FULLTEXT: correct charset support (UTF included, UCS2 - not)

code cleanup


include/m_ctype.h:
  my_mbcharlen_8bit() { return 1 }
mysql-test/r/fulltext.result:
  fulltext on UTF
mysql-test/t/fulltext.test:
  fulltext on UTF
sql/item_cmpfunc.h:
  cleanup
sql/sql_table.cc:
  FULLTEXT: UCS2 is not allowed
sql/sql_yacc.yy:
  FULLTEXT: code cleanup
strings/ctype-bin.c:
  my_mbcharlen_8bit() { return 1 }
strings/ctype-latin1.c:
  my_mbcharlen_8bit() { return 1 }
strings/ctype-simple.c:
  my_mbcharlen_8bit() { return 1 }
strings/ctype-tis620.c:
  my_mbcharlen_8bit() { return 1 }
strings/ctype-utf8.c:
  hack: (to be fixed properly later) all multi-byte sequences are considered isalpha() now

5 files changed, 24 insertions, 18 deletions

diff --git a/strings/ctype-bin.c b/strings/ctype-bin.c
index 340084ad848..cd1b1399506 100644
--- a/strings/ctype-bin.c
+++ b/strings/ctype-bin.c
@@ -118,6 +118,12 @@ static int my_strcasecmp_bin(CHARSET_INFO * cs __attribute__((unused)),
   return strcmp(s,t);
 }
 
+int my_mbcharlen_8bit(CHARSET_INFO *cs __attribute__((unused)),
+                             uint c __attribute__((unused)))
+{
+  return 1;
+}
+
 static int my_mb_wc_bin(CHARSET_INFO *cs __attribute__((unused)),
 		  my_wc_t *wc,
 		  const unsigned char *str,
@@ -264,12 +270,12 @@ static int my_strnxfrm_bin(CHARSET_INFO *cs __attribute__((unused)),
 
 static
 uint my_instr_bin(CHARSET_INFO *cs __attribute__((unused)),
-                 const char *big,   uint b_length, 
+                 const char *big,   uint b_length,
 		 const char *small, uint s_length,
 		 my_match_t *match, uint nmatch)
 {
   register const uchar *str, *search, *end, *search_end;
-  
+
   if (s_length <= b_length)
   {
     if (!s_length)
@@ -282,32 +288,32 @@ uint my_instr_bin(CHARSET_INFO *cs __attribute__((unused)),
       }
       return 1;		/* Empty string is always found */
     }
-    
+
     str= (const uchar*) big;
     search= (const uchar*) small;
     end= (const uchar*) big+b_length-s_length+1;
     search_end= (const uchar*) small + s_length;
-    
+
 skipp:
     while (str != end)
     {
       if ( (*str++) == (*search))
       {
 	register const uchar *i,*j;
-	
-	i= str; 
+
+	i= str;
 	j= search+1;
-	
+
 	while (j != search_end)
 	  if ((*i++) != (*j++))
             goto skipp;
-        
+
         if (nmatch > 0)
 	{
 	  match[0].beg= 0;
 	  match[0].end= str- (const uchar*)big-1;
 	  match[0].mblen= match[0].end;
-	  
+
 	  if (nmatch > 1)
 	  {
 	    match[1].beg= match[0].end;
@@ -338,7 +344,7 @@ MY_COLLATION_HANDLER my_collation_8bit_bin_handler =
 static MY_CHARSET_HANDLER my_charset_handler=
 {
     NULL,			/* ismbchar      */
-    NULL,			/* mbcharlen     */
+    my_mbcharlen_8bit,		/* mbcharlen     */
     my_numchars_8bit,
     my_charpos_8bit,
     my_lengthsp_8bit,
diff --git a/strings/ctype-latin1.c b/strings/ctype-latin1.c
index a8a5329f844..15798abb85b 100644
--- a/strings/ctype-latin1.c
+++ b/strings/ctype-latin1.c
@@ -170,14 +170,14 @@ int my_wc_mb_latin1(CHARSET_INFO *cs __attribute__((unused)),
 {
   if (str >= end)
     return MY_CS_TOOSMALL;
-  
+
   return ((wc < 256) && ((str[0]=uni_latin1[wc]) || (!wc))) ? 1 : MY_CS_ILUNI;
 }
 
 static MY_CHARSET_HANDLER my_charset_handler=
 {
     NULL,
-    NULL,
+    my_mbcharlen_8bit,
     my_numchars_8bit,
     my_charpos_8bit,
     my_lengthsp_8bit,
diff --git a/strings/ctype-simple.c b/strings/ctype-simple.c
index f85ce5e7a2b..ed1d2c77049 100644
--- a/strings/ctype-simple.c
+++ b/strings/ctype-simple.c
@@ -1093,7 +1093,7 @@ skipp:
 MY_CHARSET_HANDLER my_charset_8bit_handler=
 {
     NULL,			/* ismbchar      */
-    NULL,			/* mbcharlen     */
+    my_mbcharlen_8bit,		/* mbcharlen     */
     my_numchars_8bit,
     my_charpos_8bit,
     my_lengthsp_8bit,
diff --git a/strings/ctype-tis620.c b/strings/ctype-tis620.c
index a4d8a7d1f79..fd5e58ad8a7 100644
--- a/strings/ctype-tis620.c
+++ b/strings/ctype-tis620.c
@@ -717,7 +717,7 @@ static MY_COLLATION_HANDLER my_collation_ci_handler =
 static MY_CHARSET_HANDLER my_charset_handler=
 {
     NULL,		/* ismbchar  */
-    NULL,		/* mbcharlen */
+    my_mbcharlen_8bit,	/* mbcharlen */
     my_numchars_8bit,
     my_charpos_8bit,
     my_lengthsp_8bit,
diff --git a/strings/ctype-utf8.c b/strings/ctype-utf8.c
index 3ede1aa26f6..b5716c53ea2 100644
--- a/strings/ctype-utf8.c
+++ b/strings/ctype-utf8.c
@@ -1540,10 +1540,10 @@ static uchar ctype_utf8[] = {
     0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
     0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
     0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
-    0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
-    0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
-    0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
-    0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0
+    0,  0,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,
+    3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,
+    3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,
+    3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  3,  0
 };
 
 static uchar to_lower_utf8[] = {