summaryrefslogtreecommitdiff
path: root/sql/handler.cc
blob: 56db355f45e020ce7cf90f616f3961297bf814e5 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
/* Copyright (C) 2000 MySQL AB & MySQL Finland AB & TCX DataKonsult AB

   This program is free software; you can redistribute it and/or modify
   it under the terms of the GNU General Public License as published by
   the Free Software Foundation; either version 2 of the License, or
   (at your option) any later version.

   This program is distributed in the hope that it will be useful,
   but WITHOUT ANY WARRANTY; without even the implied warranty of
   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
   GNU General Public License for more details.

   You should have received a copy of the GNU General Public License
   along with this program; if not, write to the Free Software
   Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA */


/* Handler-calling-functions */

#ifdef __GNUC__
#pragma implementation				// gcc: Class implementation
#endif

#include "mysql_priv.h"
#include "ha_heap.h"
#include "ha_myisam.h"
#include "ha_myisammrg.h"
#ifdef HAVE_ISAM
#include "ha_isam.h"
#include "ha_isammrg.h"
#endif
#ifdef HAVE_BERKELEY_DB
#include "ha_berkeley.h"
#endif
#ifdef HAVE_INNOBASE_DB
#include "ha_innodb.h"
#else
#define innobase_query_caching_of_table_permitted(X,Y,Z) 1
#endif
#include <myisampack.h>
#include <errno.h>

	/* static functions defined in this file */

static int NEAR_F delete_file(const char *name,const char *ext,int extflag);

ulong ha_read_count, ha_write_count, ha_delete_count, ha_update_count,
      ha_read_key_count, ha_read_next_count, ha_read_prev_count,
      ha_read_first_count, ha_read_last_count,
      ha_commit_count, ha_rollback_count,
      ha_read_rnd_count, ha_read_rnd_next_count;

const char *ha_table_type[] = {
  "", "DIAB_ISAM","HASH","MISAM","PISAM","RMS_ISAM","HEAP", "ISAM",
  "MRG_ISAM","MYISAM", "MRG_MYISAM", "BDB", "INNODB", "GEMINI", "?", "?",NullS
};

TYPELIB ha_table_typelib=
{
  array_elements(ha_table_type)-3, "", ha_table_type
};

const char *ha_row_type[] = {
  "", "FIXED", "DYNAMIC", "COMPRESSED","?","?","?"
};

const char *tx_isolation_names[] =
{ "READ-UNCOMMITTED", "READ-COMMITTED", "REPEATABLE-READ", "SERIALIZABLE",
  NullS};
TYPELIB tx_isolation_typelib= {array_elements(tx_isolation_names)-1,"",
			       tx_isolation_names};

	/* Use other database handler if databasehandler is not incompiled */

enum db_type ha_checktype(enum db_type database_type)
{
  switch (database_type) {
#ifdef HAVE_BERKELEY_DB
  case DB_TYPE_BERKELEY_DB:
    return(berkeley_skip ? DB_TYPE_MYISAM : database_type);
#endif
#ifdef HAVE_INNOBASE_DB
  case DB_TYPE_INNODB:
    return(innodb_skip ? DB_TYPE_MYISAM : database_type);
#endif
#ifndef NO_HASH
  case DB_TYPE_HASH:
#endif
#ifdef HAVE_ISAM
  case DB_TYPE_ISAM:
    return (isam_skip ? DB_TYPE_MYISAM : database_type);
  case DB_TYPE_MRG_ISAM:
    return (isam_skip ? DB_TYPE_MRG_MYISAM : database_type);
#else
  case DB_TYPE_MRG_ISAM:
    return (DB_TYPE_MRG_MYISAM);
#endif
  case DB_TYPE_HEAP:
  case DB_TYPE_MYISAM:
  case DB_TYPE_MRG_MYISAM:
    return (database_type);			/* Database exists on system */
  default:
    break;
  }
  return(DB_TYPE_MYISAM);			/* Use this as default */
} /* ha_checktype */


handler *get_new_handler(TABLE *table, enum db_type db_type)
{
  switch (db_type) {
#ifndef NO_HASH
  return new ha_hash(table);
#endif
#ifdef HAVE_ISAM
  case DB_TYPE_MRG_ISAM:
    return new ha_isammrg(table);
  case DB_TYPE_ISAM:
    return new ha_isam(table);
#else
  case DB_TYPE_MRG_ISAM:
    return new ha_myisammrg(table);
#endif
#ifdef HAVE_BERKELEY_DB
  case DB_TYPE_BERKELEY_DB:
    return new ha_berkeley(table);
#endif
#ifdef HAVE_INNOBASE_DB
  case DB_TYPE_INNODB:
    return new ha_innobase(table);
#endif
  case DB_TYPE_HEAP:
    return new ha_heap(table);
  default:					// should never happen
  {
    enum db_type def=(enum db_type) current_thd->variables.table_type;
    /* Try first with 'default table type' */
    if (db_type != def)
      return get_new_handler(table, def);
  }
  /* Fall back to MyISAM */
  case DB_TYPE_MYISAM:
    return new ha_myisam(table);
  case DB_TYPE_MRG_MYISAM:
    return new ha_myisammrg(table);
  }
}

int ha_init()
{
#ifdef HAVE_BERKELEY_DB
  if (!berkeley_skip)
  {
    int error;
    if ((error=berkeley_init()))
      return error;
    if (!berkeley_skip)				// If we couldn't use handler
      opt_using_transactions=1;
    else
      have_berkeley_db=SHOW_OPTION_DISABLED;
  }
#endif
#ifdef HAVE_INNOBASE_DB
  if (!innodb_skip)
  {
    if (innobase_init())
      return -1;
    if (!innodb_skip)				// If we couldn't use handler
      opt_using_transactions=1;
    else
      have_innodb=SHOW_OPTION_DISABLED;
  }
#endif
  return 0;
}

	/* close, flush or restart databases */
	/* Ignore this for other databases than ours */

int ha_panic(enum ha_panic_function flag)
{
  int error=0;
#ifndef NO_HASH
  error|=h_panic(flag);			/* fix hash */
#endif
#ifdef HAVE_ISAM
  error|=mrg_panic(flag);
  error|=nisam_panic(flag);
#endif
  error|=heap_panic(flag);
  error|=mi_panic(flag);
  error|=myrg_panic(flag);
#ifdef HAVE_BERKELEY_DB
  if (!berkeley_skip)
    error|=berkeley_end();
#endif
#ifdef HAVE_INNOBASE_DB
  if (!innodb_skip)
    error|=innobase_end();
#endif
  return error;
} /* ha_panic */

void ha_drop_database(char* path)
{
#ifdef HAVE_INNOBASE_DB
  if (!innodb_skip)
    innobase_drop_database(path);
#endif
}

void ha_close_connection(THD* thd)
{
#ifdef HAVE_INNOBASE_DB
  if (!innodb_skip)
    innobase_close_connection(thd);
#endif
}

/*
  This is used to commit or rollback a single statement depending on the value
  of error. Note that if the autocommit is on, then the following call inside
  InnoDB will commit or rollback the whole transaction (= the statement). The
  autocommit mechanism built into InnoDB is based on counting locks, but if
  the user has used LOCK TABLES then that mechanism does not know to do the
  commit.
*/

int ha_autocommit_or_rollback(THD *thd, int error)
{
  DBUG_ENTER("ha_autocommit_or_rollback");
#ifdef USING_TRANSACTIONS
  if (opt_using_transactions)
  {
    if (!error)
    {
      if (ha_commit_stmt(thd))
	error=1;
    }
    else
      (void) ha_rollback_stmt(thd);

    thd->variables.tx_isolation=thd->session_tx_isolation;
  }
#endif
  DBUG_RETURN(error);
}

/*
  This function is called when MySQL writes the log segment of a
  transaction to the binlog. It is called when the LOCK_log mutex is
  reserved. Here we communicate to transactional table handlers what
  binlog position corresponds to the current transaction. The handler
  can store it and in recovery print to the user, so that the user
  knows from what position in the binlog to start possible
  roll-forward, for example, if the crashed server was a slave in
  replication. This function also calls the commit of the table
  handler, because the order of transactions in the log of the table
  handler must be the same as in the binlog.
  NOTE that to eliminate the bottleneck of the group commit, we do not
  flush the handler log files here, but only later in a call of
  ha_commit_complete().

  arguments:
  thd:           the thread handle of the current connection
  log_file_name: latest binlog file name
  end_offset:	 the offset in the binlog file up to which we wrote
  return value:  0 if success, 1 if error
*/

int ha_report_binlog_offset_and_commit(THD *thd,
				       char *log_file_name,
				       my_off_t end_offset)
{
  int  error= 0;
#ifdef HAVE_INNOBASE_DB
  THD_TRANS *trans;
  trans = &thd->transaction.all;
  if (trans->innobase_tid)
  {
    if ((error=innobase_report_binlog_offset_and_commit(thd,
							trans->innobase_tid,
							log_file_name,
							end_offset)))
    {
      my_error(ER_ERROR_DURING_COMMIT, MYF(0), error);
      error=1;
    }
  }
#endif
  return error;
}

/*
  Flushes the handler log files (if my.cnf settings do not free us from it)
  after we have called ha_report_binlog_offset_and_commit(). To eliminate
  the bottleneck from the group commit, this should be called when
  LOCK_log has been released in log.cc.

  arguments:
  thd:           the thread handle of the current connection
  return value:  always 0
*/

int ha_commit_complete(THD *thd)
{
#ifdef HAVE_INNOBASE_DB
  THD_TRANS *trans;
  trans = &thd->transaction.all;
  if (trans->innobase_tid)
  {
    innobase_commit_complete(trans->innobase_tid);

    trans->innodb_active_trans=0;
  }
#endif
  return 0;
}

/*
  This function should be called when MySQL sends rows of a SELECT result set
  or the EOF mark to the client. It releases a possible adaptive hash index
  S-latch held by thd in InnoDB and also releases a possible InnoDB query
  FIFO ticket to enter InnoDB. To save CPU time, InnoDB allows a thd to
  keep them over several calls of the InnoDB handler interface when a join
  is executed. But when we let the control to pass to the client they have
  to be released because if the application program uses mysql_use_result(),
  it may deadlock on the S-latch if the application on another connection
  performs another SQL query. In MySQL-4.1 this is even more important because
  there a connection can have several SELECT queries open at the same time.

  arguments:
  thd:           the thread handle of the current connection
  return value:  always 0
*/

int ha_release_temporary_latches(THD *thd)
{
#ifdef HAVE_INNOBASE_DB
  THD_TRANS *trans;
  trans = &thd->transaction.all;
  if (trans->innobase_tid)
    innobase_release_temporary_latches(trans->innobase_tid);
#endif
  return 0;
}

int ha_commit_trans(THD *thd, THD_TRANS* trans)
{
  int error=0;
  DBUG_ENTER("ha_commit");
#ifdef USING_TRANSACTIONS
  if (opt_using_transactions)
  {
    bool operation_done= 0;
    bool transaction_commited= 0;

    /* Update the binary log if we have cached some queries */
    if (trans == &thd->transaction.all && mysql_bin_log.is_open() &&
	my_b_tell(&thd->transaction.trans_log))
    {
      mysql_bin_log.write(thd, &thd->transaction.trans_log, 1);
      reinit_io_cache(&thd->transaction.trans_log,
		      WRITE_CACHE, (my_off_t) 0, 0, 1);
      thd->transaction.trans_log.end_of_file= max_binlog_cache_size;
    }
#ifdef HAVE_BERKELEY_DB
    if (trans->bdb_tid)
    {
      if ((error=berkeley_commit(thd,trans->bdb_tid)))
      {
	my_error(ER_ERROR_DURING_COMMIT, MYF(0), error);
	error=1;
      }
      else
	if (!(thd->options & OPTION_BEGIN))
	  transaction_commited= 1; 
      trans->bdb_tid=0;
    }
#endif
#ifdef HAVE_INNOBASE_DB
    if (trans->innobase_tid)
    {
      if ((error=innobase_commit(thd,trans->innobase_tid)))
      {
	my_error(ER_ERROR_DURING_COMMIT, MYF(0), error);
	error=1;
      }
      trans->innodb_active_trans=0;
      if (trans == &thd->transaction.all)
	operation_done= transaction_commited= 1;
    }
#endif
#ifdef HAVE_QUERY_CACHE
    if (transaction_commited && thd->transaction.changed_tables)
      query_cache.invalidate(thd->transaction.changed_tables);
#endif /*HAVE_QUERY_CACHE*/
    if (error && trans == &thd->transaction.all && mysql_bin_log.is_open())
      sql_print_error("Error: Got error during commit;  Binlog is not up to date!");
    thd->variables.tx_isolation=thd->session_tx_isolation;
    if (operation_done)
    {
      statistic_increment(ha_commit_count,&LOCK_status);
      thd->transaction.cleanup();
    }
  }
#endif // using transactions
  DBUG_RETURN(error);
}


int ha_rollback_trans(THD *thd, THD_TRANS *trans)
{
  int error=0;
  DBUG_ENTER("ha_rollback");
#ifdef USING_TRANSACTIONS
  if (opt_using_transactions)
  {
    bool operation_done=0;
#ifdef HAVE_BERKELEY_DB
    if (trans->bdb_tid)
    {
      if ((error=berkeley_rollback(thd, trans->bdb_tid)))
      {
	my_error(ER_ERROR_DURING_ROLLBACK, MYF(0), error);
	error=1;
      }
      trans->bdb_tid=0;
      operation_done=1;
    }
#endif
#ifdef HAVE_INNOBASE_DB
    if (trans->innobase_tid)
    {
      if ((error=innobase_rollback(thd, trans->innobase_tid)))
      {
	my_error(ER_ERROR_DURING_ROLLBACK, MYF(0), error);
	error=1;
      }
      trans->innodb_active_trans=0;
      operation_done=1;
    }
#endif
    if (trans == &thd->transaction.all)
    {
      /* 
         Update the binary log with a BEGIN/ROLLBACK block if we have cached some
         queries and we updated some non-transactional table. Such cases should
         be rare (updating a non-transactional table inside a transaction...).
      */
      if (unlikely((thd->options & OPTION_STATUS_NO_TRANS_UPDATE) &&
                   mysql_bin_log.is_open() &&
                   my_b_tell(&thd->transaction.trans_log)))
        mysql_bin_log.write(thd, &thd->transaction.trans_log, 0);
      /* Flushed or not, empty the binlog cache */
      reinit_io_cache(&thd->transaction.trans_log,
                      WRITE_CACHE, (my_off_t) 0, 0, 1);
      thd->transaction.trans_log.end_of_file= max_binlog_cache_size;
    }
    thd->variables.tx_isolation=thd->session_tx_isolation;
    if (operation_done)
    {
      statistic_increment(ha_rollback_count,&LOCK_status);
      thd->transaction.cleanup();
    }
  }
#endif /* USING_TRANSACTIONS */
  DBUG_RETURN(error);
}


/*
  Rolls the current transaction back to a savepoint.
  Return value: 0 if success, 1 if there was not a savepoint of the given
  name.
  NOTE: how do we handle this (unlikely but legal) case:
  [transaction] + [update to non-trans table] + [rollback to savepoint] ?
  The problem occurs when a savepoint is before the update to the
  non-transactional table. Then when there's a rollback to the savepoint, if we
  simply truncate the binlog cache, we lose the part of the binlog cache where
  the update is. If we want to not lose it, we need to write the SAVEPOINT
  command and the ROLLBACK TO SAVEPOINT command to the binlog cache. The latter
  is easy: it's just write at the end of the binlog cache, but the former should
  be *inserted* to the place where the user called SAVEPOINT. The solution is
  that when the user calls SAVEPOINT, we write it to the binlog cache (so no
  need to later insert it). As transactions are never intermixed in the binary log
  (i.e. they are serialized), we won't have conflicts with savepoint names when
  using mysqlbinlog or in the slave SQL thread.
  Then when ROLLBACK TO SAVEPOINT is called, if we updated some
  non-transactional table, we don't truncate the binlog cache but instead write
  ROLLBACK TO SAVEPOINT to it; otherwise we truncate the binlog cache (which
  will chop the SAVEPOINT command from the binlog cache, which is good as in
  that case there is no need to have it in the binlog).
*/

int ha_rollback_to_savepoint(THD *thd, char *savepoint_name)
{
  my_off_t binlog_cache_pos=0;
  bool operation_done=0;
  int error=0;
  DBUG_ENTER("ha_rollback_to_savepoint");
#ifdef USING_TRANSACTIONS
  if (opt_using_transactions)
  {
#ifdef HAVE_INNOBASE_DB
    /*
    Retrieve the trans_log binlog cache position corresponding to the
    savepoint, and if the rollback is successful inside InnoDB reset the write
    position in the binlog cache to what it was at the savepoint.
    */
    if ((error=innobase_rollback_to_savepoint(thd, savepoint_name,
						  &binlog_cache_pos)))
    {
      my_error(ER_ERROR_DURING_ROLLBACK, MYF(0), error);
      error=1;
    }
    else
    {
      /* 
         Write ROLLBACK TO SAVEPOINT to the binlog cache if we have updated some
         non-transactional table. Otherwise, truncate the binlog cache starting
         from the SAVEPOINT command.
      */
      if (unlikely((thd->options & OPTION_STATUS_NO_TRANS_UPDATE) &&
                   mysql_bin_log.is_open() &&
                   my_b_tell(&thd->transaction.trans_log)))
      {
        Query_log_event qinfo(thd, thd->query, thd->query_length, TRUE);
        if (mysql_bin_log.write(&qinfo))
          error= 1;
      }
      else
        reinit_io_cache(&thd->transaction.trans_log, WRITE_CACHE,
                        binlog_cache_pos, 0, 0);
    }
    operation_done=1;
#endif
    if (operation_done)
      statistic_increment(ha_rollback_count,&LOCK_status);
  }
#endif /* USING_TRANSACTIONS */

  DBUG_RETURN(error);
}


/*
Sets a transaction savepoint.
Return value: always 0, that is, succeeds always
*/

int ha_savepoint(THD *thd, char *savepoint_name)
{
  my_off_t binlog_cache_pos=0;
  int error=0;
  DBUG_ENTER("ha_savepoint");
#ifdef USING_TRANSACTIONS
  if (opt_using_transactions)
  {
    binlog_cache_pos=my_b_tell(&thd->transaction.trans_log);
#ifdef HAVE_INNOBASE_DB
    innobase_savepoint(thd,savepoint_name, binlog_cache_pos);
#endif
    /* Write it to the binary log (see comments of ha_rollback_to_savepoint). */
    if (mysql_bin_log.is_open())
    {
      Query_log_event qinfo(thd, thd->query, thd->query_length, TRUE);
      if (mysql_bin_log.write(&qinfo))
	error= 1;
    }
  }
#endif /* USING_TRANSACTIONS */
  DBUG_RETURN(error);
}

bool ha_flush_logs()
{
  bool result=0;
#ifdef HAVE_BERKELEY_DB
  if (!berkeley_skip && berkeley_flush_logs())
    result=1;
#endif
#ifdef HAVE_INNOBASE_DB
  if (!innodb_skip && innobase_flush_logs())
    result=1;
#endif
  return result;
}

/*
  This should return ENOENT if the file doesn't exists.
  The .frm file will be deleted only if we return 0 or ENOENT
*/

int ha_delete_table(enum db_type table_type, const char *path)
{
  handler *file=get_new_handler((TABLE*) 0, table_type);
  if (!file)
    return ENOENT;
  int error=file->delete_table(path);
  delete file;
  return error;
}

void ha_store_ptr(byte *buff, uint pack_length, my_off_t pos)
{
  switch (pack_length) {
#if SIZEOF_OFF_T > 4
  case 8: mi_int8store(buff,pos); break;
  case 7: mi_int7store(buff,pos); break;
  case 6: mi_int6store(buff,pos); break;
  case 5: mi_int5store(buff,pos); break;
#endif
  case 4: mi_int4store(buff,pos); break;
  case 3: mi_int3store(buff,pos); break;
  case 2: mi_int2store(buff,(uint) pos); break;
  case 1: buff[0]= (uchar) pos; break;
  }
  return;
}

my_off_t ha_get_ptr(byte *ptr, uint pack_length)
{
  my_off_t pos;
  switch (pack_length) {
#if SIZEOF_OFF_T > 4
  case 8:
    pos= (my_off_t) mi_uint8korr(ptr);
    break;
  case 7:
    pos= (my_off_t) mi_uint7korr(ptr);
    break;
  case 6:
    pos= (my_off_t) mi_uint6korr(ptr);
    break;
  case 5:
    pos= (my_off_t) mi_uint5korr(ptr);
    break;
#endif
  case 4:
    pos= (my_off_t) mi_uint4korr(ptr);
    break;
  case 3:
    pos= (my_off_t) mi_uint3korr(ptr);
    break;
  case 2:
    pos= (my_off_t) mi_uint2korr(ptr);
    break;
  case 1:
    pos= (my_off_t) mi_uint2korr(ptr);
    break;
  default:
    pos=0;					// Impossible
    break;
  }
 return pos;
}

/****************************************************************************
** General handler functions
****************************************************************************/

	/* Open database-handler. Try O_RDONLY if can't open as O_RDWR */
	/* Don't wait for locks if not HA_OPEN_WAIT_IF_LOCKED is set */

int handler::ha_open(const char *name, int mode, int test_if_locked)
{
  int error;
  DBUG_ENTER("handler::open");
  DBUG_PRINT("enter",("name: %s  db_type: %d  db_stat: %d  mode: %d  lock_test: %d",
		      name, table->db_type, table->db_stat, mode,
		      test_if_locked));

  if ((error=open(name,mode,test_if_locked)))
  {
    if ((error == EACCES || error == EROFS) && mode == O_RDWR &&
	(table->db_stat & HA_TRY_READ_ONLY))
    {
      table->db_stat|=HA_READ_ONLY;
      error=open(name,O_RDONLY,test_if_locked);
    }
  }
  if (error)
  {
    my_errno=error;			/* Safeguard */
    DBUG_PRINT("error",("error: %d  errno: %d",error,errno));
  }
  else
  {
    if (table->db_options_in_use & HA_OPTION_READ_ONLY_DATA)
      table->db_stat|=HA_READ_ONLY;
    (void) extra(HA_EXTRA_NO_READCHECK);	// Not needed in SQL

    if (!alloc_root_inited(&table->mem_root))	// If temporary table
      ref=(byte*) sql_alloc(ALIGN_SIZE(ref_length)*2);
    else
      ref=(byte*) alloc_root(&table->mem_root, ALIGN_SIZE(ref_length)*2);
    if (!ref)
    {
      close();
      error=HA_ERR_OUT_OF_MEM;
    }
    else
      dupp_ref=ref+ALIGN_SIZE(ref_length);
  }
  DBUG_RETURN(error);
}

int handler::check(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::backup(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::restore(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::repair(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::optimize(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::analyze(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::assign_to_keycache(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

int handler::preload_keys(THD* thd, HA_CHECK_OPT* check_opt)
{
  return HA_ADMIN_NOT_IMPLEMENTED;
}

/*
  Read first row (only) from a table
  This is never called for InnoDB or BDB tables, as these table types
  has the HA_NOT_EXACT_COUNT set.
*/

int handler::read_first_row(byte * buf, uint primary_key)
{
  register int error;
  DBUG_ENTER("handler::read_first_row");

  statistic_increment(ha_read_first_count,&LOCK_status);

  /*
    If there is very few deleted rows in the table, find the first row by
    scanning the table.
  */
  if (deleted < 10 || primary_key >= MAX_KEY || 
      !(index_flags(primary_key) & HA_READ_ORDER))
  {
    (void) rnd_init();
    while ((error= rnd_next(buf)) == HA_ERR_RECORD_DELETED) ;
    (void) rnd_end();
  }
  else
  {
    /* Find the first row through the primary key */
    (void) index_init(primary_key);
    error=index_first(buf);
    (void) index_end();
  }
  DBUG_RETURN(error);
}


/*
  The following function is only needed for tables that may be temporary tables
  during joins
*/

int handler::restart_rnd_next(byte *buf, byte *pos)
{
  return HA_ERR_WRONG_COMMAND;
}


/* Set a timestamp in record */

void handler::update_timestamp(byte *record)
{
  long skr= (long) current_thd->query_start();
#ifdef WORDS_BIGENDIAN
  if (table->db_low_byte_first)
  {
    int4store(record,skr);
  }
  else
#endif
  longstore(record,skr);
  return;
}

/*
  Updates field with field_type NEXT_NUMBER according to following:
  if field = 0 change field to the next free key in database.
*/

void handler::update_auto_increment()
{
  longlong nr;
  THD *thd;
  DBUG_ENTER("update_auto_increment");
  if (table->next_number_field->val_int() != 0 ||
      table->auto_increment_field_not_null &&
      current_thd->variables.sql_mode & MODE_NO_AUTO_VALUE_ON_ZERO)
  {
    table->auto_increment_field_not_null= false;
    auto_increment_column_changed=0;
    DBUG_VOID_RETURN;
  }
  table->auto_increment_field_not_null= false;
  thd=current_thd;
  if ((nr=thd->next_insert_id))
    thd->next_insert_id=0;			// Clear after use
  else
    nr=get_auto_increment();
  if (!table->next_number_field->store(nr))
    thd->insert_id((ulonglong) nr);
  else
    thd->insert_id(table->next_number_field->val_int());
  auto_increment_column_changed=1;
  DBUG_VOID_RETURN;
}


longlong handler::get_auto_increment()
{
  longlong nr;
  int error;

  (void) extra(HA_EXTRA_KEYREAD);
  index_init(table->next_number_index);
  if (!table->next_number_key_offset)
  {						// Autoincrement at key-start
    error=index_last(table->record[1]);
  }
  else
  {
    byte key[MAX_KEY_LENGTH];
    key_copy(key,table,table->next_number_index,
             table->next_number_key_offset);
    error=index_read(table->record[1], key, table->next_number_key_offset,
                     HA_READ_PREFIX_LAST);
  }

  if (error)
    nr=1;
  else
    nr=(longlong) table->next_number_field->
      val_int_offset(table->rec_buff_length)+1;
  index_end();
  (void) extra(HA_EXTRA_NO_KEYREAD);
  return nr;
}

	/* Print error that we got from handler function */

void handler::print_error(int error, myf errflag)
{
  DBUG_ENTER("print_error");
  DBUG_PRINT("enter",("error: %d",error));

  int textno=ER_GET_ERRNO;
  switch (error) {
  case EACCES:
    textno=ER_OPEN_AS_READONLY;
    break;
  case EAGAIN:
    textno=ER_FILE_USED;
    break;
  case ENOENT:
    textno=ER_FILE_NOT_FOUND;
    break;
  case HA_ERR_KEY_NOT_FOUND:
  case HA_ERR_NO_ACTIVE_RECORD:
  case HA_ERR_END_OF_FILE:
    textno=ER_KEY_NOT_FOUND;
    break;
  case HA_ERR_WRONG_MRG_TABLE_DEF:
    textno=ER_WRONG_MRG_TABLE;
    break;
  case HA_ERR_FOUND_DUPP_KEY:
  {
    uint key_nr=get_dup_key(error);
    if ((int) key_nr >= 0)
    {
      /* Write the dupplicated key in the error message */
      char key[MAX_KEY_LENGTH];
      String str(key,sizeof(key),system_charset_info);
      key_unpack(&str,table,(uint) key_nr);
      uint max_length=MYSQL_ERRMSG_SIZE-(uint) strlen(ER(ER_DUP_ENTRY));
      if (str.length() >= max_length)
      {
	str.length(max_length-4);
	str.append("...");
      }
      my_error(ER_DUP_ENTRY,MYF(0),str.c_ptr(),key_nr+1);
      DBUG_VOID_RETURN;
    }
    textno=ER_DUP_KEY;
    break;
  }
  case HA_ERR_FOUND_DUPP_UNIQUE:
    textno=ER_DUP_UNIQUE;
    break;
  case HA_ERR_RECORD_CHANGED:
    textno=ER_CHECKREAD;
    break;
  case HA_ERR_CRASHED:
    textno=ER_NOT_KEYFILE;
    break;
  case HA_ERR_CRASHED_ON_USAGE:
    textno=ER_CRASHED_ON_USAGE;
    break;
  case HA_ERR_CRASHED_ON_REPAIR:
    textno=ER_CRASHED_ON_REPAIR;
    break;
  case HA_ERR_OUT_OF_MEM:
    my_error(ER_OUT_OF_RESOURCES,errflag);
    DBUG_VOID_RETURN;
  case HA_ERR_WRONG_COMMAND:
    textno=ER_ILLEGAL_HA;
    break;
  case HA_ERR_OLD_FILE:
    textno=ER_OLD_KEYFILE;
    break;
  case HA_ERR_UNSUPPORTED:
    textno=ER_UNSUPPORTED_EXTENSION;
    break;
  case HA_ERR_RECORD_FILE_FULL:
    textno=ER_RECORD_FILE_FULL;
    break;
  case HA_ERR_LOCK_WAIT_TIMEOUT:
    textno=ER_LOCK_WAIT_TIMEOUT;
    break;
  case HA_ERR_LOCK_TABLE_FULL:
    textno=ER_LOCK_TABLE_FULL;
    break;
  case HA_ERR_LOCK_DEADLOCK:
    textno=ER_LOCK_DEADLOCK;
    break;
  case HA_ERR_READ_ONLY_TRANSACTION:
    textno=ER_READ_ONLY_TRANSACTION;
    break;
  case HA_ERR_CANNOT_ADD_FOREIGN:
    textno=ER_CANNOT_ADD_FOREIGN;
    break;
  case HA_ERR_ROW_IS_REFERENCED:
    textno=ER_ROW_IS_REFERENCED;
    break;
  case HA_ERR_NO_REFERENCED_ROW:
    textno=ER_NO_REFERENCED_ROW;
    break;
  default:
    {
      my_error(ER_GET_ERRNO,errflag,error);
      DBUG_VOID_RETURN;
    }
  }
  my_error(textno,errflag,table->table_name,error);
  DBUG_VOID_RETURN;
}


/* Return key if error because of duplicated keys */

uint handler::get_dup_key(int error)
{
  DBUG_ENTER("get_dup_key");
  table->file->errkey  = (uint) -1;
  if (error == HA_ERR_FOUND_DUPP_KEY || error == HA_ERR_FOUND_DUPP_UNIQUE)
    info(HA_STATUS_ERRKEY | HA_STATUS_NO_LOCK);
  DBUG_RETURN(table->file->errkey);
}


int handler::delete_table(const char *name)
{
  int error=0;
  for (const char **ext=bas_ext(); *ext ; ext++)
  {
    if (delete_file(name,*ext,2))
    {
      if ((error=errno) != ENOENT)
	break;
    }
  }
  return error;
}


int handler::rename_table(const char * from, const char * to)
{
  DBUG_ENTER("handler::rename_table");
  for (const char **ext=bas_ext(); *ext ; ext++)
  {
    if (rename_file_ext(from,to,*ext))
      DBUG_RETURN(my_errno);
  }
  DBUG_RETURN(0);
}

/*
  Tell the handler to turn on or off logging to the handler's recovery log
*/

int ha_recovery_logging(THD *thd, bool on)
{
  int error=0;

  DBUG_ENTER("ha_recovery_logging");
  DBUG_RETURN(error);
}

int handler::index_next_same(byte *buf, const byte *key, uint keylen)
{
  int error;
  if (!(error=index_next(buf)))
  {
    if (key_cmp(table, key, active_index, keylen))
    {
      table->status=STATUS_NOT_FOUND;
      error=HA_ERR_END_OF_FILE;
    }
  }
  return error;
}


/*
  This is called to delete all rows in a table
  If the handler don't support this, then this function will
  return HA_ERR_WRONG_COMMAND and MySQL will delete the rows one
  by one.
*/

int handler::delete_all_rows()
{
  return (my_errno=HA_ERR_WRONG_COMMAND);
}

bool handler::caching_allowed(THD* thd, char* table_key,
			      uint key_length, uint8 cache_type)
{
  if (cache_type == HA_CACHE_TBL_ASKTRANSACT)
    return innobase_query_caching_of_table_permitted(thd, table_key,
						     key_length);
  else
    return 1;
}

/****************************************************************************
** Some general functions that isn't in the handler class
****************************************************************************/

	/* Initiates table-file and calls apropriate database-creator */
	/* Returns 1 if something got wrong */

int ha_create_table(const char *name, HA_CREATE_INFO *create_info,
		    bool update_create_info)
{
  int error;
  TABLE table;
  DBUG_ENTER("ha_create_table");

  if (openfrm(name,"",0,(uint) READ_ALL, 0, &table))
    DBUG_RETURN(1);
  if (update_create_info)
  {
    update_create_info_from_table(create_info, &table);
    if (table.file->table_flags() & HA_DROP_BEFORE_CREATE)
      table.file->delete_table(name);		// Needed for BDB tables
  }
  error=table.file->create(name,&table,create_info);
  VOID(closefrm(&table));
  if (error)
  {
    if (table.db_type == DB_TYPE_INNODB)
    {
      /* Creation of InnoDB table cannot fail because of an OS error:
	 put error as the number */
      my_error(ER_CANT_CREATE_TABLE,MYF(ME_BELL+ME_WAITTANG),name,error);
    }
    else
      my_error(ER_CANT_CREATE_TABLE,MYF(ME_BELL+ME_WAITTANG),name,my_errno);
  }
  DBUG_RETURN(error != 0);
}

static int NEAR_F delete_file(const char *name,const char *ext,int extflag)
{
  char buff[FN_REFLEN];
  VOID(fn_format(buff,name,"",ext,extflag | 4));
  return(my_delete_with_symlink(buff,MYF(MY_WME)));
}

void st_ha_check_opt::init()
{
  flags= sql_flags= 0;
  sort_buffer_size = current_thd->variables.myisam_sort_buff_size;
}


/*****************************************************************************
  Key cache handling.

  This code is only relevant for ISAM/MyISAM tables

  key_cache->cache may be 0 only in the case where a key cache is not
  initialized or when we where not able to init the key cache in a previous
  call to ha_init_key_cache() (probably out of memory)
*****************************************************************************/

/* Init a key cache if it has not been initied before */


int ha_init_key_cache(const char *name, KEY_CACHE_VAR *key_cache)
{
  DBUG_ENTER("ha_init_key_cache");

  if (!key_cache->cache)
  {
    pthread_mutex_lock(&LOCK_global_system_variables);
    long tmp_buff_size= (long) key_cache->buff_size;
    long tmp_block_size= (long) key_cache->block_size;
    pthread_mutex_unlock(&LOCK_global_system_variables);
    DBUG_RETURN(!init_key_cache(&key_cache->cache,
				tmp_block_size,
				tmp_buff_size,
				key_cache));
  }
  DBUG_RETURN(0);
}


/* Resize key cache */

int ha_resize_key_cache(KEY_CACHE_VAR *key_cache)
{
  DBUG_ENTER("ha_resize_key_cache");

  if (key_cache->cache)
  {
    pthread_mutex_lock(&LOCK_global_system_variables);
    long tmp_buff_size= (long) key_cache->buff_size;
    long tmp_block_size= (long) key_cache->block_size;
    pthread_mutex_unlock(&LOCK_global_system_variables);
    DBUG_RETURN(!resize_key_cache(&key_cache->cache, tmp_block_size,
				  tmp_buff_size));
  }
  DBUG_RETURN(0);
}


/* Change parameters for key cache (like size) */

int ha_change_key_cache_param(KEY_CACHE_VAR *key_cache)
{
  if (key_cache->cache)
    change_key_cache_param(key_cache->cache);
  return 0;
}

/* Free memory allocated by a key cache */

int ha_end_key_cache(KEY_CACHE_VAR *key_cache)
{
  if (key_cache->cache)
  {
    end_key_cache(key_cache->cache, 1);		// Can never fail
    key_cache->cache= 0;
  }
  return 0;
}

/* Move all tables from one key cache to another one */

int ha_change_key_cache(KEY_CACHE_VAR *old_key_cache,
			KEY_CACHE_VAR *new_key_cache)
{
  mi_change_key_cache(old_key_cache, new_key_cache);
  return 0;
}