From 33f5efdf306c3f4a0289bc18905ac42b0e14f5b0 Mon Sep 17 00:00:00 2001
From: Rui Ueyama <ruiu@cs.stanford.edu>
Date: Sun, 11 Aug 2019 16:27:43 +0900
Subject: [PATCH] Add char literal

---
 tests      |  3 +++
 tokenize.c | 29 +++++++++++++++++++++++++++++
 2 files changed, 32 insertions(+)

diff --git a/tests b/tests
index 6a3ca20..1218b5e 100644
--- a/tests
+++ b/tests
@@ -329,6 +329,9 @@ int main() {
   assert(0, (long)&*(int *)0, "(long)&*(int *)0");
   assert(5, ({ int x=5; long y=(long)&x; *(int*)y; }), "int x=5; long y=(long)&x; *(int*)y");
 
+  assert(97, 'a', "'a'");
+  assert(10, '\n', "\'\\n\'");
+
   printf("OK\n");
   return 0;
 }
diff --git a/tokenize.c b/tokenize.c
index 573c3a2..7fcdcce 100644
--- a/tokenize.c
+++ b/tokenize.c
@@ -211,6 +211,28 @@ static Token *read_string_literal(Token *cur, char *start) {
   return tok;
 }
 
+static Token *read_char_literal(Token *cur, char *start) {
+  char *p = start + 1;
+  if (*p == '\0')
+    error_at(start, "unclosed char literal");
+
+  char c;
+  if (*p == '\\') {
+    p++;
+    c = get_escape_char(*p++);
+  } else {
+    c = *p++;
+  }
+
+  if (*p != '\'')
+    error_at(start, "char literal too long");
+  p++;
+
+  Token *tok = new_token(TK_NUM, cur, start, p - start);
+  tok->val = c;
+  return tok;
+}
+
 // Tokenize `user_input` and returns new tokens.
 Token *tokenize(void) {
   char *p = user_input;
@@ -248,6 +270,13 @@ Token *tokenize(void) {
       continue;
     }
 
+    // Character literal
+    if (*p == '\'') {
+      cur = read_char_literal(cur, p);
+      p += cur->len;
+      continue;
+    }
+
     // Keywords or multi-letter punctuators
     char *kw = starts_with_reserved(p);
     if (kw) {
-- 
GitLab