WL#916: Unicode collations for some languages

15 files changed, 33 insertions, 6 deletions

diff --git a/strings/ctype-big5.c b/strings/ctype-big5.c
index f024fa0cc14..7a3c4503d74 100644
--- a/strings/ctype-big5.c
+++ b/strings/ctype-big5.c
@@ -6313,6 +6313,7 @@ CHARSET_INFO my_charset_big5_chinese_ci=
     "big5",		/* cs name    */
     "big5_chinese_ci",	/* name       */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_big5,
     to_lower_big5,
     to_upper_big5,
@@ -6339,6 +6340,7 @@ CHARSET_INFO my_charset_big5_bin=
     "big5",		/* cs name    */
     "big5_bin",		/* name       */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_big5,
     to_lower_big5,
     to_upper_big5,
diff --git a/strings/ctype-bin.c b/strings/ctype-bin.c
index 7cac8c7c337..48323018cca 100644
--- a/strings/ctype-bin.c
+++ b/strings/ctype-bin.c
@@ -378,6 +378,7 @@ CHARSET_INFO my_charset_bin =
     "binary",			/* cs name    */
     "binary",			/* name          */
     "",				/* comment       */
+    NULL,			/* tailoring     */
     ctype_bin,			/* ctype         */
     bin_char_array,		/* to_lower      */
     bin_char_array,		/* to_upper      */
diff --git a/strings/ctype-czech.c b/strings/ctype-czech.c
index 2eb2fac46e9..dede737f361 100644
--- a/strings/ctype-czech.c
+++ b/strings/ctype-czech.c
@@ -589,6 +589,7 @@ CHARSET_INFO my_charset_latin2_czech_ci =
     "latin2",				/* cs name    */
     "latin2_czech_cs",			/* name */
     "",					/* comment    */
+    NULL,				/* tailoring */
     ctype_czech,
     to_lower_czech,
     to_upper_czech,
diff --git a/strings/ctype-euc_kr.c b/strings/ctype-euc_kr.c
index 8f955c15a73..2d4c68978a3 100644
--- a/strings/ctype-euc_kr.c
+++ b/strings/ctype-euc_kr.c
@@ -8681,6 +8681,7 @@ CHARSET_INFO my_charset_euckr_korean_ci=
     "euckr",		/* cs name    */
     "euckr_korean_ci",	/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_euc_kr,
     to_lower_euc_kr,
     to_upper_euc_kr,
@@ -8707,6 +8708,7 @@ CHARSET_INFO my_charset_euckr_bin=
     "euckr",		/* cs name    */
     "euckr_bin",	/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_euc_kr,
     to_lower_euc_kr,
     to_upper_euc_kr,
diff --git a/strings/ctype-gb2312.c b/strings/ctype-gb2312.c
index b76511fc4f3..49ca736a3c2 100644
--- a/strings/ctype-gb2312.c
+++ b/strings/ctype-gb2312.c
@@ -5732,6 +5732,7 @@ CHARSET_INFO my_charset_gb2312_chinese_ci=
     "gb2312",		/* cs name    */
     "gb2312_chinese_ci",/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_gb2312,
     to_lower_gb2312,
     to_upper_gb2312,
@@ -5757,6 +5758,7 @@ CHARSET_INFO my_charset_gb2312_bin=
     "gb2312",		/* cs name    */
     "gb2312_bin",	/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_gb2312,
     to_lower_gb2312,
     to_upper_gb2312,
diff --git a/strings/ctype-gbk.c b/strings/ctype-gbk.c
index 0dc00a73fa3..0273feb4c2c 100644
--- a/strings/ctype-gbk.c
+++ b/strings/ctype-gbk.c
@@ -9962,6 +9962,7 @@ CHARSET_INFO my_charset_gbk_chinese_ci=
     "gbk",		/* cs name    */
     "gbk_chinese_ci",	/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_gbk,
     to_lower_gbk,
     to_upper_gbk,
@@ -9987,6 +9988,7 @@ CHARSET_INFO my_charset_gbk_bin=
     "gbk",		/* cs name    */
     "gbk_bin",		/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_gbk,
     to_lower_gbk,
     to_upper_gbk,
diff --git a/strings/ctype-latin1.c b/strings/ctype-latin1.c
index 0b439964c7c..fe39303e2ac 100644
--- a/strings/ctype-latin1.c
+++ b/strings/ctype-latin1.c
@@ -412,6 +412,7 @@ CHARSET_INFO my_charset_latin1=
     "latin1",				/* cs name    */
     "latin1_swedish_ci",		/* name      */
     "",					/* comment   */
+    NULL,				/* tailoring */
     ctype_latin1,
     to_lower_latin1,
     to_upper_latin1,
@@ -690,6 +691,7 @@ CHARSET_INFO my_charset_latin1_german2_ci=
   "latin1",				/* cs name    */
   "latin1_german2_ci",			/* name      */
   "",					/* comment   */
+  NULL,					/* tailoring */
   ctype_latin1,
   to_lower_latin1,
   to_upper_latin1,
@@ -715,6 +717,7 @@ CHARSET_INFO my_charset_latin1_bin=
   "latin1",				/* cs name    */
   "latin1_bin",				/* name      */
   "",					/* comment   */
+  NULL,					/* tailoring */
   ctype_latin1,
   to_lower_latin1,
   to_upper_latin1,
diff --git a/strings/ctype-sjis.c b/strings/ctype-sjis.c
index 72666175a1f..22c58360348 100644
--- a/strings/ctype-sjis.c
+++ b/strings/ctype-sjis.c
@@ -4579,6 +4579,7 @@ CHARSET_INFO my_charset_sjis_japanese_ci=
     "sjis",		/* cs name    */
     "sjis_japanese_ci",	/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_sjis,
     to_lower_sjis,
     to_upper_sjis,
@@ -4604,6 +4605,7 @@ CHARSET_INFO my_charset_sjis_bin=
     "sjis",		/* cs name    */
     "sjis_bin",		/* name */
     "",			/* comment    */
+    NULL,		/* tailoring */
     ctype_sjis,
     to_lower_sjis,
     to_upper_sjis,
diff --git a/strings/ctype-tis620.c b/strings/ctype-tis620.c
index e2a138300c3..b2b1ab98352 100644
--- a/strings/ctype-tis620.c
+++ b/strings/ctype-tis620.c
@@ -951,6 +951,7 @@ CHARSET_INFO my_charset_tis620_thai_ci=
     "tis620",		/* cs name    */
     "tis620_thai_ci",	/* name      */
     "",			/* comment   */
+    NULL,		/* tailoring */
     ctype_tis620,
     to_lower_tis620,
     to_upper_tis620,
@@ -976,6 +977,7 @@ CHARSET_INFO my_charset_tis620_bin=
     "tis620",		/* cs name    */
     "tis620_bin",	/* name      */
     "",			/* comment   */
+    NULL,		/* tailoring */
     ctype_tis620,
     to_lower_tis620,
     to_upper_tis620,
diff --git a/strings/ctype-uca.c b/strings/ctype-uca.c
index 81073d47554..75e2c06eec2 100644
--- a/strings/ctype-uca.c
+++ b/strings/ctype-uca.c
@@ -7055,6 +7055,7 @@ CHARSET_INFO my_charset_ucs2_general_uca=
     "ucs2",		/* cs name    */
     "ucs2_general_uca",	/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     NULL,		/* ctype        */
     NULL,		/* to_lower     */
     NULL,		/* to_upper     */
diff --git a/strings/ctype-ucs2.c b/strings/ctype-ucs2.c
index 67340fdd4f4..d1ba63b8b84 100644
--- a/strings/ctype-ucs2.c
+++ b/strings/ctype-ucs2.c
@@ -1431,6 +1431,7 @@ CHARSET_INFO my_charset_ucs2_general_ci=
     "ucs2",		/* cs name    */
     "ucs2_general_ci",	/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     ctype_ucs2,		/* ctype        */
     to_lower_ucs2,	/* to_lower     */
     to_upper_ucs2,	/* to_upper     */
@@ -1456,6 +1457,7 @@ CHARSET_INFO my_charset_ucs2_bin=
     "ucs2",		/* cs name    */
     "ucs2_bin",		/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     ctype_ucs2,		/* ctype        */
     to_lower_ucs2,	/* to_lower     */
     to_upper_ucs2,	/* to_upper     */
diff --git a/strings/ctype-ujis.c b/strings/ctype-ujis.c
index fd3692553be..668dc7beb8b 100644
--- a/strings/ctype-ujis.c
+++ b/strings/ctype-ujis.c
@@ -8468,6 +8468,7 @@ CHARSET_INFO my_charset_ujis_japanese_ci=
     "ujis",		/* cs name    */
     "ujis_japanese_ci",	/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     ctype_ujis,
     to_lower_ujis,
     to_upper_ujis,
@@ -8494,6 +8495,7 @@ CHARSET_INFO my_charset_ujis_bin=
     "ujis",		/* cs name    */
     "ujis_bin",		/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     ctype_ujis,
     to_lower_ujis,
     to_upper_ujis,
diff --git a/strings/ctype-utf8.c b/strings/ctype-utf8.c
index 09b918b0777..2d0feb1c890 100644
--- a/strings/ctype-utf8.c
+++ b/strings/ctype-utf8.c
@@ -2090,6 +2090,7 @@ CHARSET_INFO my_charset_utf8_general_ci=
     "utf8",		/* cs name      */
     "utf8_general_ci",	/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     ctype_utf8,		/* ctype        */
     to_lower_utf8,	/* to_lower     */
     to_upper_utf8,	/* to_upper     */
@@ -2116,6 +2117,7 @@ CHARSET_INFO my_charset_utf8_bin=
     "utf8",		/* cs name      */
     "utf8_bin",		/* name         */
     "",			/* comment      */
+    NULL,		/* tailoring    */
     ctype_utf8,		/* ctype        */
     to_lower_utf8,	/* to_lower     */
     to_upper_utf8,	/* to_upper     */
diff --git a/strings/ctype-win1250ch.c b/strings/ctype-win1250ch.c
index 2eefb570170..bb287eb695e 100644
--- a/strings/ctype-win1250ch.c
+++ b/strings/ctype-win1250ch.c
@@ -623,6 +623,7 @@ CHARSET_INFO my_charset_cp1250_czech_ci =
   "cp1250",			/* cs name    */
   "cp1250_czech_cs",		/* name      */
   "",				/* comment   */
+  NULL,				/* tailoring */
   ctype_win1250ch,
   to_lower_win1250ch,
   to_upper_win1250ch,
diff --git a/strings/ctype.c b/strings/ctype.c
index 44bf20ada5c..4454d3c45e1 100644
--- a/strings/ctype.c
+++ b/strings/ctype.c
@@ -124,6 +124,7 @@ static struct my_cs_file_section_st * cs_file_sec(const char *attr, uint len)
 }
 
 #define MY_CS_CSDESCR_SIZE	64
+#define MY_CS_TAILORING_SIZE	128
 
 typedef struct my_cs_file_info
 {
@@ -135,7 +136,8 @@ typedef struct my_cs_file_info
   uchar  sort_order[MY_CS_SORT_ORDER_TABLE_SIZE];
   uint16 tab_to_uni[MY_CS_TO_UNI_TABLE_SIZE];
   char   comment[MY_CS_CSDESCR_SIZE];
-  size_t sort_order_length;
+  char   tailoring[MY_CS_TAILORING_SIZE];
+  size_t tailoring_length;
   CHARSET_INFO cs;
   int (*add_collation)(CHARSET_INFO *cs);
 } MY_CHARSET_LOADER;
@@ -186,7 +188,7 @@ static int cs_enter(MY_XML_PARSER *st,const char *attr, uint len)
     bzero(&i->cs,sizeof(i->cs));
   
   if (s && (s->state == _CS_COLLATION))
-    i->sort_order_length= 0;
+    i->tailoring_length= 0;
 
   return MY_XML_OK;
 }
@@ -283,12 +285,12 @@ static int cs_value(MY_XML_PARSER *st,const char *attr, uint len)
       */
       char arg[16];
       const char *cmd[]= {"&","<","<<","<<<"};
-      i->cs.sort_order= i->sort_order;
+      i->cs.tailoring= i->tailoring;
       mstr(arg,attr,len,sizeof(arg)-1);
-      if (i->sort_order_length + 20 < sizeof(i->sort_order))
+      if (i->tailoring_length + 20 < sizeof(i->tailoring))
       {
-        char *dst= i->sort_order_length + i->sort_order;
-        i->sort_order_length+= sprintf(dst," %s %s",cmd[state-_CS_RESET],arg);
+        char *dst= i->tailoring_length + i->tailoring;
+        i->tailoring_length+= sprintf(dst," %s %s",cmd[state-_CS_RESET],arg);
       }
     }
   }