From be9157f3e9b156dd62c73637ce20e2908ab00eda Mon Sep 17 00:00:00 2001
From: "Dirk-Jan C. Binnema" <djcb@djcbsoftware.nl>
Date: Fri, 12 Nov 2010 22:36:13 +0200
Subject: [PATCH] * mu-msg-str.c: add Latin-Extended-A for normalization,
 update unit test

---
 src/mu-msg-str.c            | 331 ++++++++++++++++++++++++++++--------
 src/tests/test-mu-msg-str.c |   1 +
 2 files changed, 261 insertions(+), 71 deletions(-)

diff --git a/src/mu-msg-str.c b/src/mu-msg-str.c
index 65d46fa0..c75f23ef 100644
--- a/src/mu-msg-str.c
+++ b/src/mu-msg-str.c
@@ -26,6 +26,8 @@
 
 
 
+
+
 char*
 mu_msg_str_normalize (const char *str, gboolean downcase)
 {
@@ -41,92 +43,279 @@ mu_msg_str_normalize (const char *str, gboolean downcase)
 	output = g_new0 (char, 2 * strlen(str));
 	
 	for (i = 0, cur = (const guchar*)str; *cur; ++cur) {
-		if (*cur != 0xc3) { /* != latin-1 supplement? */
+		if (*cur < 0xc3 || *cur > 0xc5) { 
 			output[i++] = *cur;
 			continue; 
 		}
-		++cur;
 
-		switch (*cur) {
+		if (*cur == 0xc3) { /* latin-1 supplement */
+			++cur;
+			switch (*cur) {
+				
+			case 0x80:
+			case 0x81:
+			case 0x82:
+			case 0x83:
+			case 0x84:
+			case 0x85: output[i++] = 'A'; break;
 
-		case 0x80:
-		case 0x81:
-		case 0x82:
-		case 0x83:
-		case 0x84:
-		case 0x85: output[i++] = 'A'; break;
-
-		case 0x86: output[i++] = 'A'; output[i++] = 'e'; break;
-		case 0x87: output[i++] = 'C'; break;
+			case 0x86: output[i++] = 'A'; output[i++] = 'e'; break;
+			case 0x87: output[i++] = 'C'; break;
+				
+			case 0x88:
+			case 0x89:
+			case 0x8a:
+			case 0x8b: output[i++] = 'E'; break;
+				
+			case 0x8c:
+			case 0x8d:
+			case 0x8e:
+			case 0x8f: output[i++] = 'I'; break;
+				
+			case 0x90: output[i++] = 'D'; break;
+			case 0x91: output[i++] = 'N'; break;
+				
+			case 0x92:
+			case 0x93:
+			case 0x94:
+			case 0x95:
+			case 0x96: output[i++] = 'O'; break;
+				
+			case 0x99:
+			case 0x9a:
+			case 0x9b:
+			case 0x9c: output[i++] = 'U'; break;
 			
-		case 0x88:
-		case 0x89:
-		case 0x8a:
-		case 0x8b: output[i++] = 'E'; break;
+			case 0x9d: output[i++] = 'Y'; break;
+			case 0x9e: output[i++] = 'T'; output[i++] = 'h'; break;
+			case 0x9f: output[i++] = 's'; output[i++] = 's'; break;
+				
+			case 0xa0:
+			case 0xa1:
+			case 0xa2:
+			case 0xa3:
+			case 0xa4:
+			case 0xa5: output[i++] = 'a'; break;
+				
+			case 0xa6: output[i++] = 'a'; output[i++] = 'e'; break;
+			case 0xa7: output[i++] = 'c'; break;
+				
+			case 0xa8:
+			case 0xa9:
+			case 0xaa:
+			case 0xab: output[i++] = 'e'; break;
+				
+			case 0xac:
+			case 0xad:
+			case 0xae:
+			case 0xaf: output[i++] = 'i'; break;
+				
+			case 0xb0: output[i++] = 'd'; break;
+			case 0xb1: output[i++] = 'n'; break;
+				
+			case 0xb2:
+			case 0xb3:
+			case 0xb4:
+			case 0xb5:
+			case 0xb6: output[i++] = 'o'; break;
+				
+			case 0xb9:
+			case 0xba:
+			case 0xbb:
+			case 0xbc: output[i++] = 'u'; break;
+				
+			case 0xbd: output[i++] = 'y'; break;
+			case 0xbe: output[i++] = 't'; output[i++] = 'h'; break;
+			case 0xbf: output[i++] = 'y'; break;
+				
+			default:
+				output[i++] = *cur;
+			}
+		} else if (*cur == 0xc4) {  /* Latin Extended-A (0x04) */
+			++cur;
+			switch (*cur) {	
+			case 0x80:
+			case 0x82:
+			case 0x84: output[i++] = 'A'; break;
 
-		case 0x8c:
-		case 0x8d:
-		case 0x8e:
-		case 0x8f: output[i++] = 'I'; break;
+			case 0x86:
+			case 0x88:
+			case 0x8a:
+			case 0x8c: output[i++] = 'C'; break;
 
-		case 0x90: output[i++] = 'D'; break;
-		case 0x91: output[i++] = 'N'; break;
-			
-		case 0x92:
-		case 0x93:
-		case 0x94:
-		case 0x95:
-		case 0x96: output[i++] = 'O'; break;
+			case 0x8e:
+			case 0x90: output[i++] = 'D'; break;
 
-		case 0x99:
-		case 0x9a:
-		case 0x9b:
-		case 0x9c: output[i++] = 'U'; break;
-			
-		case 0x9d: output[i++] = 'Y'; break;
-		case 0x9e: output[i++] = 'T'; output[i++] = 'h'; break;
-		case 0x9f: output[i++] = 's'; output[i++] = 's'; break;
-			
-		case 0xa0:
-		case 0xa1:
-		case 0xa2:
-		case 0xa3:
-		case 0xa4:
-		case 0xa5: output[i++] = 'a'; break;
+			case 0x92:
+			case 0x94:
+			case 0x96:
+			case 0x98:
+			case 0x9a: output[i++] = 'E'; break;
 
-		case 0xa6: output[i++] = 'a'; output[i++] = 'e'; break;
-		case 0xa7: output[i++] = 'c'; break;
-			
-		case 0xa8:
-		case 0xa9:
-		case 0xaa:
-		case 0xab: output[i++] = 'e'; break;
+			case 0x9c:
+			case 0x9e:
+			case 0xa0:
+			case 0xa2: output[i++] = 'G'; break;
 
-		case 0xac:
-		case 0xad:
-		case 0xae:
-		case 0xaf: output[i++] = 'i'; break;
+			case 0xa4:
+			case 0xa6: output[i++] = 'H'; break;
 
-		case 0xb0: output[i++] = 'd'; break;
-		case 0xb1: output[i++] = 'n'; break;
-			
-		case 0xb2:
-		case 0xb3:
-		case 0xb4:
-		case 0xb5:
-		case 0xb6: output[i++] = 'o'; break;
+			case 0xa8:
+			case 0xaa:
+			case 0xac:
+			case 0xae:
+			case 0xb0: output[i++] = 'I'; break;
+	
+			case 0xb2: output[i++] = 'I'; output[i++] = 'J'; break;		
 
-		case 0xb9:
-		case 0xba:
-		case 0xbb:
-		case 0xbc: output[i++] = 'u'; break;
+			case 0xb4: output[i++] = 'J'; break;
 
-		case 0xbd: output[i++] = 'y'; break;
-		case 0xbe: output[i++] = 't'; output[i++] = 'h'; break;
-		case 0xbf: output[i++] = 'y'; break;
+			case 0xb6: output[i++] = 'K'; break;
 
-		default:
-			output[i++] = *cur;
+			case 0xb9:
+			case 0xbb:
+			case 0xbd:
+			case 0xbf: output[i++] = 'L'; break;
+				
+			case 0x81:
+			case 0x83:
+			case 0x85: output[i++] = 'a'; break;
+
+			case 0x87:
+			case 0x89:
+			case 0x8b:
+			case 0x8d: output[i++] = 'c'; break;
+
+			case 0x8f:
+			case 0x91: output[i++] = 'd'; break;
+
+			case 0x93:
+			case 0x95:
+			case 0x97:
+			case 0x99:
+			case 0x9b: output[i++] = 'e'; break;
+
+			case 0x9d:
+			case 0x9f:
+			case 0xa1:
+			case 0xa: output[i++] = 'g'; break;
+
+			case 0xa5:
+			case 0xa7: output[i++] = 'h'; break;
+
+			case 0xa9:
+			case 0xab:
+			case 0xad:
+			case 0xaf:
+			case 0xb1: output[i++] = 'i'; break;
+
+			case 0xb3: output[i++] = 'i'; output[i++] = 'j'; break;		
+
+			case 0xb5: output[i++] = 'j'; break;
+
+			case 0xb7:
+			case 0xb8: output[i++] = 'k'; break;
+
+			case 0xba:
+			case 0xbc:
+			case 0xbe: output[i++] = 'l'; break;
+
+			default:
+				output[i++] = *cur;
+			}
+
+		} else { /* Latin Extended-A (0xc5) */
+			++cur;
+			switch (*cur) {
+			case 0x81: output[i++] = 'L'; break;
+				
+			case 0x83:
+			case 0x85:
+			case 0x87: output[i++] = 'N'; break;
+
+			case 0x8c:
+			case 0x8e:
+			case 0x90: output[i++] = 'O'; break;
+				
+			case 0x92: output[i++] = 'O'; output[i++] = 'e'; break;		
+
+			case 0x94:
+			case 0x96:
+			case 0x98: output[i++] = 'R'; break;
+
+			case 0x9a:
+			case 0x9c:
+			case 0x9e:
+			case 0xa0: output[i++] = 'S'; break;
+
+			case 0xa2:
+			case 0xa4:
+			case 0xa6: output[i++] = 'T'; break;
+
+			case 0xa8:
+			case 0xaa:
+			case 0xac:
+			case 0xae:
+			case 0xb0:	
+			case 0xb2: output[i++] = 'U'; break;				
+
+			case 0xb4: output[i++] = 'W'; break;				
+
+			case 0xb6:
+			case 0xb8:
+				output[i++] = 'Y'; break;				
+
+			case 0xb9:
+			case 0xbb: 
+			case 0xbd: output[i++] = 'Z'; break;
+				
+			case 0x80:
+			case 0x82: output[i++] = 'l'; break;
+
+			case 0x84:
+			case 0x86:
+			case 0x88:
+			case 0x89:
+			case 0x8a:
+			case 0x8b: output[i++] = 'n'; break;
+
+			case 0x8d:
+			case 0x8f:
+			case 0x91: output[i++] = 'o'; break;
+
+			case 0x93: output[i++] = 'o'; output[i++] = 'e'; break;		
+
+			case 0x95:
+			case 0x97:
+			case 0x99: output[i++] = 'r'; break;
+
+			case 0x9b:
+			case 0x9d:
+			case 0x9f:
+			case 0xa1: output[i++] = 's'; break;
+
+			case 0xa3:
+			case 0xa5:
+			case 0xa7: output[i++] = 't'; break;
+
+			case 0xa9:
+			case 0xab:
+			case 0xad:
+			case 0xaf:
+			case 0xb1:	
+			case 0xb3: output[i++] = 'u'; break;
+				
+			case 0xb5: output[i++] = 'w'; break;
+
+			case 0xb7: output[i++] = 'y'; break;
+
+			case 0xba:
+			case 0xbc: 
+			case 0xbe: output[i++] = 'z'; break;
+
+			case 0xbf: output[i++] = 's'; break;
+				
+			}
 		}
 	}
 
diff --git a/src/tests/test-mu-msg-str.c b/src/tests/test-mu-msg-str.c
index 70e7794e..691606a0 100644
--- a/src/tests/test-mu-msg-str.c
+++ b/src/tests/test-mu-msg-str.c
@@ -137,6 +137,7 @@ test_mu_msg_str_normalize_01 (void)
 		{ "dantès", "dantes"}, 
 		{ "foo", "foo" },
 		{ "Föö", "foo" },
+		{ "číslo", "cislo" },
 		{ "hÆvý mëÐal ümláõt", "haevy medal umlaot"}
 	};