* improve support for non-latin languages (cyrillic etc.) (WIP)

- change the various escaping / normalization functions to better deal with
    non-ascii, non-latin languages, such as Russian.

    It seems. now we can match 'Тесла' or 'Аркона' without problem.

  - added unit test.

  - WIP -- needs more testing.

src/mu-msg-fields.c

@@ -139,7 +139,7 @@ static const MuMsgField FIELD_DATA[] = {
 		MU_MSG_FIELD_ID_FILE,
 		MU_MSG_FIELD_TYPE_STRING,
 		"file" , 'j', 'J',
-		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_NORMALIZE |
+		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_XAPIAN_ESCAPE |
 		FLAG_DONT_CACHE | FLAG_XAPIAN_PREFIX_ONLY
 	},
 
@@ -164,7 +164,8 @@ static const MuMsgField FIELD_DATA[] = {
 		MU_MSG_FIELD_TYPE_STRING,
 		"path", 'l', 'L',   /* 'l' for location */
 		FLAG_GMIME | FLAG_XAPIAN_VALUE |
-		FLAG_XAPIAN_BOOLEAN  | FLAG_XAPIAN_PREFIX_ONLY
+		FLAG_XAPIAN_BOOLEAN  | FLAG_XAPIAN_PREFIX_ONLY |
+		FLAG_XAPIAN_ESCAPE
 	},
 
 	{
@@ -172,7 +173,7 @@ static const MuMsgField FIELD_DATA[] = {
 		MU_MSG_FIELD_TYPE_STRING,
 		"maildir", 'm', 'M',
 		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_XAPIAN_VALUE |
-		FLAG_NORMALIZE | FLAG_XAPIAN_ESCAPE | FLAG_XAPIAN_PREFIX_ONLY
+		FLAG_XAPIAN_ESCAPE | FLAG_XAPIAN_PREFIX_ONLY
 	},
 
 
@@ -204,7 +205,7 @@ static const MuMsgField FIELD_DATA[] = {
 		MU_MSG_FIELD_TYPE_STRING,
 		"subject", 's', 'S',
 		FLAG_GMIME | FLAG_XAPIAN_INDEX | FLAG_XAPIAN_VALUE |
-		FLAG_XAPIAN_TERM | FLAG_NORMALIZE | FLAG_XAPIAN_ESCAPE
+		FLAG_XAPIAN_TERM  | FLAG_XAPIAN_ESCAPE
 	},
 
 	{
@@ -234,7 +235,7 @@ static const MuMsgField FIELD_DATA[] = {
 		MU_MSG_FIELD_TYPE_STRING_LIST,
 		"tag", 'x', 'X',
 		FLAG_GMIME | FLAG_XAPIAN_TERM | FLAG_XAPIAN_PREFIX_ONLY |
-		FLAG_NORMALIZE | FLAG_XAPIAN_ESCAPE
+		FLAG_XAPIAN_ESCAPE
 	},
 
 	{	/* special, internal field, to get a unique key */