From c2cc1d3c4500caa34da5e68eb62b7474caf96fe2 Mon Sep 17 00:00:00 2001
From: Rui Ueyama <ruiu@cs.stanford.edu>
Date: Wed, 7 Oct 2020 20:14:14 +0900
Subject: [PATCH] Add \x<hexadecimal-sequence>

---
 test.sh    |  4 ++++
 tokenize.c | 21 +++++++++++++++++++++
 2 files changed, 25 insertions(+)

diff --git a/test.sh b/test.sh
index 67eb37e..e070ef9 100755
--- a/test.sh
+++ b/test.sh
@@ -202,5 +202,9 @@ assert 0 'int main() { return "\0"[0]; }'
 assert 16 'int main() { return "\20"[0]; }'
 assert 65 'int main() { return "\101"[0]; }'
 assert 104 'int main() { return "\1500"[0]; }'
+assert 0 'int main() { return "\x00"[0]; }'
+assert 119 'int main() { return "\x77"[0]; }'
+assert 165 'int main() { return "\xA5"[0]; }'
+assert 255 'int main() { return "\x00ff"[0]; }'
 
 echo OK
diff --git a/tokenize.c b/tokenize.c
index 403211f..2b1accb 100644
--- a/tokenize.c
+++ b/tokenize.c
@@ -79,6 +79,14 @@ static bool is_ident2(char c) {
   return is_ident1(c) || ('0' <= c && c <= '9');
 }
 
+static int from_hex(char c) {
+  if ('0' <= c && c <= '9')
+    return c - '0';
+  if ('a' <= c && c <= 'f')
+    return c - 'a' + 10;
+  return c - 'A' + 10;
+}
+
 // Read a punctuator token from p and returns its length.
 static int read_punct(char *p) {
   if (startswith(p, "==") || startswith(p, "!=") ||
@@ -112,6 +120,19 @@ static int read_escaped_char(char **new_pos, char *p) {
     return c;
   }
 
+  if (*p == 'x') {
+    // Read a hexadecimal number.
+    p++;
+    if (!isxdigit(*p))
+      error_at(p, "invalid hex escape sequence");
+
+    int c = 0;
+    for (; isxdigit(*p); p++)
+      c = (c << 4) + from_hex(*p);
+    *new_pos = p;
+    return c;
+  }
+
   *new_pos = p + 1;
 
   // Escape sequences are defined using themselves here. E.g.