From 416bede7f778b56239f5b6bf41519fa9a54d1332 Mon Sep 17 00:00:00 2001
From: magnum <john.magnum@hushmail.com>
Date: Sat, 24 Oct 2015 10:56:38 +0200
Subject: [PATCH] Mask mode: Fix incorrect use to CTYPE macros, so ?W will
 case- toggle according to our active encoding - not just ASCII.

Closes #1847, as other uses of ctype macros was audited too and
no other needed fixing.
---
 src/mask.c    | 14 +++++++-------
 src/unicode.c | 11 ++++++++++-
 src/unicode.h | 12 ++++++++++--
 3 files changed, 27 insertions(+), 10 deletions(-)

diff --git a/src/mask.c b/src/mask.c
index 36ef20d815..bd7fb56a94 100644
--- a/src/mask.c
+++ b/src/mask.c
@@ -2145,21 +2145,21 @@ int do_mask_crack(const char *extern_key)
 		i = 0;
 		while(template_key_offsets[i] != -1) {
 			int offset = template_key_offsets[i] & 0xffff;
-			unsigned char is_lower =  (template_key_offsets[i++] >> 16)
-				== 'w';
+			unsigned char toggle =  (template_key_offsets[i++] >> 16) == 'W';
 			int cpy_len = max_keylen - offset;
+
 			cpy_len = cpy_len > key_len ? key_len : cpy_len;
-			if (is_lower)
-			memcpy(template_key + offset, extern_key, cpy_len);
+			if (!toggle)
+				memcpy(template_key + offset, extern_key, cpy_len);
 			else {
 				int z;
 				for (z = 0; z < cpy_len; ++z) {
-					if (islower(ARCH_INDEX(extern_key[z])))
+					if (enc_islower(extern_key[z]))
 						template_key[offset + z] =
-							toupper(ARCH_INDEX(extern_key[z]));
+							enc_toupper(extern_key[z]);
 					else
 						template_key[offset + z] =
-							tolower(ARCH_INDEX(extern_key[z]));
+							enc_tolower(extern_key[z]);
 				}
 			}
 		}
diff --git a/src/unicode.c b/src/unicode.c
index 30921a06be..27332a85fe 100644
--- a/src/unicode.c
+++ b/src/unicode.c
@@ -49,7 +49,9 @@ UTF16 CP_to_Unicode[0x100];
 static UTF8 CP_from_Unicode[0x10000];
 
 UTF8 CP_up[0x100];
+UTF8 CP_ups[0x100];
 UTF8 CP_down[0x100];
+UTF8 CP_lows[0x100];
 
 #ifndef UNICODE_NO_OPTIONS
 static int UnicodeType = -1;
@@ -1044,7 +1046,7 @@ void initUnicode(int type)
 {
 
 #ifndef UNICODE_NO_OPTIONS
-	unsigned i;
+	unsigned i, j;
 	unsigned char *cpU, *cpL, *Sep, *Letter;
 	unsigned char *pos;
 	int encoding;
@@ -1546,6 +1548,13 @@ void initUnicode(int type)
 		CP_down[0x91] = 0x91;
 	}
 #endif
+	j = 0;
+	for (i = 0; i < 256; i++) {
+		if (CP_up[i] != CP_down[i]) {
+			CP_ups[j] = CP_up[i];
+			CP_lows[j++] = CP_down[i];
+		}
+	}
 	return;
 }
 
diff --git a/src/unicode.h b/src/unicode.h
index 4e504a440c..6211793188 100644
--- a/src/unicode.h
+++ b/src/unicode.h
@@ -246,13 +246,21 @@ extern char *enc_strupper(char *s);
 extern UTF16 CP_to_Unicode[0x100];
 
 /* Used by various formats uc/lc */
-extern UTF8  CP_up[0x100];
-extern UTF8  CP_down[0x100];
+extern UTF8 CP_up[0x100]; /* upper-case lookup table */
+extern UTF8 CP_ups[0x100]; /* all upper-case letters */
+extern UTF8 CP_down[0x100]; /* lower-case lookup table */
+extern UTF8 CP_lows[0x100]; /* all lower-case letters */
 
 /* Used by single.c and loader.c */
 extern UTF8 CP_isLetter[0x100];
 extern UTF8 CP_isSeparator[0x100];
 
+/* These are encoding-aware but not LC_CTYPE */
+#define enc_islower(c) (pers_opts.internal_cp == ASCII ? (c > 'a' && c < 'z') : (strchr((char*)CP_lows, ARCH_INDEX(c)) != NULL))
+#define enc_isupper(c) (pers_opts.internal_cp == ASCII ? (c > 'A' && c < 'Z') : ((strchr((char*)CP_ups, ARCH_INDEX(c)) != NULL))
+#define enc_tolower(c) (char)CP_down[ARCH_INDEX(c)]
+#define enc_toupper(c) (char)CP_up[ARCH_INDEX(c)]
+
 /* Conversion between encoding names and integer id */
 extern int cp_name2id(char *encoding);
 extern char *cp_id2name(int encoding);