New myisamchk option --sort-recover

Allow delete of crashed MyISAM tables
Fixed bug when BLOB was first part of key
Fixed bug when using result from CASE in GROUP BY
Fixed core-dump bug in monthname()
Optimized calling of check_db_name()


Docs/manual.texi:
  Added more information about myisamchk
client/mysqladmin.c:
  Added error message for CREATE database and fixed possible overflow bug
include/myisam.h:
  New myisamchk option --sort-recover
libmysql/libmysql.c:
  Removed commented code
  Don't define getpwuid (breaks on SCO 3.2)
myisam/mi_check.c:
  Fixed (new) bug when using --recover --optimize
myisam/mi_delete_table.c:
  Allow delete of crashed tables
myisam/mi_key.c:
  Fixed bug when BLOB was first part of key
myisam/myisamchk.c:
  New myisamchk option --sort-recover
mysql-test/r/case.result:
  New test cases to check for reported bugs
mysql-test/r/func_time.result:
  New test cases to check for reported bugs
mysql-test/r/type_blob.result:
  New test cases to check for reported bugs
mysql-test/r/type_datetime.result:
  New test cases to check for reported bugs
mysql-test/t/case.test:
  New test cases to check for reported bugs
mysql-test/t/func_time.test:
  New test cases to check for reported bugs
mysql-test/t/type_blob.test:
  New test cases to check for reported bugs
mysql-test/t/type_datetime.test:
  New test cases to check for reported bugs
mysys/my_bitmap.c:
  Optimize
sql-bench/limits/ms-sql.cfg:
  Updated limits
sql/item_cmpfunc.cc:
  Fixed bug when using result from CASE in GROUP BY
sql/item_cmpfunc.h:
  Fixed bug when using result from CASE in GROUP BY
sql/item_timefunc.cc:
  Fixed core-dump bug in monthname()
sql/sql_db.cc:
  Optimized calling of check_db_name()
sql/sql_parse.cc:
  Optimized calling of check_db_name()
sql/table.cc:
  Fixed typo

Docs/manual.texi

@@ -9952,7 +9952,7 @@ for this is that negative numbers caused problems when wrapping from -1 to 0.
 @code{AUTO_INCREMENT} is now for MyISAM tables handled at a lower level and
 is much faster than before. For MyISAM tables old numbers are also not reused
 anymore, even if you delete some rows from the table.
-@item @code{INNER}, @code{DELAYED}, @code{RIGHT}, @code{CASE}, @code{THEN}, @code{WHEN}, @code{ELSE}, @code{END} and @code{WHEN} are now reserved words.
+@item @code{CASE}, @code{DELAYED}, @code{ELSE}, @code{END}, @code{FULLTEXT}, @code{INNER}, @code{RIGHT}, @code{THEN} and @code{WHEN} are now reserved words.
 @item @code{FLOAT(X)} is now a true floating-point type and not a value with
 a fixed number of decimals.
 @item When declaring @code{DECIMAL(length,dec)} the length argument no
@@ -15814,6 +15814,10 @@ mysql> SELECT CASE BINARY "B" when "a" then 1 when "b" then 2 END;
 @end example
 @end table
 
+The type of the return value (@code{INTEGER}, @code{DOUBLE} or
+@code{STRING}) is the same as the type of the first returned value (the
+expression after the first @code{THEN}).
+
 @findex mathematical functions
 @findex functions, mathematical
 @node Mathematical functions, String functions, Control flow functions, Functions
@@ -30619,11 +30623,11 @@ utility. @xref{Table types}.
 In the following text we will talk about @code{myisamchk}, but everything
 also applies to the old @code{isamchk}.
 
-You can use the @code{myisamchk} utility to get information about your database
+You can use the @code{myisamchk} utility to get information about your
-tables, check and repair them, or optimize them.  The following sections
+database tables, check and repair them, or optimize them.  The following
-describe how to invoke @code{myisamchk} (including a description of its
+sections describe how to invoke @code{myisamchk} (including a
-options), how to set up a table maintenance schedule, and how to use
+description of its options), how to set up a table maintenance schedule,
-@code{myisamchk} to perform its various functions.
+and how to use @code{myisamchk} to perform its various functions.
 
 You can, in most cases, also use the command @code{OPTIMIZE TABLES} to
 optimize and repair tables, but this is not as fast or reliable (in case
@@ -30713,9 +30717,9 @@ the file or that has died without closing the file properly.
 
 If you @code{mysqld} is running, you must force a sync/close of all
 tables with @code{FLUSH TABLES} and ensure that no one is using the
-tables while you are running @code{myisamchk}.  In MySQL Version 3.23 the easiest
+tables while you are running @code{myisamchk}.  In @strong{MySQL} Version 3.23
-way to avoid this problem is to use @code{CHECK TABLE} instead of
+the easiest way to avoid this problem is to use @code{CHECK TABLE}
-@code{myisamchk} to check tables.
+instead of @code{myisamchk} to check tables.
 
 @cindex options, @code{myisamchk}
 @cindex @code{myisamchk}, options
@@ -30740,9 +30744,30 @@ for myisamchk can be examined with @code{myisamchk --help}:
 @item decode_bits @tab 9
 @end multitable
 
-@code{key_buffer_size} is only used when you check the table with @code{-e} or
+@code{sort_buffer_size} is used when the keys are reparied by sorting
-repair it with @code{-o}.
+keys, which is the normal case when you use @code{--recover}.
-@code{sort_buffer_size} is used when you repair the table with @code{-r}.
+
+@code{key_buffer_size} is used when you are checking the table with
+@code{--extended-check} or when the keys are repaired by inserting key
+row by row in to the table (like when doing normal inserts). Repairing
+through the key buffer is used in the following cases:
+
+@itemize @bullet
+@item
+If you use @code{--safe-recover}.
+@item
+If you are using a @code{FULLTEXT} index.
+@item
+If the temporary files needed to sort the keys would be more than twice
+as big as when creating the key file directly.  This is often the case
+when you have big @code{CHAR}, @code{VARCHAR} or @code{TEXT} keys as the
+sort needs to store the whole keys during sorting. If you have lots
+of temporary space and you can force @code{myisamchk} to repair by sorting
+you can use the @code{--sort-recover} option.
+@end itemize
+
+Reparing through the key buffer takes much less disk space than using
+sorting, but is also much slower.
 
 If you want a faster repair, set the above variables to about 1/4 of your
 available memory.  You can set both variables to big values, as only one
@@ -30851,6 +30876,11 @@ space than @code{-r}. Normally one should always first repair with
 
 If you have lots of memory, you should increase the size of
 @code{key_buffer_size}!
+@item -n or --sort-recover
+Force @code{myisamchk} to use sorting to resolve the keys even if the
+temporary files should be very big.  This will not have any effect if you have
+fulltext keys in the table.
+
 @item --character-sets-dir=...
 Directory where character sets are stored.
 @item --set-character-set=name
@@ -30933,8 +30963,9 @@ Space for the new index file that replaces the old one. The old
 index file is truncated at start, so one usually ignore this space.
 This space is needed on the same disk as the original index file!
 @item
-When using @code{--repair} (but not when using @code{--safe-repair}, you
+When using @code{--recover} or @code{--sort-recover}
-will need space for a sort buffer for:
+(but not when using @code{--safe-recover}, you will need space for a
+sort buffer for:
 @code{(largest_key + row_pointer_length)*number_of_rows * 2}.
 You can check the length of the keys and the row_pointer_length with
 @code{myisamchk -dv table}.
@@ -30943,7 +30974,7 @@ This space is allocated on the temporary disk (specified by @code{TMPDIR} or
 @end itemize
 
 If you have a problem with disk space during repair, you can try to use
-@code{--safe-repair} instead of @code{--repair}.
+@code{--safe-recover} instead of @code{--recover}.
 
 @cindex maintaining, tables
 @cindex tables, maintenance regimen
@@ -39486,6 +39517,8 @@ javascript, 2d/3d graphics, and PHP3/MySQL. All pages are generated from
 a database.
 @c Added 990614; EMAIL: downey@image.dk (Rune Madsen)
 
+@item @uref{http://www.softwarezrus.com/, Softwarezrus.com}
+Ecommerce site that is selling computers.
 @end itemize
 
 @cindex consultants, list of
@@ -39782,6 +39815,10 @@ OLEDB handler for @code{MySQL}. By SWsoft.
 Examples and documentation for MyOLEDB. By SWsoft.
 @item @uref{http://www.mysql.com/Downloads/Win32/Myoledb.zip, Myoledb.zip}
 Source for MyOLEDB. By SWsoft.
+@item @uref{http://www.mysql.com/Downloads/Win32/MyOLEDB.chm, MyOLEDB.chm}
+Help files for MyOLEDB.
+@item @uref{http://www.mysql.com/Downloads/Win32/libmyodbc.zip, libmyodbc.zip}
+Static MyODBC library used for build MyOLEDB. Based on MyODBC code.
 @end itemize
 
 @cindex C++
@@ -40950,6 +40987,13 @@ not yet 100 % confident in this code.
 @appendixsubsec Changes in release 3.23.33
 @itemize bullet
 @item
+Fixed 'no found rows' bug in @code{MyISAM} tables when a @code{BLOB} was
+first part of a multi-part key.
+@item
+Fixed bug where @code{CASE} didn't work with @code{GROUP BY}.
+@item
+Added option @code{--sort-recover} to @code{myisamchk}.
+@item
 @code{myisamchk -S} and @code{OPTIMIZE TABLE} now works on Windows.
 @item
 Fixed bug when using @code{DISTINCT} on results from functions that refered to

client/mysqladmin.c

@@ -28,7 +28,7 @@
 #include <my_pthread.h>				/* because of signal()	*/
 #endif
 
-#define ADMIN_VERSION "8.14"
+#define ADMIN_VERSION "8.15"
 #define MAX_MYSQL_VAR 64
 #define MAX_TIME_TO_WAIT 3600			/* Wait for shutdown */
 #define MAX_TRUNC_LENGTH 3
@@ -404,7 +404,11 @@ static my_bool execute_commands(MYSQL *mysql,int argc, char **argv)
       }
       sprintf(buff,"create database %.*s",FN_REFLEN,argv[1]);
       if (mysql_query(mysql,buff))
+      {
+	my_printf_error(0,"Create failed; error: '%-.200s'",MYF(ME_BELL),
+			mysql_error(mysql));
 	return 1;
+      }
       else
       {
 	argc--; argv++;
@@ -762,7 +766,7 @@ static my_bool execute_commands(MYSQL *mysql,int argc, char **argv)
       mysql->reconnect=1;	/* Automatic reconnect is default */
       break;
     default:
-      my_printf_error(0,"Unknown command: '%s'",MYF(ME_BELL),argv[0]);
+      my_printf_error(0,"Unknown command: '%-.60s'",MYF(ME_BELL),argv[0]);
       return 1;
     }
   }

include/myisam.h

@@ -324,7 +324,7 @@ typedef struct st_mi_check_param
   uint testflag;
   uint8 language;
   my_bool using_global_keycache, opt_lock_memory, opt_follow_links;
-  my_bool retry_repair,retry_without_quick;
+  my_bool retry_repair,retry_without_quick, force_sort;
   char temp_filename[FN_REFLEN],*isam_file_name,*tmpdir;
   int tmpfile_createflag;
   myf myf_rw;

libmysql/libmysql.c

@@ -485,7 +485,7 @@ static void free_old_query(MYSQL *mysql)
   DBUG_VOID_RETURN;
 }
 
-#ifdef HAVE_GETPWUID
+#if defined(HAVE_GETPWUID) && defined(NO_GETPWUID_DECL)
 struct passwd *getpwuid(uid_t);
 char* getlogin(void);
 #endif
@@ -501,14 +501,6 @@ static void read_user_name(char *name)
 #ifdef HAVE_GETPWUID
     struct passwd *skr;
     const char *str;
-/*#ifdef __cplusplus
-    extern "C" struct passwd *getpwuid(uid_t);
-    extern "C" { char* getlogin(void); }
-#else
-    char * getlogin();
-    struct passwd *getpwuid(uid_t);
-#endif
-*/
     if ((str=getlogin()) == NULL)
     {
       if ((skr=getpwuid(geteuid())) != NULL)

myisam/mi_check.c

@@ -1464,6 +1464,7 @@ int mi_sort_index(MI_CHECK *param, register MI_INFO *info, my_string name)
   my_off_t index_pos[MI_MAX_POSSIBLE_KEY];
   uint r_locks,w_locks;
   MYISAM_SHARE *share=info->s;
+  MI_STATE_INFO old_state;
   DBUG_ENTER("sort_index");
 
   if (!(param->testflag & T_SILENT))
@@ -1502,9 +1503,10 @@ int mi_sort_index(MI_CHECK *param, register MI_INFO *info, my_string name)
   /* Flush key cache for this file if we are calling this outside myisamchk */
   flush_key_blocks(share->kfile, FLUSH_IGNORE_CHANGED);
 
-	/* Put same locks as old file */
   share->state.version=(ulong) time((time_t*) 0);
+  old_state=share->state;			/* save state if not stored */
   r_locks=share->r_locks; w_locks=share->w_locks;
+	/* Put same locks as old file */
   share->r_locks=share->w_locks=0;
   (void) _mi_writeinfo(info,WRITEINFO_UPDATE_KEYFILE);
   VOID(my_close(share->kfile,MYF(MY_WME)));
@@ -1518,6 +1520,7 @@ int mi_sort_index(MI_CHECK *param, register MI_INFO *info, my_string name)
   _mi_readinfo(info,F_WRLCK,0);			/* Will lock the table */
   info->lock_type=F_WRLCK;
   share->r_locks=r_locks; share->w_locks=w_locks;
+  share->state=old_state;			/* Restore old state */
 
   info->state->key_file_length=param->new_file_pos;
   info->update= (short) (HA_STATE_CHANGED | HA_STATE_ROW_CHANGED);
@@ -3144,7 +3147,7 @@ void mi_disable_non_unique_index(MI_INFO *info, ha_rows rows)
 */
 
 my_bool mi_test_if_sort_rep(MI_INFO *info, ha_rows rows, 
-			    my_bool force __attribute__((unused)))
+			    my_bool force)
 {
   MYISAM_SHARE *share=info->s;
   uint i;
@@ -3160,7 +3163,8 @@ my_bool mi_test_if_sort_rep(MI_INFO *info, ha_rows rows,
    is very time-consuming process, it's better to leave it to repair stage
    but this repair shouldn't be repair_by_sort (serg)
  */
-    if (mi_too_big_key_for_sort(key,rows) || (key->flag & HA_FULLTEXT))
+    if (!force && mi_too_big_key_for_sort(key,rows) ||
+	(key->flag & HA_FULLTEXT))
       return FALSE;
   }
   return TRUE;

myisam/mi_delete_table.c

@@ -37,7 +37,8 @@ int mi_delete_table(const char *name)
 #ifdef USE_RAID
   {
     MI_INFO *info;
-    if (!(info=mi_open(name, O_RDONLY, 0)))
+    /* we use 'open_for_repair' to be able to delete a crashed table */
+    if (!(info=mi_open(name, O_RDONLY, HA_OPEN_FOR_REPAIR)))
       DBUG_RETURN(my_errno);
     raid_type =      info->s->base.raid_type;
     raid_chunks =    info->s->base.raid_chunks;

myisam/mi_key.c

@@ -184,8 +184,9 @@ uint _mi_pack_key(register MI_INFO *info, uint keynr, uchar *key, uchar *old,
       /* Length of key-part used with mi_rkey() always 2 */
       uint tmp_length=uint2korr(pos);
       k_length-= 2+length;
-      set_if_smaller(length,tmp_length);
+      set_if_smaller(length,tmp_length);	/* Safety */
       store_key_length_inc(key,length);
+      old+=2;					/* Skipp length */
       memcpy((byte*) key, pos+2,(size_t) length);
       key+= length;
       continue;

myisam/myisamchk.c

@@ -189,6 +189,7 @@ static struct option long_options[] =
   {"silent",	       no_argument,	  0, 's'},
   {"sort-index",       no_argument,	  0, 'S'},
   {"sort-records",     required_argument, 0, 'R'},
+  {"sort-recover",     no_argument,	  0, 'n'},
   {"tmpdir",	       required_argument, 0, 't'},
   {"update-state",     no_argument,	  0, 'U'},
   {"unpack",	       no_argument,	  0, 'u'},
@@ -200,7 +201,7 @@ static struct option long_options[] =
 
 static void print_version(void)
 {
-  printf("%s  Ver 1.41 for %s at %s\n",my_progname,SYSTEM_TYPE,
+  printf("%s  Ver 1.42 for %s at %s\n",my_progname,SYSTEM_TYPE,
 	 MACHINE_TYPE);
 }
 
@@ -255,9 +256,11 @@ static void usage(void)
 		      myisamchk repairs the table a symlink points at.\n\
   -r, --recover       Can fix almost anything except unique keys that aren't\n\
                       unique.\n\
+  -n, --sort-recover  Force recovering with sorting even if the temporary\n\
+		      file would be very big.\n\
   -o, --safe-recover  Uses old recovery method; Slower than '-r' but can\n\
-		      handle a couple of cases where '-r' reports that it can't\n\
+		      handle a couple of cases where '-r' reports that it\n\
-		      fix the data file.\n\
+		      can't fix the data file.\n\
   --character-sets-dir=...\n\
                       Directory where character sets are\n\
   --set-character-set=name\n\
@@ -306,7 +309,8 @@ static void get_options(register int *argc,register char ***argv)
   set_all_changeable_vars(changeable_vars);
   if (isatty(fileno(stdout)))
     check_param.testflag|=T_WRITE_LOOP;
-  while ((c=getopt_long(*argc,*argv,"aBcCdeifF?lqrmosSTuUvVw#:b:D:k:O:R:A::t:",
+  while ((c=getopt_long(*argc,*argv,
+			"aBcCdeifF?lqrmnosSTuUvVw#:b:D:k:O:R:A::t:",
 			long_options, &option_index)) != EOF)
   {
     switch(c) {
@@ -374,8 +378,13 @@ static void get_options(register int *argc,register char ***argv)
       break;
     case 'o':
       check_param.testflag= (check_param.testflag & ~T_REP_BY_SORT) | T_REP;
+      check_param.force_sort=0;
       my_disable_async_io=1;		/* More safety */
       break;
+    case 'n':
+      check_param.testflag= (check_param.testflag & ~T_REP) | T_REP_BY_SORT;
+      check_param.force_sort=1;
+      break;
     case 'q':
       check_param.opt_rep_quick++;
       break;
@@ -683,7 +692,8 @@ static int myisamchk(MI_CHECK *param, my_string filename)
 	if ((param->testflag & T_REP_BY_SORT) &&
 	    (share->state.key_map ||
 	     (rep_quick && !param->keys_in_use && !recreate)) &&
-	    mi_test_if_sort_rep(info, info->state->records, 1))
+	    mi_test_if_sort_rep(info, info->state->records,
+				check_param.force_sort))
 	{
 	  error=mi_repair_by_sort(&check_param,info,fixed_name,rep_quick);
 	  state_updated=1;

mysql-test/r/case.result

@@ -32,3 +32,11 @@ case when 1>0 then "TRUE" else "FALSE" END
 TRUE
 case when 1<0 then "TRUE" else "FALSE" END
 FALSE
+fcase	count(*)
+0	2
+2	1
+3	1
+fcase	count(*)
+nothing	2
+one	1
+two	1

mysql-test/r/func_time.result

@@ -186,3 +186,9 @@ extract(SECOND FROM "1999-01-02 10:11:12")
 12
 ctime	hour(ctime)
 2001-01-12 12:23:40	12
+monthname(date)
+NULL
+January
+monthname(date)
+NULL
+January

mysql-test/r/type_blob.result

@@ -257,3 +257,7 @@ hello
 hello word
 count(*)
 2
+foobar	boggle
+fish	10
+foobar	boggle
+fish	10

mysql-test/r/type_datetime.result

@@ -13,6 +13,8 @@ t
 9999-12-31 23:59:59
 Table	Op	Msg_type	Msg_text
 test.t1	optimize	status	OK
+Table	Op	Msg_type	Msg_text
+test.t1	check	status	OK
 t
 2000-01-01 00:00:00
 2069-12-31 00:00:00

mysql-test/t/case.test

@@ -2,6 +2,8 @@
 # Testing of CASE
 #
 
+drop table if exists t1;
+
 select CASE "b" when "a" then 1 when "b" then 2 END;
 select CASE "c" when "a" then 1 when "b" then 2 END;
 select CASE "c" when "a" then 1 when "b" then 2 ELSE 3 END;
@@ -19,3 +21,12 @@ select (case 1/0 when "a" then "true" END) | 0;
 select (case 1/0 when "a" then "true" END) + 0.0;
 select case when 1>0 then "TRUE" else "FALSE" END;
 select case when 1<0 then "TRUE" else "FALSE" END;
+
+#
+# Test bug when using GROUP BY on CASE
+#
+create table t1 (a int);
+insert into t1 values(1),(2),(3),(4);
+select case a when 1 then 2 when 2 then 3 else 0 end as fcase, count(*) from t1 group by fcase;
+select case a when 1 then "one" when 2 then "two" else "nothing" end as fcase, count(*) from t1 group by fcase;
+drop table t1;

mysql-test/t/func_time.test

@@ -1,6 +1,8 @@
 #
 # time functions
 #
+drop table if exists t1,t2;
+
 select from_days(to_days("960101")),to_days(960201)-to_days("19960101"),to_days(date_add(curdate(), interval 1 day))-to_days(curdate()),weekday("1997-11-29");
 select period_add("9602",-12),period_diff(199505,"9404") ;
 select now()-now(),weekday(curdate())-weekday(now()),unix_timestamp()-unix_timestamp(now());
@@ -99,9 +101,21 @@ select extract(MINUTE FROM "10:11:12");
 select extract(MINUTE_SECOND FROM "10:11:12");
 select extract(SECOND FROM "1999-01-02 10:11:12");
 
-drop table if exists t1;
 create table t1 (ctime varchar(20));
 insert into t1 values ('2001-01-12 12:23:40');
 select ctime, hour(ctime) from t1;
 drop table t1;
 
+#
+# Test bug with monthname() and NULL
+#
+
+create table t1 (id int);
+create table t2 (id int, date date);
+insert into t1 values (1);
+insert into t2 values (1, "0000-00-00");
+insert into t1 values (2);
+insert into t2 values (2, "2000-01-01");
+select monthname(date) from t1 inner join t2 on t1.id = t2.id;
+select monthname(date) from t1 inner join t2 on t1.id = t2.id order by t1.id;
+drop table t1,t2;

mysql-test/t/type_blob.test

@@ -239,3 +239,13 @@ INSERT INTO t1 VALUES (0,'traktor','1111111111111');
 INSERT INTO t1 VALUES (1,'traktor','1111111111111111111111111');
 select count(*) from t1 where f2='traktor';
 drop table t1;
+
+#
+# Test of found bug when blob is first key part
+#
+
+create table t1 (foobar tinyblob not null, boggle smallint not null, key (foobar(32), boggle));
+insert into t1 values ('fish', 10),('bear', 20);
+select foobar, boggle from t1 where foobar = 'fish';
+select foobar, boggle from t1 where foobar = 'fish' and boggle = 10;
+drop table t1;

mysql-test/t/type_datetime.test

@@ -1,11 +1,14 @@
 #
 # testing different DATETIME ranges
 #
+
+drop table if exists t1;
 create table t1 (t datetime);
 insert into t1 values(101),(691231),(700101),(991231),(10000101),(99991231),(101000000),(691231000000),(700101000000),(991231235959),(10000101000000),(99991231235959);
 select * from t1;
 delete from t1 where t > 0;
 optimize table t1;
+check table t1;
 insert into t1 values("000101"),("691231"),("700101"),("991231"),("00000101"),("00010101"),("99991231"),("00101000000"),("691231000000"),("700101000000"),("991231235959"),("10000101000000"),("99991231235959");
 select * from t1;
 drop table t1;

mysys/my_bitmap.c

@@ -17,6 +17,13 @@
 
 /*
   Handling of uchar arrays as large bitmaps.
+  We assume that the size of the used bitmap is less than ~(uint) 0
+
+  TODO:
+
+  create an unique structure for this that includes the mutex and bitmap size
+  make a init function that will allocate the bitmap and init the mutex
+  make an end function that will free everything
 */
 
 #include "mysys_priv.h"
@@ -24,37 +31,51 @@
 
 pthread_mutex_t LOCK_bitmap;
 
-void bitmap_set_bit(uchar *bitmap, uint bitmap_size, uint bitmap_bit) {
+void bitmap_set_bit(uchar *bitmap, uint bitmap_size, uint bitmap_bit)
-  if((bitmap_bit != MY_BIT_NONE) && (bitmap_bit < bitmap_size*8)) {
+{
+  if (bitmap_bit < bitmap_size*8)
+  {
     pthread_mutex_lock(&LOCK_bitmap);
-    bitmap[bitmap_bit / 8] |= (1 << bitmap_bit % 8);
+    bitmap[bitmap_bit / 8] |= (1 << (bitmap_bit & 7));
     pthread_mutex_unlock(&LOCK_bitmap);
-  };
+  }
-};
+}
 
-uint bitmap_set_next(uchar *bitmap, uint bitmap_size) {
+uint bitmap_set_next(uchar *bitmap, uint bitmap_size)
+{
   uint bit_found = MY_BIT_NONE;
-  int i, b;
+  uint i;
 
   pthread_mutex_lock(&LOCK_bitmap);
-  for(i=0; (i<bitmap_size) && (bit_found==MY_BIT_NONE); i++) {
+  for (i=0; i < bitmap_size ; i++, bitmap++)
-    if(bitmap[i] == 0xff) continue;
+  {
-    for(b=0; (b<8) && (bit_found==MY_BIT_NONE); b++)
+    if (*bitmap != 0xff)
-      if((bitmap[i] & 1<<b) == 0) {
+    {						/* Found slot with free bit */
-        bit_found = (i*8)+b;
+      uint b;
-        bitmap[i] |= 1<<b;
+      for (b=0; ; b++)
-      };
+      {
-  };
+	if (!(*bitmap & (1 << b)))
+	{
+	  *bitmap |= 1<<b;
+	  bit_found = (i*8)+b;
+	  break;
+	}
+      }
+      break;					/* Found bit */
+    }
+  }
   pthread_mutex_unlock(&LOCK_bitmap);
-
   return bit_found;
-};
+}
 
-void bitmap_clear_bit(uchar *bitmap, uint bitmap_size, uint bitmap_bit) {
+
-  if((bitmap_bit != MY_BIT_NONE) && (bitmap_bit < bitmap_size*8)) {
+void bitmap_clear_bit(uchar *bitmap, uint bitmap_size, uint bitmap_bit)
+{
+  if (bitmap_bit < bitmap_size*8)
+  {
     pthread_mutex_lock(&LOCK_bitmap);
-    bitmap[bitmap_bit / 8] &= ~(1 << bitmap_bit % 8);
+    bitmap[bitmap_bit / 8] &= ~ (1 << (bitmap_bit & 7));
     pthread_mutex_unlock(&LOCK_bitmap);
-  };
+  }
-};
+}
 

sql-bench/limits/ms-sql.cfg

@@ -1,4 +1,4 @@
-#This file is automaticly generated by crash-me 1.45
+#This file is automaticly generated by crash-me 1.54
 
 NEG=yes					# update of column= -column
 Need_cast_for_null=no			# Need to cast NULL for arithmetic
@@ -19,25 +19,25 @@ alter_modify_col=no			# Alter table modify column
 alter_rename_table=no			# Alter table rename table
 atomic_updates=yes			# atomic updates
 atomic_updates_with_rollback=yes	# atomic_updates_with_rollback
-automatic_rowid=no			# Automatic rowid
+automatic_rowid=no			# Automatic row id
 binary_numbers=yes			# binary numbers (0b1001)
 binary_strings=no			# binary strings (b'0110')
-case_insensitive_strings=yes		# case insensitive compare
+case_insensitive_strings=yes		# Case insensitive compare
 char_is_space_filled=no			# char are space filled
 column_alias=yes			# Column alias
 columns_in_group_by=+64			# number of columns in group by
 columns_in_order_by=+64			# number of columns in order by
 comment_#=no				# # as comment
-comment_--=yes				# -- as comment
+comment_--=yes				# -- as comment (ANSI)
 comment_/**/=yes			# /* */ as comment
-comment_//=no				# // as comment
+comment_//=no				# // as comment (ANSI)
 compute=yes				# Compute
 connections=1000			# Simultaneous connections (installation default)
 constraint_check=no			# Column constraints
 constraint_check_table=yes		# Table constraints
 constraint_null=yes			# NULL constraint (SyBase style)
-crash_me_safe=no			# crash me safe
+crash_me_safe=yes			# crash me safe
-crash_me_version=1.45			# crash me version
+crash_me_version=1.54			# crash me version
 create_default=yes			# default value for column
 create_default_func=no			# default value function for column
 create_if_not_exists=no			# create table if not exists
@@ -46,14 +46,18 @@ create_schema=yes			# Create SCHEMA
 create_table_select=no			# create table from select
 cross_join=yes				# cross join (same as from a,b)
 date_as_string=no			# String functions on date columns
+date_infinity=no			# Supports 'infinity dates
 date_last=no				# Supports 9999-12-31 dates
 date_one=no				# Supports 0001-01-01 dates
 date_with_YY=no				# Supports YY-MM-DD dates
 date_zero=no				# Supports 0000-00-00 dates
 domains=no				# Domains (ANSI SQL)
+dont_require_cast_to_float=yes		# No need to cast from integer to float
 double_quotes=yes			# Double '' as ' in strings
 drop_if_exists=no			# drop table if exists
 drop_index=with 'table.index'		# drop index
+drop_requires_cascade=no		# drop table require cascade/restrict
+drop_restrict=no			# drop table with cascade/restrict
 end_colon=yes				# allows end ';'
 except=no				# except
 except_all=no				# except all
@@ -161,6 +165,7 @@ func_extra_version=no			# Function VERSION
 func_extra_weekday=no			# Function WEEKDAY
 func_extra_|=yes			# Function | (bitwise or)
 func_extra_||=no			# Function OR as '||'
+func_extra_~*=no			# Function ~* (case insensitive compare)
 func_odbc_abs=yes			# Function ABS
 func_odbc_acos=yes			# Function ACOS
 func_odbc_ascii=yes			# Function ASCII
@@ -243,7 +248,8 @@ func_sql_extract_sql=no			# Function EXTRACT
 func_sql_localtime=no			# Function LOCALTIME
 func_sql_localtimestamp=no		# Function LOCALTIMESTAMP
 func_sql_lower=yes			# Function LOWER
-func_sql_nullif=yes			# Function NULLIF
+func_sql_nullif_num=yes			# Function NULLIF with numbers
+func_sql_nullif_string=yes		# Function NULLIF with strings
 func_sql_octet_length=no		# Function OCTET_LENGTH
 func_sql_position=no			# Function POSITION
 func_sql_searched_case=yes		# Function searched CASE
@@ -273,7 +279,7 @@ func_where_unique=no			# Function UNIQUE
 functions=yes				# Functions
 group_by=yes				# Group by
 group_by_alias=no			# Group by alias
-group_by_null=yes			# group on column with null values
+group_by_null=yes			# Group on column with null values
 group_by_position=no			# Group by position
 group_distinct_functions=yes		# Group functions with distinct
 group_func_extra_bit_and=no		# Group function BIT_AND
@@ -282,28 +288,33 @@ group_func_extra_count_distinct_list=no	# Group function COUNT(DISTINCT expr,exp
 group_func_extra_std=no			# Group function STD
 group_func_extra_stddev=no		# Group function STDDEV
 group_func_extra_variance=no		# Group function VARIANCE
+group_func_sql_any=no			# Group function ANY
 group_func_sql_avg=yes			# Group function AVG
 group_func_sql_count_*=yes		# Group function COUNT (*)
 group_func_sql_count_column=yes		# Group function COUNT column name
 group_func_sql_count_distinct=yes	# Group function COUNT(DISTINCT expr)
+group_func_sql_every=no			# Group function EVERY
 group_func_sql_max=yes			# Group function MAX on numbers
 group_func_sql_max_str=yes		# Group function MAX on strings
 group_func_sql_min=yes			# Group function MIN on numbers
 group_func_sql_min_str=yes		# Group function MIN on strings
+group_func_sql_some=no			# Group function SOME
 group_func_sql_sum=yes			# Group function SUM
 group_functions=yes			# Group functions
+group_on_unused=yes			# Group on unused column
 has_true_false=no			# TRUE and FALSE
 having=yes				# Having
 having_with_alias=no			# Having on alias
 having_with_group=yes			# Having with group function
 hex_numbers=yes				# hex numbers (0x41)
 hex_strings=no				# hex strings (x'1ace')
-ignore_end_space=yes			# ignore end space in compare
+ignore_end_space=yes			# Ignore end space in compare
 index_in_create=no			# index in create table
 index_namespace=yes			# different namespace for index
 index_parts=no				# index on column part (extension)
 inner_join=yes				# inner join
 insert_empty_string=yes			# insert empty string
+insert_multi_value=no			# INSERT with Value lists
 insert_select=yes			# insert INTO ... SELECT ...
 insert_with_set=no			# INSERT with set syntax
 intersect=no				# intersect
@@ -317,12 +328,12 @@ like_with_column=yes			# column LIKE column
 like_with_number=yes			# LIKE on numbers
 lock_tables=no				# lock table
 logical_value=not supported		# Value of logical operation (1=1)
-max_big_expressions=7			# big expressions
+max_big_expressions=10			# big expressions
 max_char_size=8000			# max char() size
 max_column_name=128			# column name length
 max_columns=1024			# Columns in table
-max_conditions=10922			# OR and AND in WHERE
+max_conditions=13104			# OR and AND in WHERE
-max_expressions=2783			# simple expressions
+max_expressions=656			# simple expressions
 max_index=+64				# max index
 max_index_length=900			# index length
 max_index_name=128			# index name length
@@ -345,7 +356,6 @@ multi_drop=no				# many tables to drop table
 multi_strings=no			# Multiple line strings
 multi_table_delete=no			# DELETE FROM table1,table2...
 multi_table_update=no			# Update with many tables
-insert_multi_value=no			# Value lists in INSERT
 natural_join=no				# natural join
 natural_join_incompat=no		# natural join (incompatible lists)
 natural_left_outer_join=no		# natural left outer join
@@ -354,6 +364,7 @@ null_concat_expr=no			# Is concat('a',NULL) = NULL
 null_in_index=yes			# null in index
 null_in_unique=no			# null in unique index
 null_num_expr=yes			# Is 1+NULL = NULL
+nulls_in_unique=no			# null combination in unique index
 odbc_left_outer_join=yes		# left outer join odbc style
 operating_system=Microsoft Windows 2000 [Version 5.00.2195]	# crash-me tested on
 order_by=yes				# Order by
@@ -361,6 +372,7 @@ order_by_alias=yes			# Order by alias
 order_by_function=yes			# Order by function
 order_by_position=yes			# Order by position
 order_by_remember_desc=no		# Order by DESC is remembered
+order_on_unused=yes			# Order by on unused column
 primary_key_in_create=yes		# primary key in create table
 psm_functions=no			# PSM functions (ANSI SQL)
 psm_modules=no				# PSM modules (ANSI SQL)
@@ -374,7 +386,7 @@ quote_with_"=no				# Allows ' and " as string markers
 recursive_subqueries=40			# recursive subqueries
 remember_end_space=no			# Remembers end space in char()
 remember_end_space_varchar=yes		# Remembers end space in varchar()
-repeat_string_size=8000			# return string size from function
+rename_table=no				# rename table
 right_outer_join=yes			# right outer join
 rowid=no				# Type for row id
 select_constants=yes			# Select constants
@@ -383,15 +395,16 @@ select_limit2=no			# SELECT with LIMIT #,#
 select_string_size=16777207		# constant string size in SELECT
 select_table_update=yes			# Update with sub select
 select_without_from=yes			# SELECT without FROM
-server_version=Microsoft SQL Server  7.00 - 7.00.842 (Intel X86) 	# server version
+server_version=Microsoft SQL Server  2000 - 8.00.194 (Intel X86) 	# server version
 simple_joins=yes			# ANSI SQL simple joins
 storage_of_float=round			# Storage of float values
 subqueries=yes				# subqueries
 table_alias=yes				# Table alias
 table_name_case=yes			# case independent table names
 table_wildcard=yes			# Select table_name.*
-tempoary_table=no			# temporary tables
+temporary_table=no			# temporary tables
 transactions=yes			# transactions
+truncate_table=no			# truncate
 type_extra_abstime=no			# Type abstime
 type_extra_bfile=no			# Type bfile
 type_extra_blob=no			# Type blob
@@ -399,6 +412,7 @@ type_extra_bool=no			# Type bool
 type_extra_box=no			# Type box
 type_extra_byte=no			# Type byte
 type_extra_char(1_arg)_binary=no	# Type char(1 arg) binary
+type_extra_cidr=no			# Type cidr
 type_extra_circle=no			# Type circle
 type_extra_clob=no			# Type clob
 type_extra_datetime=yes			# Type datetime
@@ -408,6 +422,7 @@ type_extra_float(2_arg)=no		# Type float(2 arg)
 type_extra_float4=no			# Type float4
 type_extra_float8=no			# Type float8
 type_extra_image=yes			# Type image
+type_extra_inet=no			# Type inet
 type_extra_int(1_arg)_zerofill=no	# Type int(1 arg) zerofill
 type_extra_int1=no			# Type int1
 type_extra_int2=no			# Type int2
@@ -424,6 +439,7 @@ type_extra_long_raw=no			# Type long raw
 type_extra_long_varbinary=no		# Type long varbinary
 type_extra_long_varchar(1_arg)=no	# Type long varchar(1 arg)
 type_extra_lseg=no			# Type lseg
+type_extra_macaddr=no			# Type macaddr
 type_extra_mediumint=no			# Type mediumint
 type_extra_mediumtext=no		# Type mediumtext
 type_extra_middleint=no			# Type middleint
@@ -451,7 +467,7 @@ type_extra_timespan=no			# Type timespan
 type_extra_uint=no			# Type uint
 type_extra_varchar2(1_arg)=no		# Type varchar2(1 arg)
 type_extra_year=no			# Type year
-type_odbc_bigint=no			# Type bigint
+type_odbc_bigint=yes			# Type bigint
 type_odbc_binary(1_arg)=yes		# Type binary(1 arg)
 type_odbc_datetime=yes			# Type datetime
 type_odbc_tinyint=yes			# Type tinyint
@@ -459,6 +475,7 @@ type_odbc_varbinary(1_arg)=yes		# Type varbinary(1 arg)
 type_sql_bit=yes			# Type bit
 type_sql_bit(1_arg)=no			# Type bit(1 arg)
 type_sql_bit_varying(1_arg)=no		# Type bit varying(1 arg)
+type_sql_boolean=no			# Type boolean
 type_sql_char(1_arg)=yes		# Type char(1 arg)
 type_sql_char_varying(1_arg)=yes	# Type char varying(1 arg)
 type_sql_character(1_arg)=yes		# Type character(1 arg)

sql/item_cmpfunc.cc

@@ -721,17 +721,41 @@ double Item_func_case::val()
 bool
 Item_func_case::fix_fields(THD *thd,TABLE_LIST *tables)
 {
-
   if (first_expr && first_expr->fix_fields(thd,tables) ||
       else_expr && else_expr->fix_fields(thd,tables))
     return 1;
   if (Item_func::fix_fields(thd,tables))
     return 1;
+  if (first_expr)
+  {
+    used_tables_cache|=(first_expr)->used_tables();
+    const_item_cache&= (first_expr)->const_item();
+  }
+  if (else_expr)
+  {
+    used_tables_cache|=(else_expr)->used_tables();
+    const_item_cache&= (else_expr)->const_item();
+  }
   if (!else_expr || else_expr->maybe_null)
     maybe_null=1;				// The result may be NULL
   return 0;
 }
 
+void Item_func_case::update_used_tables()
+{
+  Item_func::update_used_tables();
+  if (first_expr)
+  {
+    used_tables_cache|=(first_expr)->used_tables();
+    const_item_cache&= (first_expr)->const_item();
+  }
+  if (else_expr)
+  {
+    used_tables_cache|=(else_expr)->used_tables();
+    const_item_cache&= (else_expr)->const_item();
+  }
+}
+
 
 void Item_func_case::fix_length_and_dec()
 {
@@ -750,6 +774,7 @@ void Item_func_case::fix_length_and_dec()
   }
 }
 
+/* TODO:  Fix this so that it prints the whole CASE expression */
 
 void Item_func_case::print(String *str)
 {

sql/item_cmpfunc.h

@@ -254,6 +254,7 @@ public:
   longlong val_int();
   String *val_str(String *);
   void fix_length_and_dec();
+  void update_used_tables();
   enum Item_result result_type () const { return cached_result_type; }
   const char *func_name() const { return "case"; }
   void print(String *str);

sql/item_timefunc.cc

@@ -135,7 +135,11 @@ String* Item_func_monthname::val_str(String* str)
 {
   uint month=(uint) Item_func_month::val_int();
   if (!month)					// This is also true for NULL
+  {
+    null_value=1;
     return (String*) 0;
+  }
+  null_value=0;
   return &month_names[month-1];
 }
 

sql/sql_db.cc

@@ -28,6 +28,8 @@
 static long mysql_rm_known_files(THD *thd, MY_DIR *dirp, const char *path,
 				 uint level);
 
+/* db-name is already validated when we come here */
+
 void mysql_create_db(THD *thd, char *db, uint create_options)
 {
   char	 path[FN_REFLEN+16];
@@ -35,11 +37,6 @@ void mysql_create_db(THD *thd, char *db, uint create_options)
   long result=1;
   DBUG_ENTER("mysql_create_db");
   
-  if (!stripp_sp(db) || check_db_name(db))
-  {
-    net_printf(&thd->net,ER_WRONG_DB_NAME, db);
-    DBUG_VOID_RETURN;
-  }
   VOID(pthread_mutex_lock(&LOCK_mysql_create_db));
 
   /* Check directory */
@@ -96,6 +93,8 @@ static TYPELIB deletable_extentions=
 {array_elements(del_exts)-1,"del_exts", del_exts};
 
 
+/* db-name is already validated when we come here */
+
 void mysql_rm_db(THD *thd,char *db,bool if_exists)
 {
   long deleted=0;
@@ -103,12 +102,6 @@ void mysql_rm_db(THD *thd,char *db,bool if_exists)
   MY_DIR *dirp;
   DBUG_ENTER("mysql_rm_db");
 
-  if (!stripp_sp(db) || check_db_name(db))
-  {
-    net_printf(&thd->net,ER_WRONG_DB_NAME, db);
-    DBUG_VOID_RETURN;
-  }
-  
   VOID(pthread_mutex_lock(&LOCK_mysql_create_db));
   VOID(pthread_mutex_lock(&LOCK_open));
 

sql/sql_parse.cc

@@ -510,7 +510,7 @@ int mysql_table_dump(THD* thd, char* db, char* tbl_name, int fd)
   int error = 0;
   DBUG_ENTER("mysql_table_dump");
   db = (db && db[0]) ? db : thd->db;
-  if(!(table_list = (TABLE_LIST*) sql_calloc(sizeof(TABLE_LIST))))
+  if (!(table_list = (TABLE_LIST*) sql_calloc(sizeof(TABLE_LIST))))
      DBUG_RETURN(1); // out of memory
   table_list->db = db;
   table_list->real_name = table_list->name = tbl_name;
@@ -518,9 +518,14 @@ int mysql_table_dump(THD* thd, char* db, char* tbl_name, int fd)
   table_list->next = 0;
   remove_escape(table_list->real_name);
 
-  if(!(table=open_ltable(thd, table_list, TL_READ_NO_INSERT)))
+  if (!(table=open_ltable(thd, table_list, TL_READ_NO_INSERT)))
     DBUG_RETURN(1);
 
+  if (!db || check_db_name(db))
+  {
+    net_printf(&thd->net,ER_WRONG_DB_NAME, db ? db : "NULL");
+    goto err;
+  }
   if (check_access(thd, SELECT_ACL, db, &table_list->grant.privilege))
     goto err;
   if (grant_option && check_grant(thd, SELECT_ACL, table_list))
@@ -710,6 +715,12 @@ bool do_command(THD *thd)
   case COM_CREATE_DB:
     {
       char *db=thd->strdup(packet+1);
+      // null test to handle EOM
+      if (!db || !stripp_sp(db) || check_db_name(db))
+      {
+	net_printf(&thd->net,ER_WRONG_DB_NAME, db ? db : "NULL");
+	break;
+      }
       if (check_access(thd,CREATE_ACL,db,0,1))
 	break;
       mysql_log.write(thd,command,packet+1);
@@ -719,6 +730,12 @@ bool do_command(THD *thd)
   case COM_DROP_DB:
     {
       char *db=thd->strdup(packet+1);
+      // null test to handle EOM
+      if (!db || !stripp_sp(db) || check_db_name(db))
+      {
+	net_printf(&thd->net,ER_WRONG_DB_NAME, db ? db : "NULL");
+	break;
+      }
       if (check_access(thd,DROP_ACL,db,0,1) || end_active_trans(thd))
 	break;
       mysql_log.write(thd,command,db);
@@ -1503,10 +1520,10 @@ mysql_execute_command(void)
 	goto error;				/* purecov: inspected */
       }
       remove_escape(db);				// Fix escaped '_'
-      if (strlen(db) > NAME_LEN)
+      if (check_db_name(db))
       {
-	net_printf(&thd->net,ER_WRONG_DB_NAME, db);
+        net_printf(&thd->net,ER_WRONG_DB_NAME, db);
-	goto error;
+        goto error;
       }
       if (check_access(thd,SELECT_ACL,db,&thd->col_access))
 	goto error;				/* purecov: inspected */
@@ -1666,6 +1683,11 @@ mysql_execute_command(void)
     break;
   case SQLCOM_CREATE_DB:
     {
+      if (!stripp_sp(lex->name) || check_db_name(lex->name))
+      {
+	net_printf(&thd->net,ER_WRONG_DB_NAME, lex->name);
+	break;
+      }
       if (check_access(thd,CREATE_ACL,lex->name,0,1))
 	break;
       mysql_create_db(thd,lex->name,lex->create_info.options);
@@ -1673,6 +1695,11 @@ mysql_execute_command(void)
     }
   case SQLCOM_DROP_DB:
     {
+      if (!stripp_sp(lex->name) || check_db_name(lex->name))
+      {
+	net_printf(&thd->net,ER_WRONG_DB_NAME, lex->name);
+	break;
+      }
       if (check_access(thd,DROP_ACL,lex->name,0,1) ||
 	  end_active_trans(thd))
 	break;
@@ -1887,12 +1914,6 @@ check_access(THD *thd,uint want_access,const char *db, uint *save_priv,
   if (db == any_db)
     return FALSE;				// Allow select on anything
   
-  if (strlen(db) > NAME_LEN || check_db_name(db))
-    {
-      net_printf(&thd->net,ER_WRONG_DB_NAME, db);
-      return TRUE;
-    }
-
   if (db && (!thd->db || strcmp(db,thd->db)))
     db_access=acl_get(thd->host, thd->ip, (char*) &thd->remote.sin_addr,
 		      thd->priv_user, db); /* purecov: inspected */
@@ -1970,7 +1991,8 @@ static bool check_db_used(THD *thd,TABLE_LIST *tables)
 }
 
 
-static bool check_merge_table_access(THD *thd, char *db, TABLE_LIST *table_list)
+static bool check_merge_table_access(THD *thd, char *db,
+				     TABLE_LIST *table_list)
 {
   int error=0;
   if (table_list)
@@ -2463,8 +2485,8 @@ TABLE_LIST *add_table_to_list(Table_ident *table, LEX_STRING *alias,
     DBUG_RETURN(0);				// End of memory
   alias_str= alias ? alias->str : table->table.str;
   if (table->table.length > NAME_LEN ||
-      table->db.str && table->db.length > NAME_LEN ||
+      check_table_name(table->table.str,table->table.length) ||
-      check_table_name(table->table.str,table->table.length))
+      table->db.str && check_db_name(table->db.str))
   {
     net_printf(&thd->net,ER_WRONG_TABLE_NAME,table->table.str);
     DBUG_RETURN(0);

sql/table.cc

@@ -1033,7 +1033,7 @@ char *get_field(MEM_ROOT *mem, TABLE *table, uint fieldnr)
 
 bool check_db_name(const char *name)
 {
-  const char *start=end;
+  const char *start=name;
   while (*name)
   {
 #if defined(USE_MB) && defined(USE_MB_IDENT)