size_t *index) {
       error_if_finished(file, tokens, index);
       size_t keyIndex = *index;
       Token *keyToken = darray_get(tokens, *index);
-      ParsedValue *key = parse_token(file, tokens, index, true);
+      ParsedValue *key;
       if (keyToken->type == TOKEN_IDENTIFIER) {
-        key->type = AST_STRING;
+        (*index)++;
+        key = parse_string(keyToken, false);
+      } else {
+        key = parse_token(file, tokens, index, true);
       }
       skip_newlines_and_indents(tokens, index);
       error_if_finished(file, tokens, index);

src/parser/number/number.c

@@ -3,12 +3,156 @@
 #include "../parser.h"
 #include "../../memory.h"
 #include <gmp.h>
+#include <string.h>
+// #include <stdio.h>
+// #include <stdlib.h>
+// #include <string.h>
+// #include <ctype.h>
+
+
+// int parse_exponent(const char *exp_str, long *exp_val) {
+//     char *endptr;
+//     long val = strtol(exp_str, &endptr, 10);
+//     if (*endptr != '\0') {
+//         // exponent contains invalid chars or decimal point → reject
+//         return -1;
+//     }
+//     *exp_val = val;
+//     return 0;
+// }
+
+// int mpq_set_decimal_str_exp(mpq_t r, const char *str) {
+//     // Skip leading whitespace
+//     while (isspace(*str)) str++;
+
+//     // Handle sign
+//     int negative = 0;
+//     if (*str == '-') {
+//         negative = 1;
+//         str++;
+//     } else if (*str == '+') {
+//         str++;
+//     }
+
+//     // Copy input to a buffer for manipulation
+//     size_t len = strlen(str);
+//     char *buf = malloc(len + 1);
+//     if (!buf) return -1;
+//     strcpy(buf, str);
+
+//     // Find 'e' or 'E'
+//     char *e_ptr = strchr(buf, 'e');
+//     if (!e_ptr) e_ptr = strchr(buf, 'E');
+
+//     char *exp_str = NULL;
+//     if (e_ptr) {
+//         *e_ptr = '\0';
+//         exp_str = e_ptr + 1;
+//     }
+
+//     // Validate decimal part (digits and one dot)
+//     int dot_count = 0;
+//     for (char *p = buf; *p; p++) {
+//         if (*p == '.') {
+//             if (++dot_count > 1) { free(buf); return -1; }
+//             continue;
+//         }
+//         if (!isdigit((unsigned char)*p)) { free(buf); return -1; }
+//     }
+
+//     // Extract integer and fractional parts
+//     char *dot = strchr(buf, '.');
+//     size_t int_len = dot ? (size_t)(dot - buf) : strlen(buf);
+//     size_t frac_len = dot ? strlen(dot + 1) : 0;
+
+//     // Validate exponent if present
+//     int exp_negative = 0;
+//     long exp_val = 0;
+//     if (exp_str) {
+//         // Skip leading spaces in exponent (not in regex but safe)
+//         while (isspace(*exp_str)) exp_str++;
+
+//         if (*exp_str == '-') {
+//             exp_negative = 1;
+//             exp_str++;
+//         } else if (*exp_str == '+') {
+//             exp_str++;
+//         }
+
+//         if (!isdigit((unsigned char)*exp_str)) {
+//             free(buf);
+//             return -1;
+//         }
+
+//         char *endptr;
+//         exp_val = strtol(exp_str, &endptr, 10);
+//         if (*endptr != '\0') {
+//             free(buf);
+//             return -1;
+//         }
+//         if (exp_negative) exp_val = -exp_val;
+//     }
+
+//     // Build numerator string (integer part + fractional part)
+//     size_t num_len = int_len + frac_len;
+//     if (num_len == 0) { free(buf); return -1; }
+
+//     char *num_str = malloc(num_len + 1);
+//     if (!num_str) { free(buf); return -1; }
+
+//     if (int_len > 0) memcpy(num_str, buf, int_len);
+//     if (frac_len > 0) memcpy(num_str + int_len, dot + 1, frac_len);
+//     num_str[num_len] = '\0';
+
+//     // Calculate denominator exponent considering exponent part
+//     long denom_exp = frac_len - exp_val;
+
+//     mpz_t numerator, denominator;
+//     mpz_init(numerator);
+//     mpz_init(denominator);
+
+//     if (mpz_set_str(numerator, num_str, 10) != 0) {
+//         free(num_str);
+//         free(buf);
+//         mpz_clear(numerator);
+//         mpz_clear(denominator);
+//         return -1;
+//     }
+//     free(num_str);
+//     free(buf);
+
+//     if (denom_exp >= 0) {
+//         mpz_ui_pow_ui(denominator, 10, (unsigned long)denom_exp);
+//     } else {
+//         // denom_exp < 0 means multiply numerator by 10^(-denom_exp)
+//         mpz_ui_pow_ui(denominator, 10, 0);
+//         mpz_ui_pow_ui(numerator, 10, (unsigned long)(-denom_exp));
+//     }
+
+//     if (denom_exp < 0) {
+//         mpz_t temp;
+//         mpz_init(temp);
+//         mpz_ui_pow_ui(temp, 10, (unsigned long)(-denom_exp));
+//         mpz_mul(numerator, numerator, temp);
+//         mpz_clear(temp);
+//         mpz_set_ui(denominator, 1);
+//     }
+
+//     mpq_set_num(r, numerator);
+//     mpq_set_den(r, denominator);
+//     mpq_canonicalize(r);
+
+//     if (negative) mpq_neg(r, r);
+
+//     mpz_clear(numerator);
+//     mpz_clear(denominator);
+
+//     return 0;
+// }
 
 ParsedValue *parse_number(Token *token) {
   ParsedValue *parsedValue = checked_malloc(sizeof(ParsedValue));
-  mpz_t *number = checked_malloc(sizeof(mpz_t));
-  mpz_init_set_str(*number, token->value, 10);
   parsedValue->type = AST_NUMBER;
-  parsedValue->data = number;
+  parsedValue->data = strdup(token->value);
   return parsedValue;
 }

src/parser/operations/operations.c

@@ -1,6 +1,6 @@
 #include "operations.h"
-#include "../parser.h"
 #include "../../memory.h"
+#include "../parser.h"
 #include <stddef.h>
 #include <stdio.h>
 #include <stdlib.h>
@@ -23,22 +23,30 @@ ParsedValue *convert_to_operation(DArray * to_operate_on, DArray * operations) {
     }
     darray_push(&positions, &i);
   }
-  size_t last_position = operations->size-1;
-  darray_push(&positions, &last_position);
   ParsedValue *parsedValue = checked_malloc(sizeof(ParsedValue));
   parsedValue->type = AST_OPERATION;
   ParsedOperation *operationStruct = checked_malloc(sizeof(ParsedOperation));
   parsedValue->data = operationStruct;
   operationStruct->operation = operation;
   darray_init(&operationStruct->to_operate_on, sizeof(ParsedValue));
-  last_position = 0;
+  size_t last_position = 0;
+  size_t to_operate_on_last_position = 0;
   for (size_t i = 0; i < positions.size; i++) {
     size_t *position = darray_get(&positions, i);
-    DArray to_operate_on_slice = darray_slice(to_operate_on, last_position, *position+1);
+    DArray to_operate_on_slice = darray_slice(
-    DArray operations_slice = darray_slice(operations, last_position, *position);
+        to_operate_on, to_operate_on_last_position, (*position) + 1);
-    darray_push(&operationStruct->to_operate_on, convert_to_operation(&to_operate_on_slice, &operations_slice));
+    DArray operations_slice =
-    last_position = *position;
+        darray_slice(operations, last_position, *position);
+    darray_push(&operationStruct->to_operate_on,
+                convert_to_operation(&to_operate_on_slice, &operations_slice));
+    last_position = (*position);
+    to_operate_on_last_position = (*position) + 1;
   }
+  DArray to_operate_on_slice =
+      darray_slice(to_operate_on, to_operate_on_last_position, to_operate_on->size);
+  DArray operations_slice = darray_slice(operations, last_position, operations->size);
+  darray_push(&operationStruct->to_operate_on,
+              convert_to_operation(&to_operate_on_slice, &operations_slice));
   darray_free(&positions, NULL);
   return parsedValue;
 }
@@ -48,6 +56,7 @@ ParsedValue *parse_operations(char *file, DArray *tokens, size_t *index,
   DArray to_operate_on;
   darray_init(&to_operate_on, sizeof(ParsedValue));
   darray_push(&to_operate_on, first_parsed_value);
+  free(first_parsed_value);
 
   DArray operations;
   darray_init(&operations, sizeof(TokenType));
@@ -66,7 +75,10 @@ ParsedValue *parse_operations(char *file, DArray *tokens, size_t *index,
     darray_push(&operations, &token->type);
     (*index)++;
     error_if_finished(file, tokens, index);
-    darray_push(&to_operate_on, parse_token_full(file, tokens, index, true, false));
+    ParsedValue *parsedValue =
+        parse_token_full(file, tokens, index, true, false);
+    darray_push(&to_operate_on, parsedValue);
+    free(parsedValue);
   }
   ParsedValue *output = convert_to_operation(&to_operate_on, &operations);
   darray_free(&to_operate_on, NULL);

src/parser/parser.c

@@ -82,7 +82,7 @@ ParsedValue *parse_token_full(char *file, DArray *tokens, size_t *index,
     break;
   case TOKEN_STRING:
     (*index)++;
-    output = parse_string(token);
+    output = parse_string(token, true);
     break;
   case TOKEN_NEW_LINE:
     (*index)++;
@@ -196,9 +196,12 @@ void free_parsed(void *ptr) {
   ParsedValue *parsed = ptr;
   switch (parsed->type) {
   case AST_IDENTIFIER:
-  case AST_STRING:
+  case AST_NUMBER:
     free(parsed->data);
     break;
+  case AST_STRING:
+    free_parsed_string(parsed);
+    break;
   case AST_ASSIGN:
     free_parse_assign(parsed);
     break;
@@ -211,9 +214,6 @@ void free_parsed(void *ptr) {
   case AST_ACCESS:
     free_parse_access(parsed);
     break;
-  case AST_NUMBER:
-    mpz_clear(parsed->data);
-    break;
   case AST_NULL:
   case AST_BOOLEAN:
     break;

src/parser/string/string.c

@@ -245,12 +245,24 @@ char *unquote(char *str, size_t *decoded_len) {
   return unescaped;
 }
 
-ParsedValue *parse_string(Token* token) {
+ParsedValue *parse_string(Token* token, bool to_unquote) {
   ParsedValue *parsedValue = checked_malloc(sizeof(ParsedValue));
   parsedValue->type = AST_STRING;
   ParsedString *parsedString = checked_malloc(sizeof(ParsedString));
   parsedValue->data = parsedString;
+  if (to_unquote) {
     parsedString->length = 0;
     parsedString->string = unquote(token->value, &parsedString->length);
+  } else {
+    parsedString->string = strdup(token->value);
+    parsedString->length = token->length;
+  }
   return parsedValue;
 }
+
+void free_parsed_string(void *ptr) {
+  ParsedValue *parsedValue = ptr;
+  ParsedString *parsedString = parsedValue->data;
+  free(parsedString->string);
+  free(parsedString);
+}

src/parser/string/string.h

@@ -15,6 +15,8 @@ char *swap_quotes(char *input, char quote);
 
 char *unquote(char *str, size_t *decoded_len);
 
-ParsedValue *parse_string(Token* token);
+ParsedValue *parse_string(Token* token, bool to_unquote);
+
+void free_parsed_string(void *ptr);
 
 #endif // STRING_UTILS_H

src/translator/declaration/declaration.c

@@ -16,14 +16,20 @@ size_t translate_parsed_declaration(Translated *translated,
     if (i==0) first = temp;
     size_t length = strlen(singleDeclaration->name);
     size_t offset = arena_push(&translated->constants, singleDeclaration->name, length);
+    
     push_instruction_code(translated, OP_LOAD_CONST);
     push_instruction_code(translated, 1);
     push_instruction_code(translated, TYPE_OP_STRING);
     push_instruction_code(translated,length);
     push_instruction_code(translated, offset);
+
     push_instruction_code(translated, OP_DECLARE);
     push_instruction_code(translated, 0);
     push_instruction_code(translated, 1);
   }
+  if (delcarations->size != 1) {
+    push_instruction_code(translated, OP_LOAD_NULL);
+    push_instruction_code(translated, 0);
+  }
   return first;
 }

src/translator/number/number.c

@@ -0,0 +1,21 @@
+#include "../translator.h"
+#include "number.h"
+#include <gmp.h>
+#include <stddef.h>
+#include <stdio.h>
+#include <string.h>
+
+size_t translate_parsed_number(Translated *translated, ParsedValue *parsedValue) {
+  char *number_str = (char*)parsedValue->data;
+  size_t length = strlen(number_str);
+  size_t number_pos = arena_push(&translated->constants, number_str, length);
+  set_registers(translated, 1);
+  
+  size_t start = push_instruction_code(translated, OP_LOAD_CONST);
+  push_instruction_code(translated, 0);
+
+  push_instruction_code(translated, TYPE_OP_NUMBER);
+  push_instruction_code(translated,length);
+  push_instruction_code(translated, number_pos);
+  return start;
+}

src/translator/number/number.h

@@ -0,0 +1,7 @@
+#ifndef BYTECODE_NUMBER_H
+#define BYTECODE_NUMBER_H
+#include "../translator.h"
+
+size_t translate_parsed_number(Translated *translated, ParsedValue *parsedValue);
+
+#endif

src/translator/string/string.c

@@ -14,7 +14,5 @@ size_t translate_parsed_string(Translated *translated, ParsedValue *parsedValue)
   push_instruction_code(translated, TYPE_OP_STRING);
   push_instruction_code(translated,parsedString->length);
   push_instruction_code(translated, string_pos);
-  fwrite(parsedString->string, 1, parsedString->length, stdout);
-  putchar('\n');
   return start;
 }

src/translator/translator.c

@@ -1,6 +1,7 @@
 #include "translator.h"
-#include "string/string.h"
 #include "declaration/declaration.h"
+#include "number/number.h"
+#include "string/string.h"
 #include <stddef.h>
 #include <stdint.h>
 #include <stdio.h>
@@ -19,13 +20,13 @@ void arena_resize(ConstantArena *arena, size_t new_size) {
     return;
   arena->data = realloc(arena->data, new_capacity);
   if (!arena->data) {
-    fprintf(stderr, "error: failed to resize arena from %zu to %zu\n", new_capacity, arena->capacity);
+    fprintf(stderr, "error: failed to resize arena from %zu to %zu\n",
+            new_capacity, arena->capacity);
     exit(EXIT_FAILURE);
   }
   arena->capacity = new_capacity;
 }
 
-
 void arena_free(ConstantArena *arena) {
   free(arena->data);
   arena->capacity = 0;
@@ -37,6 +38,13 @@ void * arena_get(ConstantArena *arena, size_t offset) {
 }
 
 size_t arena_push(ConstantArena *arena, const void *data, size_t length) {
+  if (arena->size >= length) {
+    for (size_t i = 0; i <= (arena->size - length); i++) {
+      if (memcmp(data, arena->data + i, length) == 0) {
+        return i;
+      }
+    }
+  }
   arena_resize(arena, arena->size + length);
   size_t offset = arena->size;
   memcpy(arena->data + arena->size, data, length);
@@ -52,7 +60,8 @@ Translated init_translator() {
   return translated;
 }
 
-void set_instruction_code(Translated * translator, size_t offset, uint64_t code) {
+void set_instruction_code(Translated *translator, size_t offset,
+                          uint64_t code) {
   code = htole64(code);
   size_t *ptr = (translator->bytecode.data + offset);
   *ptr = code;
@@ -66,23 +75,31 @@ size_t push_instruction_code(Translated * translator, uint64_t code) {
 }
 
 void set_registers(Translated *translator, size_t count) {
-  if (count>translator->registerCount) translator->registerCount = count;
+  if (count > translator->registerCount)
+    translator->registerCount = count;
 }
 
-size_t translate_parsed(Translated * translator, ParsedValue * parsedValue) {
+size_t translate_parsed(Translated *translated, ParsedValue *parsedValue) {
   switch (parsedValue->type) {
   case AST_STRING:
-      return translate_parsed_string(translator,parsedValue);
+    return translate_parsed_string(translated, parsedValue);
   case AST_DECLARATION:
-      return translate_parsed_declaration(translator,parsedValue);
+    return translate_parsed_declaration(translated, parsedValue);
+  case AST_NUMBER:
+    return translate_parsed_number(translated, parsedValue);
+  case AST_NULL:
+    set_registers(translated, 1);
+    size_t output = push_instruction_code(translated, OP_LOAD_NULL);
+    push_instruction_code(translated, 0);
+    return output;
   }
   return 0;
 }
 
-void translate(Translated * translator, DArray *ast) {
+void translate(Translated *translated, DArray *ast) {
   for (size_t i = 0; i < ast->size; i++) {
     ParsedValue *parsedValue = darray_get(ast, i);
-    translate_parsed(translator,parsedValue);
+    translate_parsed(translated, parsedValue);
   }
 }
 

src/translator/translator.h

@@ -8,7 +8,7 @@
 #include <stdint.h>
 
 typedef enum { OP_LOAD_CONST = 255, OP_DECLARE, OP_LOAD_NULL, OP_JUMP } OperationType;
-typedef enum { TYPE_OP_STRING = 255 } types;
+typedef enum { TYPE_OP_STRING = 255, TYPE_OP_NUMBER } types;
 
 typedef struct {
   void *data;

test.ar

@@ -17,8 +17,10 @@
 "🇬🇧"
 "hello\u0000world"
 
+1.24323234e2312324
+
 let a, 
-    b = 1, 
+    b = "hello", 
     c, 
     d = 42, 
     temp_result, 
@@ -57,7 +59,7 @@ else term.log("bruh")
 
 
 
-mm=1/2/4/2/4354/534/534//534//3422*404203420234+3432423324&&430234230||4320423040230423^384239423043024923%4432042304920.3432423423
+mm=x/2/4/2/4354/534/534//534//3422*404203420234+3432423324&&430234230||4320423040230423^384239423043024923%4432042304920.3432423423
 
 let x = [
     'hello world',
@@ -68,7 +70,7 @@ let x = [
 term.log(x[0:1:1])
 
 let y = {
-    'hello':10,
+    'hello':test,
     world:'nice'
 }