Merge svojtovich@bk-internal.mysql.com:/home/bk/mysql-5.1

into  may.pils.ru:/home/svoj/devel/mysql/BUG18036/mysql-5.1

include/my_global.h

@@ -797,6 +797,9 @@ typedef SOCKET_SIZE_TYPE size_socket;
 #define DBL_MAX		1.79769313486231470e+308
 #define FLT_MAX		((float)3.40282346638528860e+38)
 #endif
+#ifndef SSIZE_MAX
+#define SSIZE_MAX ((~((size_t) 0)) / 2)
+#endif
 
 #if !defined(HAVE_ISINF) && !defined(isinf)
 #define isinf(X)    0

mysql-test/r/fulltext2.result

@@ -241,3 +241,11 @@ select * from t1 where match a against('ab c' in boolean mode);
 a
 drop table t1;
 set names latin1;
+SET NAMES utf8;
+CREATE TABLE t1(a VARCHAR(255), FULLTEXT(a)) ENGINE=MyISAM DEFAULT CHARSET=utf8;
+INSERT INTO t1 VALUES('„MySQL“');
+SELECT a FROM t1 WHERE MATCH a AGAINST('“MySQL„' IN BOOLEAN MODE);
+a
+„MySQL“
+DROP TABLE t1;
+SET NAMES latin1;

mysql-test/t/fulltext2.test

@@ -221,3 +221,13 @@ drop table t1;
 set names latin1;
 
 # End of 4.1 tests
+
+#
+# BUG#19580 - FULLTEXT search produces wrong results on UTF-8 columns
+#
+SET NAMES utf8;
+CREATE TABLE t1(a VARCHAR(255), FULLTEXT(a)) ENGINE=MyISAM DEFAULT CHARSET=utf8;
+INSERT INTO t1 VALUES('„MySQL“');
+SELECT a FROM t1 WHERE MATCH a AGAINST('“MySQL„' IN BOOLEAN MODE);
+DROP TABLE t1;
+SET NAMES latin1;

sql/mysqld.cc

@@ -6156,7 +6156,8 @@ The minimum value for this variable is 4096.",
    "Each thread that does a sequential scan allocates a buffer of this size for each table it scans. If you do many sequential scans, you may want to increase this value.",
    (gptr*) &global_system_variables.read_buff_size,
    (gptr*) &max_system_variables.read_buff_size,0, GET_ULONG, REQUIRED_ARG,
-   128*1024L, IO_SIZE*2+MALLOC_OVERHEAD, ~0L, MALLOC_OVERHEAD, IO_SIZE, 0},
+   128*1024L, IO_SIZE*2+MALLOC_OVERHEAD, SSIZE_MAX, MALLOC_OVERHEAD, IO_SIZE,
+   0},
   {"read_only", OPT_READONLY,
    "Make all non-temporary tables read-only, with the exception for replication (slave) threads and users with the SUPER privilege",
    (gptr*) &opt_readonly,
@@ -6167,12 +6168,12 @@ The minimum value for this variable is 4096.",
    (gptr*) &global_system_variables.read_rnd_buff_size,
    (gptr*) &max_system_variables.read_rnd_buff_size, 0,
    GET_ULONG, REQUIRED_ARG, 256*1024L, IO_SIZE*2+MALLOC_OVERHEAD,
-   ~0L, MALLOC_OVERHEAD, IO_SIZE, 0},
+   SSIZE_MAX, MALLOC_OVERHEAD, IO_SIZE, 0},
   {"record_buffer", OPT_RECORD_BUFFER,
    "Alias for read_buffer_size",
    (gptr*) &global_system_variables.read_buff_size,
    (gptr*) &max_system_variables.read_buff_size,0, GET_ULONG, REQUIRED_ARG,
-   128*1024L, IO_SIZE*2+MALLOC_OVERHEAD, ~0L, MALLOC_OVERHEAD, IO_SIZE, 0},
+   128*1024L, IO_SIZE*2+MALLOC_OVERHEAD, SSIZE_MAX, MALLOC_OVERHEAD, IO_SIZE, 0},
 #ifdef HAVE_REPLICATION
   {"relay_log_purge", OPT_RELAY_LOG_PURGE,
    "0 = do not purge relay logs. 1 = purge them as soon as they are no more needed.",

sql/sql_show.cc

@@ -438,6 +438,7 @@ mysql_find_files(THD *thd,List<char> *files, const char *db,const char *path,
   uint col_access=thd->col_access;
 #endif
   TABLE_LIST table_list;
+  char tbbuff[FN_REFLEN];
   DBUG_ENTER("mysql_find_files");
 
   if (wild && !wild[0])
@@ -454,6 +455,8 @@ mysql_find_files(THD *thd,List<char> *files, const char *db,const char *path,
     DBUG_RETURN(-1);
   }
 
+  VOID(tablename_to_filename(tmp_file_prefix, tbbuff, sizeof(tbbuff)));
+
   for (i=0 ; i < (uint) dirp->number_off_files  ; i++)
   {
     char uname[NAME_LEN*3+1];                   /* Unencoded name */
@@ -491,7 +494,7 @@ mysql_find_files(THD *thd,List<char> *files, const char *db,const char *path,
     {
         // Return only .frm files which aren't temp files.
       if (my_strcasecmp(system_charset_info, ext=fn_rext(file->name),reg_ext) ||
-          is_prefix(file->name,tmp_file_prefix))
+          is_prefix(file->name,tbbuff))
         continue;
       *ext=0;
       VOID(filename_to_tablename(file->name, uname, sizeof(uname)));

storage/myisam/ft_parser.c

@@ -111,6 +111,7 @@ byte ft_get_word(CHARSET_INFO *cs, byte **start, byte *end,
                  FT_WORD *word, MYSQL_FTPARSER_BOOLEAN_INFO *param)
 {
   byte *doc=*start;
+  int ctype;
   uint mwc, length, mbl;
 
   param->yesno=(FTB_YES==' ') ? 1 : (param->quot != 0);
@@ -119,9 +120,11 @@ byte ft_get_word(CHARSET_INFO *cs, byte **start, byte *end,
 
   while (doc<end)
   {
-    for (;doc<end;doc++)
+    for (; doc < end; doc+= (mbl > 0 ? mbl : 1))
     {
-      if (true_word_char(cs,*doc)) break;
+      mbl= cs->cset->ctype(cs, &ctype, (uchar*)doc, (uchar*)end);
+      if (true_word_char(ctype, *doc))
+        break;
       if (*doc == FTB_RQUOT && param->quot)
       {
         param->quot=doc;
@@ -155,14 +158,16 @@ byte ft_get_word(CHARSET_INFO *cs, byte **start, byte *end,
     }
 
     mwc=length=0;
-    for (word->pos=doc; doc<end; length++, mbl=my_mbcharlen(cs, *(uchar *)doc), doc+=(mbl ? mbl : 1))
-      if (true_word_char(cs,*doc))
+    for (word->pos= doc; doc < end; length++, doc+= (mbl > 0 ? mbl : 1))
+    {
+      mbl= cs->cset->ctype(cs, &ctype, (uchar*)doc, (uchar*)end);
+      if (true_word_char(ctype, *doc))
         mwc=0;
       else if (!misc_word_char(*doc) || mwc)
         break;
       else
         mwc++;
-
+    }
     param->prev='A'; /* be sure *prev is true_word_char */
     word->len= (uint)(doc-word->pos) - mwc;
     if ((param->trunc=(doc<end && *doc == FTB_TRUNC)))
@@ -197,24 +202,31 @@ byte ft_simple_get_word(CHARSET_INFO *cs, byte **start, const byte *end,
 {
   byte *doc= *start;
   uint mwc, length, mbl;
+  int ctype;
   DBUG_ENTER("ft_simple_get_word");
 
   do
   {
-    for (;; doc++)
+    for (;; doc+= (mbl > 0 ? mbl : 1))
     {
-      if (doc >= end) DBUG_RETURN(0);
-      if (true_word_char(cs, *doc)) break;
+      if (doc >= end)
+        DBUG_RETURN(0);
+      mbl= cs->cset->ctype(cs, &ctype, (uchar*)doc, (uchar*)end);
+      if (true_word_char(ctype, *doc))
+        break;
     }
 
     mwc= length= 0;
-    for (word->pos=doc; doc<end; length++, mbl=my_mbcharlen(cs, *(uchar *)doc), doc+=(mbl ? mbl : 1))
-      if (true_word_char(cs,*doc))
+    for (word->pos= doc; doc < end; length++, doc+= (mbl > 0 ? mbl : 1))
+    {
+      mbl= cs->cset->ctype(cs, &ctype, (uchar*)doc, (uchar*)end);
+      if (true_word_char(ctype, *doc))
         mwc= 0;
       else if (!misc_word_char(*doc) || mwc)
         break;
       else
         mwc++;
+    }
 
     word->len= (uint)(doc-word->pos) - mwc;
 

storage/myisam/ft_update.c

@@ -174,11 +174,6 @@ int _mi_ft_cmp(MI_INFO *info, uint keynr, const byte *rec1, const byte *rec2)
   FT_SEG_ITERATOR ftsi1, ftsi2;
   CHARSET_INFO *cs=info->s->keyinfo[keynr].seg->charset;
   DBUG_ENTER("_mi_ft_cmp");
-#ifndef MYSQL_HAS_TRUE_CTYPE_IMPLEMENTATION
-  if (cs->mbmaxlen > 1)
-    DBUG_RETURN(THOSE_TWO_DAMN_KEYS_ARE_REALLY_DIFFERENT);
-#endif
-
   _mi_ft_segiterator_init(info, keynr, rec1, &ftsi1);
   _mi_ft_segiterator_init(info, keynr, rec2, &ftsi2);
 

storage/myisam/ftdefs.h

@@ -24,9 +24,10 @@
 #include <queues.h>
 #include <mysql/plugin.h>
 
-#define true_word_char(s,X)	(my_isalnum(s,X) || (X)=='_')
+#define true_word_char(ctype, character) \
+                      ((ctype) & (_MY_U | _MY_L | _MY_NMR) || \
+                       (character) == '_')
 #define misc_word_char(X)	0
-#define word_char(s,X)		(true_word_char(s,X) || misc_word_char(X))
 
 #define FT_MAX_WORD_LEN_FOR_SORT 31