Add, and use, macros to do locale-independent case mapping.

This means we get the same behavior in Turkish locales (where, if we
aren't in the C locale, we might get lower-case "i" mapped to upper-case
"I with dot" and upper-case "I" mapped to lower-case "i without dot),
and may also suppress some shortening warnings from MSVC.

3 files changed, 23 insertions, 5 deletions

diff --git a/netdissect.h b/netdissect.h
index 0bacbfc7..36adf852 100644
--- a/netdissect.h
+++ b/netdissect.h
@@ -350,15 +350,33 @@ extern void txtproto_print(netdissect_options *, const u_char *, u_int,
 
 /*
  * Locale-independent macros for testing character properties and
- * stripping the 8th bit from characters.  Assumed to be handed
- * a value between 0 and 255, i.e. don't hand them a char, as
- * those might be in the range -128 to 127.
+ * stripping the 8th bit from characters.
+ *
+ * Byte values outside the ASCII range are considered unprintable, so
+ * both ND_ISPRINT() and ND_ISGRAPH() return "false" for them.
+ *
+ * Assumed to be handed a value between 0 and 255, i.e. don't hand them
+ * a char, as those might be in the range -128 to 127.
  */
 #define ND_ISASCII(c)	(!((c) & 0x80))	/* value is an ASCII code point */
 #define ND_ISPRINT(c)	((c) >= 0x20 && (c) <= 0x7E)
 #define ND_ISGRAPH(c)	((c) > 0x20 && (c) <= 0x7E)
 #define ND_TOASCII(c)	((c) & 0x7F)
 
+/*
+ * Locale-independent macros for coverting to upper or lower case.
+ *
+ * Byte values outside the ASCII range are not converted.  Byte values
+ * *in* the ASCII range are converted to byte values in the ASCII range;
+ * in particular, 'i' is upper-cased to 'I" and 'I' is lower-cased to 'i',
+ * even in Turkish locales.
+ *
+ * Assumed to be handed a value between 0 and 255, i.e. don't hand
+ * them a char, as those might be in the range -128 to 127.
+ */
+#define ND_TOLOWER(c)	(((c) >= 'A' && (c) <= 'Z') ? (c) - 'A' + 'a' : (c))
+#define ND_TOUPPER(c)	(((c) >= 'a' && (c) <= 'z') ? (c) - 'a' + 'z' : (c))
+
 #if (defined(__i386__) || defined(_M_IX86) || defined(__X86__) || defined(__x86_64__) || defined(_M_X64)) || \
     (defined(__arm__) || defined(_M_ARM) || defined(__aarch64__)) || \
     (defined(__m68k__) && (!defined(__mc68000__) && !defined(__mc68010__))) || \
diff --git a/print-zephyr.c b/print-zephyr.c
index bce7394b..d2fbf4d0 100644
--- a/print-zephyr.c
+++ b/print-zephyr.c
@@ -134,7 +134,7 @@ str_to_lower(const char *string)
 
     zb_string = z_buf;
     while (*zb_string) {
-	*zb_string = tolower((unsigned char)(*zb_string));
+	*zb_string = ND_TOLOWER((unsigned char)(*zb_string));
 	zb_string++;
     }
 
diff --git a/util-print.c b/util-print.c
index 5a8c8fe5..aa6cf87f 100644
--- a/util-print.c
+++ b/util-print.c
@@ -892,7 +892,7 @@ txtproto_print(netdissect_options *ndo, const u_char *pptr, u_int len,
 
 	/* Capitalize the protocol name */
 	for (pnp = protoname; *pnp != '\0'; pnp++)
-		ND_PRINT("%c", toupper((u_char)*pnp));
+		ND_PRINT("%c", ND_TOUPPER((u_char)*pnp));
 
 	if (print_this) {
 		/*