From 94abb9e3d43e7559befb0aa02894227ec02eab41 Mon Sep 17 00:00:00 2001
From: Aryadev Chavali <aryadev@aryadevchavali.com>
Date: Wed, 25 Oct 2023 21:43:13 +0100
Subject: Separated lexer from main file in asm

---
 asm/lexer.c | 159 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 159 insertions(+)
 create mode 100644 asm/lexer.c

(limited to 'asm/lexer.c')

diff --git a/asm/lexer.c b/asm/lexer.c
new file mode 100644
index 0000000..03f7d05
--- /dev/null
+++ b/asm/lexer.c
@@ -0,0 +1,159 @@
+/* Copyright (C) 2023 Aryadev Chavali
+
+ * You may distribute and modify this code under the terms of the GPLv2
+ * license.  You should have received a copy of the GPLv2 license with
+ * this file.  If not, please write to: aryadev@aryadevchavali.com.
+
+ * Created: 2023-10-24
+ * Author: Aryadev Chavali
+ * Description: Lexer for assembly language
+ */
+
+#include <ctype.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <string.h>
+
+#include "./lexer.h"
+
+const char *token_type_as_cstr(token_type_t type)
+{
+  switch (type)
+  {
+  case TOKEN_LITERAL_BYTE:
+    return "LITERAL_BYTE";
+  case TOKEN_LITERAL_HWORD:
+    return "LITERAL_HWORD";
+  case TOKEN_LITERAL_WORD:
+    return "LITERAL_WORD";
+  case TOKEN_LITERAL_CHAR:
+    return "LITERAL_CHAR";
+  case TOKEN_SYMBOL:
+    return "SYMBOL";
+  }
+  return "";
+}
+
+size_t space_left(buffer_t *buffer)
+{
+  if (buffer->available == buffer->used)
+    return 0;
+  return buffer->available - 1 - buffer->used;
+}
+
+bool is_symbol(char c)
+{
+  return isalpha(c) || c == '-' || c == '_';
+}
+
+token_t tokenise_symbol(buffer_t *buffer)
+{
+  token_t token = {.type = TOKEN_SYMBOL, .str_size = 0};
+  for (; token.str_size < space_left(buffer) &&
+         is_symbol(buffer->data[buffer->used + token.str_size]);
+       ++token.str_size)
+    continue;
+  token.str = calloc(token.str_size + 1, 1);
+  memcpy(token.str, buffer->data + buffer->used, token.str_size);
+  token.str[token.str_size] = '\0';
+  buffer->used += token.str_size;
+  return token;
+}
+
+void tokenise_literal(buffer_t *buffer, token_t *token)
+{
+  token->str_size = 0;
+  for (; token->str_size < space_left(buffer) &&
+         isdigit(buffer->data[buffer->used + token->str_size]);
+       ++token->str_size)
+    continue;
+  token->str = calloc(token->str_size + 1, 1);
+  memcpy(token->str, buffer->data + buffer->used, token->str_size);
+  token->str[token->str_size] = '\0';
+  buffer->used += token->str_size;
+}
+
+token_t tokenise_byte_literal(buffer_t *buffer)
+{
+  token_t token = {.type = TOKEN_LITERAL_BYTE};
+  tokenise_literal(buffer, &token);
+  return token;
+}
+
+token_t tokenise_hword_literal(buffer_t *buffer)
+{
+  token_t token = {.type = TOKEN_LITERAL_HWORD};
+  tokenise_literal(buffer, &token);
+  return token;
+}
+
+token_t tokenise_word_literal(buffer_t *buffer)
+{
+  token_t token = {.type = TOKEN_LITERAL_WORD};
+  tokenise_literal(buffer, &token);
+  return token;
+}
+
+token_t tokenise_char_literal(buffer_t *buffer)
+{
+  token_t token = {.type = TOKEN_LITERAL_CHAR, .str_size = 1};
+  token.str     = calloc(1, 1);
+  token.str[0]  = buffer->data[buffer->used + 1];
+  buffer->used += 3;
+  return token;
+}
+
+token_t *tokenise_buffer(buffer_t *buffer, size_t *n_tokens)
+{
+  buffer_t tokens = {0};
+  darr_init(&tokens, sizeof(token_t));
+  while (space_left(buffer) != 0)
+  {
+    bool is_token = true;
+    token_t t     = {0};
+    char c        = buffer->data[buffer->used];
+    if (isspace(c) || c == '\0')
+    {
+      // Clean whitespace
+      for (; space_left(buffer) > 0 && (isspace(c) || c == '\0');
+           ++buffer->used, c = buffer->data[buffer->used])
+        continue;
+      is_token = false;
+    }
+    else if (space_left(buffer) > 1 && isdigit(buffer->data[buffer->used + 1]))
+    {
+      // Parsing numeric literals
+      switch (c)
+      {
+      case 'b':
+        buffer->used++;
+        t = tokenise_byte_literal(buffer);
+        break;
+      case 'h':
+        buffer->used++;
+        t = tokenise_hword_literal(buffer);
+        break;
+      case 'w':
+        buffer->used++;
+        t = tokenise_word_literal(buffer);
+        break;
+      default:
+        // TODO: Lex Error (INVALID_NUMERIC_LITERAL)
+        exit(1);
+      }
+    }
+    else if (is_symbol(c))
+      t = tokenise_symbol(buffer);
+    else if (c == '\'')
+    {
+      if (space_left(buffer) < 2 || buffer->data[buffer->used + 2] != '\'')
+        // TODO: Lex Error (INVALID_CHAR_LITERAL)
+        exit(1);
+      t = tokenise_char_literal(buffer);
+    }
+    if (is_token)
+      darr_append_bytes(&tokens, (byte *)&t, sizeof(t));
+  }
+  *n_tokens = tokens.used / sizeof(token_t);
+  return (token_t *)tokens.data;
+}
-- 
cgit v1.2.3-13-gbd6f