summaryrefslogtreecommitdiffstats
path: root/extra/replace.c
blob: 8b20f812be005a208107583a8a2ca1089c1da461 (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
/*
   Copyright (c) 2000, 2014, Oracle and/or its affiliates

   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; version 2 of
   the License.

   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., 51 Franklin St, Fifth Floor, Boston, MA
   02110-1335  USA */

/*
  Replace strings in textfile

  This program replaces strings in files or from stdin to stdout.
  It accepts a list of from-string/to-string pairs and replaces
  each occurrence of a from-string with the corresponding to-string.
  The first occurrence of a found string is matched. If there is more
  than one possibility for the string to replace, longer matches
  are preferred before shorter matches.

  Special characters in from string:
  \^    Match start of line.
  \$	Match end of line.
  \b	Match space-character, start of line or end of line.
        For end \b the next replace starts locking at the end space-character.
        An \b alone or in a string matches only a space-character.
  \r, \t, \v as in C.
  The programs make a DFA-state-machine of the strings and the speed isn't
  dependent on the count of replace-strings (only of the number of replaces).
  A line is assumed ending with \n or \0.
  There are no limit except memory on length of strings.

  Written by Monty.
  fill_buffer_retaining() is taken from gnu-grep and modified.
*/

#include <my_global.h>
#include <m_ctype.h>
#include <my_sys.h>
#include <m_string.h>
#include <errno.h>

#define PC_MALLOC		256	/* Bytes for pointers */
#define PS_MALLOC		512	/* Bytes for data */

typedef struct st_pointer_array {		/* when using array-strings */
  TYPELIB typelib;				/* Pointer to strings */
  uchar *str;					/* Strings is here */
  uint8	*flag;					/* Flag about each var. */
  uint  array_allocs,max_count,length,max_length;
} POINTER_ARRAY;

#define SPACE_CHAR	256
#define START_OF_LINE	257
#define END_OF_LINE	258
#define LAST_CHAR_CODE	259

typedef struct st_replace {
  uint8 found;
  struct st_replace *next[256];
} REPLACE;

typedef struct st_replace_found {
  my_bool found;
  char *replace_string;
  uint to_offset;
  int from_offset;
} REPLACE_STRING;

#ifndef WORD_BIT
#define WORD_BIT (8*sizeof(uint))
#endif

	/* functions defined in this file */

static int static_get_options(int *argc,char * * *argv);
static int get_replace_strings(int *argc,char * * *argv,
				   POINTER_ARRAY *from_array,
				   POINTER_ARRAY *to_array);
static int insert_pointer_name(POINTER_ARRAY *pa, char * name);
static void free_pointer_array(POINTER_ARRAY *pa);
static int convert_pipe(REPLACE *,FILE *,FILE *);
static int convert_file(REPLACE *, char *);
static REPLACE *init_replace(char * *from, char * *to,uint count,
                             char * word_end_chars);
static uint replace_strings(REPLACE *rep, char * *start,uint *max_length,
                            char * from);
static int initialize_buffer(void);
static void reset_buffer(void);
static void free_buffer(void);

static int silent=0,verbose=0,updated=0;

	/* The main program */

int main(int argc, char *argv[])
{
  int i,error;
  char word_end_chars[256],*pos;
  POINTER_ARRAY from,to;
  REPLACE *replace;
  MY_INIT(argv[0]);

  if (static_get_options(&argc,&argv))
    exit(1);
  if (get_replace_strings(&argc,&argv,&from,&to))
    exit(1);

  for (i=1,pos=word_end_chars ; i < 256 ; i++)
    if (my_isspace(&my_charset_latin1,i))
      *pos++= (char) i;
  *pos=0;
  if (!(replace=init_replace((char**) from.typelib.type_names,
			     (char**) to.typelib.type_names,
			     (uint) from.typelib.count,word_end_chars)))
    exit(1);
  free_pointer_array(&from);
  free_pointer_array(&to);
  if (initialize_buffer())
    return 1;

  error=0;
  if (argc == 0)
    error=convert_pipe(replace,stdin,stdout);
  else
  {
    while (argc--)
    {
      error=convert_file(replace,*(argv++));
    }
  }
  free_buffer();
  my_free(replace);
  my_end(verbose ? MY_CHECK_ERROR | MY_GIVE_INFO : MY_CHECK_ERROR);
  exit(error ? 2 : 0);
  return 0;					/* No compiler warning */
} /* main */


	/* reads options */
	/* Initiates DEBUG - but no debugging here ! */

static int static_get_options(int *argc, char***argv)
{
  int help,version;
  char *pos;

  silent=verbose=help=0;

  while (--*argc > 0 && *(pos = *(++*argv)) == '-' && pos[1] != '-') {
    while (*++pos)
    {
      version=0;
      switch((*pos)) {
      case 's':
	silent=1;
	break;
      case 'v':
	verbose=1;
	break;
      case '#':
	DBUG_PUSH (++pos);
	pos= (char*) " ";			/* Skip rest of arguments */
	break;
      case 'V':
	version=1;
        /* fall through */
      case 'I':
      case '?':
	help=1;					/* Help text written */
	printf("%s  Ver 1.4 for %s at %s\n",my_progname,SYSTEM_TYPE,
	       MACHINE_TYPE);
	if (version)
	  break;
	puts("This software comes with ABSOLUTELY NO WARRANTY. This is free software,\nand you are welcome to modify and redistribute it under the GPL license\n");
	puts("This program replaces strings in files or from stdin to stdout.\n"
	     "It accepts a list of from-string/to-string pairs and replaces\n"
	     "each occurrence of a from-string with the corresponding to-string.\n"
         "The first occurrence of a found string is matched. If there is\n"
         "more than one possibility for the string to replace, longer\n"
         "matches are preferred before shorter matches.\n\n"
	     "A from-string can contain these special characters:\n"
	     "  \\^      Match start of line.\n"
	     "  \\$      Match end of line.\n"
	     "  \\b      Match space-character, start of line or end of line.\n"
	     "          For a end \\b the next replace starts locking at the end\n"
	     "          space-character. A \\b alone in a string matches only a\n"
	     "          space-character.\n");
	  printf("Usage: %s [-?svIV] from to from to ... -- [files]\n", my_progname);
	puts("or");
	  printf("Usage: %s [-?svIV] from to from to ... < fromfile > tofile\n", my_progname);
	puts("");
	puts("Options: -? or -I \"Info\"  -s \"silent\"      -v \"verbose\"");
	break;
      default:
	fprintf(stderr,"illegal option: -%c\n",*pos);
	break;
      }
    }
  }
  if (*argc == 0)
  {
    if (!help)
      my_message(0,"No replace options given",MYF(ME_BELL));
    exit(0);					/* Don't use as pipe */
  }
  return(0);
} /* static_get_options */


static int get_replace_strings(int *argc, char ***argv,
                               POINTER_ARRAY *from_array,
                               POINTER_ARRAY *to_array)
{
  char *pos;

  bzero((char*) from_array,sizeof(from_array[0]));
  bzero((char*) to_array,sizeof(to_array[0]));
  while (*argc > 0 && (*(pos = *(*argv)) != '-' || pos[1] != '-' || pos[2]))
  {
    insert_pointer_name(from_array,pos);
    (*argc)--;
    (*argv)++;
    if (!*argc || !strcmp(**argv,"--"))
    {
      my_message(0,"No to-string for last from-string",MYF(ME_BELL));
      return 1;
    }
    insert_pointer_name(to_array,**argv);
    (*argc)--;
    (*argv)++;
  }
  if (*argc)
  {					/* Skip "--" argument */
    (*argc)--;
    (*argv)++;
  }
  return 0;
}

static int insert_pointer_name(reg1 POINTER_ARRAY *pa,char * name)
{
  uint i,length,old_count;
  uchar *new_pos;
  const char **new_array;
  DBUG_ENTER("insert_pointer_name");

  if (! pa->typelib.count)
  {
    if (!(pa->typelib.type_names=(const char **)
	  my_malloc(PSI_NOT_INSTRUMENTED, ((PC_MALLOC-MALLOC_OVERHEAD)/
		     (sizeof(char *)+sizeof(*pa->flag))*
		     (sizeof(char *)+sizeof(*pa->flag))),MYF(MY_WME))))
      DBUG_RETURN(-1);
    if (!(pa->str= (uchar*) my_malloc(PSI_NOT_INSTRUMENTED,
                                      PS_MALLOC-MALLOC_OVERHEAD, MYF(MY_WME))))
    {
      my_free((void*) pa->typelib.type_names);
      DBUG_RETURN (-1);
    }
    pa->max_count=(PC_MALLOC-MALLOC_OVERHEAD)/(sizeof(uchar*)+
					       sizeof(*pa->flag));
    pa->flag= (uint8*) (pa->typelib.type_names+pa->max_count);
    pa->length=0;
    pa->max_length=PS_MALLOC-MALLOC_OVERHEAD;
    pa->array_allocs=1;
  }
  length=(uint) strlen(name)+1;
  if (pa->length+length >= pa->max_length)
  {
    pa->max_length=(pa->length+length+MALLOC_OVERHEAD+PS_MALLOC-1)/PS_MALLOC;
    pa->max_length=pa->max_length*PS_MALLOC-MALLOC_OVERHEAD;
    if (!(new_pos= (uchar*) my_realloc(PSI_NOT_INSTRUMENTED, (uchar*) pa->str,
                                      (uint) pa->max_length, MYF(MY_WME))))
      DBUG_RETURN(1);
    if (new_pos != pa->str)
    {
      my_ptrdiff_t diff=PTR_BYTE_DIFF(new_pos,pa->str);
      for (i=0 ; i < pa->typelib.count ; i++)
	pa->typelib.type_names[i]= ADD_TO_PTR(pa->typelib.type_names[i],diff,
					      char*);
      pa->str=new_pos;
    }
  }
  if (pa->typelib.count >= pa->max_count-1)
  {
    int len;
    pa->array_allocs++;
    len=(PC_MALLOC*pa->array_allocs - MALLOC_OVERHEAD);
    if (!(new_array=(const char **) my_realloc(PSI_NOT_INSTRUMENTED, (void*)(pa->typelib.type_names),
					       (uint) len/
					 (sizeof(uchar*)+sizeof(*pa->flag))*
					 (sizeof(uchar*)+sizeof(*pa->flag)),
					 MYF(MY_WME))))
      DBUG_RETURN(1);
    pa->typelib.type_names=new_array;
    old_count=pa->max_count;
    pa->max_count=len/(sizeof(uchar*) + sizeof(*pa->flag));
    pa->flag= (uint8*) (pa->typelib.type_names+pa->max_count);
    memcpy((uchar*) pa->flag,(char *) (pa->typelib.type_names+old_count),
	   old_count*sizeof(*pa->flag));
  }
  pa->flag[pa->typelib.count]=0;			/* Reset flag */
  pa->typelib.type_names[pa->typelib.count++]= (char*) (pa->str+pa->length);
  pa->typelib.type_names[pa->typelib.count]= NullS;	/* Put end-mark */
  (void) strmov((char*) pa->str + pa->length, name);
  pa->length+=length;
  DBUG_RETURN(0);
} /* insert_pointer_name */


	/* free pointer array */

static void free_pointer_array(reg1 POINTER_ARRAY *pa)
{
  if (pa->typelib.count)
  {
    pa->typelib.count=0;
    my_free((void*) pa->typelib.type_names);
    pa->typelib.type_names=0;
    my_free(pa->str);
  }
  return;
} /* free_pointer_array */


	/* Code for replace rutines */

#define SET_MALLOC_HUNC 64

typedef struct st_rep_set {
  uint  *bits;				/* Pointer to used sets */
  short	next[LAST_CHAR_CODE];		/* Pointer to next sets */
  uint	found_len;			/* Best match to date */
  int	found_offset;
  uint  table_offset;
  uint  size_of_bits;			/* For convinience */
} REP_SET;

typedef struct st_rep_sets {
  uint		count;			/* Number of sets */
  uint		extra;			/* Extra sets in buffer */
  uint		invisible;		/* Sets not chown */
  uint		size_of_bits;
  REP_SET	*set,*set_buffer;
  uint		*bit_buffer;
} REP_SETS;

typedef struct st_found_set {
  uint table_offset;
  int found_offset;
} FOUND_SET;

typedef struct st_follow {
  int chr;
  uint table_offset;
  uint len;
} FOLLOWS;


static int init_sets(REP_SETS *sets,uint states);
static REP_SET *make_new_set(REP_SETS *sets);
static void make_sets_invisible(REP_SETS *sets);
static void free_last_set(REP_SETS *sets);
static void free_sets(REP_SETS *sets);
static void internal_set_bit(REP_SET *set, uint bit);
static void internal_clear_bit(REP_SET *set, uint bit);
static void or_bits(REP_SET *to,REP_SET *from);
static void copy_bits(REP_SET *to,REP_SET *from);
static int cmp_bits(REP_SET *set1,REP_SET *set2);
static int get_next_bit(REP_SET *set,uint lastpos);
static short find_set(REP_SETS *sets,REP_SET *find);
static short find_found(FOUND_SET *found_set,uint table_offset,
                        int found_offset);
static uint start_at_word(char * pos);
static uint end_of_word(char * pos);
static uint replace_len(char * pos);

static uint found_sets=0;


	/* Init a replace structure for further calls */

static REPLACE *init_replace(char * *from, char * *to,uint count,
                             char * word_end_chars)
{
  uint i,j,states,set_nr,len,result_len,max_length,found_end,bits_set,bit_nr;
  int used_sets,chr;
  short default_state;
  char used_chars[LAST_CHAR_CODE],is_word_end[256];
  char * pos, *to_pos, **to_array;
  REP_SETS sets;
  REP_SET *set,*start_states,*word_states,*new_set;
  FOLLOWS *follow,*follow_ptr;
  REPLACE *replace;
  FOUND_SET *found_set;
  REPLACE_STRING *rep_str;
  DBUG_ENTER("init_replace");

  /* Count number of states */
  for (i=result_len=max_length=0 , states=2 ; i < count ; i++)
  {
    len=replace_len(from[i]);
    if (!len)
    {
      errno=EINVAL;
      my_message(0,"No to-string for last from-string",MYF(ME_BELL));
      DBUG_RETURN(0);
    }
    states+=len+1;
    result_len+=(uint) strlen(to[i])+1;
    if (len > max_length)
      max_length=len;
  }
  bzero((char*) is_word_end,sizeof(is_word_end));
  for (i=0 ; word_end_chars[i] ; i++)
    is_word_end[(uchar) word_end_chars[i]]=1;

  if (init_sets(&sets,states))
    DBUG_RETURN(0);
  found_sets=0;
  if (!(found_set= (FOUND_SET*) my_malloc(PSI_NOT_INSTRUMENTED,
                                          sizeof(FOUND_SET)*max_length*count,
					  MYF(MY_WME))))
  {
    free_sets(&sets);
    DBUG_RETURN(0);
  }
  (void) make_new_set(&sets);			/* Set starting set */
  make_sets_invisible(&sets);			/* Hide previus sets */
  used_sets=-1;
  word_states=make_new_set(&sets);		/* Start of new word */
  start_states=make_new_set(&sets);		/* This is first state */
  if (!(follow=(FOLLOWS*) my_malloc(PSI_NOT_INSTRUMENTED,
                                    (states+2)*sizeof(FOLLOWS), MYF(MY_WME))))
  {
    free_sets(&sets);
    my_free(found_set);
    DBUG_RETURN(0);
  }

	/* Init follow_ptr[] */
  for (i=0, states=1, follow_ptr=follow+1 ; i < count ; i++)
  {
    if (from[i][0] == '\\' && from[i][1] == '^')
    {
      internal_set_bit(start_states,states+1);
      if (!from[i][2])
      {
	start_states->table_offset=i;
	start_states->found_offset=1;
      }
    }
    else if (from[i][0] == '\\' && from[i][1] == '$')
    {
      internal_set_bit(start_states,states);
      internal_set_bit(word_states,states);
      if (!from[i][2] && start_states->table_offset == (uint) ~0)
      {
	start_states->table_offset=i;
	start_states->found_offset=0;
      }
    }
    else
    {
      internal_set_bit(word_states,states);
      if (from[i][0] == '\\' && (from[i][1] == 'b' && from[i][2]))
	internal_set_bit(start_states,states+1);
      else
	internal_set_bit(start_states,states);
    }
    for (pos=from[i], len=0; *pos ; pos++)
    {
      if (*pos == '\\' && *(pos+1))
      {
	pos++;
	switch (*pos) {
	case 'b':
	  follow_ptr->chr = SPACE_CHAR;
	  break;
	case '^':
	  follow_ptr->chr = START_OF_LINE;
	  break;
	case '$':
	  follow_ptr->chr = END_OF_LINE;
	  break;
	case 'r':
	  follow_ptr->chr = '\r';
	  break;
	case 't':
	  follow_ptr->chr = '\t';
	  break;
	case 'v':
	  follow_ptr->chr = '\v';
	  break;
	default:
	  follow_ptr->chr = (uchar) *pos;
	  break;
	}
      }
      else
	follow_ptr->chr= (uchar) *pos;
      follow_ptr->table_offset=i;
      follow_ptr->len= ++len;
      follow_ptr++;
    }
    follow_ptr->chr=0;
    follow_ptr->table_offset=i;
    follow_ptr->len=len;
    follow_ptr++;
    states+=(uint) len+1;
  }


  for (set_nr=0,pos=0 ; set_nr < sets.count ; set_nr++)
  {
    set=sets.set+set_nr;
    default_state= 0;				/* Start from beginning */

    /* If end of found-string not found or start-set with current set */

    for (i= (uint) ~0; (i=get_next_bit(set,i)) ;)
    {
      if (!follow[i].chr)
      {
	if (! default_state)
	  default_state= find_found(found_set,set->table_offset,
				    set->found_offset+1);
      }
    }
    copy_bits(sets.set+used_sets,set);		/* Save set for changes */
    if (!default_state)
      or_bits(sets.set+used_sets,sets.set);	/* Can restart from start */

    /* Find all chars that follows current sets */
    bzero((char*) used_chars,sizeof(used_chars));
    for (i= (uint) ~0; (i=get_next_bit(sets.set+used_sets,i)) ;)
    {
      used_chars[follow[i].chr]=1;
      if ((follow[i].chr == SPACE_CHAR && !follow[i+1].chr &&
	   follow[i].len > 1) || follow[i].chr == END_OF_LINE)
	used_chars[0]=1;
    }

    /* Mark word_chars used if \b is in state */
    if (used_chars[SPACE_CHAR])
      for (pos= word_end_chars ; *pos ; pos++)
	used_chars[(int) (uchar) *pos] = 1;

    /* Handle other used characters */
    for (chr= 0 ; chr < 256 ; chr++)
    {
      if (! used_chars[chr])
	set->next[chr]= (short) (chr ? default_state : -1);
      else
      {
	new_set=make_new_set(&sets);
	set=sets.set+set_nr;			/* if realloc */
	new_set->table_offset=set->table_offset;
	new_set->found_len=set->found_len;
	new_set->found_offset=set->found_offset+1;
	found_end=0;

	for (i= (uint) ~0 ; (i=get_next_bit(sets.set+used_sets,i)) ; )
	{
	  if (!follow[i].chr || follow[i].chr == chr ||
	      (follow[i].chr == SPACE_CHAR &&
	       (is_word_end[chr] ||
		(!chr && follow[i].len > 1 && ! follow[i+1].chr))) ||
	      (follow[i].chr == END_OF_LINE && ! chr))
	  {
	    if ((! chr || (follow[i].chr && !follow[i+1].chr)) &&
		follow[i].len > found_end)
	      found_end=follow[i].len;
	    if (chr && follow[i].chr)
	      internal_set_bit(new_set,i+1);		/* To next set */
	    else
	      internal_set_bit(new_set,i);
	  }
	}
	if (found_end)
	{
	  new_set->found_len=0;			/* Set for testing if first */
	  bits_set=0;
	  for (i= (uint) ~0; (i=get_next_bit(new_set,i)) ;)
	  {
	    if ((follow[i].chr == SPACE_CHAR ||
		 follow[i].chr == END_OF_LINE) && ! chr)
	      bit_nr=i+1;
	    else
	      bit_nr=i;
	    if (follow[bit_nr-1].len < found_end ||
		(new_set->found_len &&
		 (chr == 0 || !follow[bit_nr].chr)))
	      internal_clear_bit(new_set,i);
	    else
	    {
	      if (chr == 0 || !follow[bit_nr].chr)
	      {					/* best match  */
		new_set->table_offset=follow[bit_nr].table_offset;
		if (chr || (follow[i].chr == SPACE_CHAR ||
			    follow[i].chr == END_OF_LINE))
		  new_set->found_offset=found_end;	/* New match */
		new_set->found_len=found_end;
	      }
	      bits_set++;
	    }
	  }
	  if (bits_set == 1)
	  {
	    set->next[chr] = find_found(found_set,
					new_set->table_offset,
					new_set->found_offset);
	    free_last_set(&sets);
	  }
	  else
	    set->next[chr] = find_set(&sets,new_set);
	}
	else
	  set->next[chr] = find_set(&sets,new_set);
      }
    }
  }

	/* Alloc replace structure for the replace-state-machine */

  if ((replace=(REPLACE*) my_malloc(PSI_NOT_INSTRUMENTED,
                                    sizeof(REPLACE)*(sets.count)+
				    sizeof(REPLACE_STRING)*(found_sets+1)+
				    sizeof(char *)*count+result_len,
				    MYF(MY_WME | MY_ZEROFILL))))
  {
    rep_str=(REPLACE_STRING*) (replace+sets.count);
    to_array=(char **) (rep_str+found_sets+1);
    to_pos=(char *) (to_array+count);
    for (i=0 ; i < count ; i++)
    {
      to_array[i]=to_pos;
      to_pos=strmov(to_pos,to[i])+1;
    }
    rep_str[0].found=1;
    rep_str[0].replace_string=0;
    for (i=1 ; i <= found_sets ; i++)
    {
      pos=from[found_set[i-1].table_offset];
      /*
        Test if we are matching start of string (\^)
        We can't use bcmp() here as pos may be only 1 character and
        that would confuse MSAN.
      */
      rep_str[i].found= (uint8) ((pos[0] == '\\' && pos[1] == '^' &&
                                  pos[2] == 0) ? 2 : 1);
      rep_str[i].replace_string=to_array[found_set[i-1].table_offset];
      rep_str[i].to_offset=found_set[i-1].found_offset-start_at_word(pos);
      rep_str[i].from_offset=found_set[i-1].found_offset-replace_len(pos)+
	end_of_word(pos);
    }
    for (i=0 ; i < sets.count ; i++)
    {
      for (j=0 ; j < 256 ; j++)
	if (sets.set[i].next[j] >= 0)
	  replace[i].next[j]=replace+sets.set[i].next[j];
	else
	  replace[i].next[j]=(REPLACE*) (rep_str+(-sets.set[i].next[j]-1));
    }
  }
  my_free(follow);
  free_sets(&sets);
  my_free(found_set);
  DBUG_PRINT("exit",("Replace table has %d states",sets.count));
  DBUG_RETURN(replace);
}


static int init_sets(REP_SETS *sets,uint states)
{
  bzero((char*) sets,sizeof(*sets));
  sets->size_of_bits=((states+7)/8);
  if (!(sets->set_buffer=(REP_SET*) my_malloc(PSI_NOT_INSTRUMENTED,
                                              sizeof(REP_SET)*SET_MALLOC_HUNC,
					      MYF(MY_WME))))
    return 1;
  if (!(sets->bit_buffer=(uint*) my_malloc(PSI_NOT_INSTRUMENTED,
                                           sizeof(uint)*sets->size_of_bits*
					   SET_MALLOC_HUNC,MYF(MY_WME))))
  {
    my_free(sets->set);
    return 1;
  }
  return 0;
}

	/* Make help sets invisible for nicer codeing */

static void make_sets_invisible(REP_SETS *sets)
{
  sets->invisible=sets->count;
  sets->set+=sets->count;
  sets->count=0;
}

static REP_SET *make_new_set(REP_SETS *sets)
{
  uint i,count,*bit_buffer;
  REP_SET *set;
  if (sets->extra)
  {
    sets->extra--;
    set=sets->set+ sets->count++;
    bzero((char*) set->bits,sizeof(uint)*sets->size_of_bits);
    bzero((char*) &set->next[0],sizeof(set->next[0])*LAST_CHAR_CODE);
    set->found_offset=0;
    set->found_len=0;
    set->table_offset= (uint) ~0;
    set->size_of_bits=sets->size_of_bits;
    return set;
  }
  count=sets->count+sets->invisible+SET_MALLOC_HUNC;
  if (!(set=(REP_SET*) my_realloc(PSI_NOT_INSTRUMENTED, sets->set_buffer,
                                  sizeof(REP_SET)*count, MYF(MY_WME))))
    return 0;
  sets->set_buffer=set;
  sets->set=set+sets->invisible;
  if (!(bit_buffer=(uint*) my_realloc(PSI_NOT_INSTRUMENTED, sets->bit_buffer,
                                      (sizeof(uint)*sets->size_of_bits)*count,
                                      MYF(MY_WME))))
    return 0;
  sets->bit_buffer=bit_buffer;
  for (i=0 ; i < count ; i++)
  {
    sets->set_buffer[i].bits=bit_buffer;
    bit_buffer+=sets->size_of_bits;
  }
  sets->extra=SET_MALLOC_HUNC;
  return make_new_set(sets);
}

static void free_last_set(REP_SETS *sets)
{
  sets->count--;
  sets->extra++;
  return;
}

static void free_sets(REP_SETS *sets)
{
  my_free(sets->set_buffer);
  my_free(sets->bit_buffer);
  return;
}

static void internal_set_bit(REP_SET *set, uint bit)
{
  set->bits[bit / WORD_BIT] |= 1 << (bit % WORD_BIT);
  return;
}

static void internal_clear_bit(REP_SET *set, uint bit)
{
  set->bits[bit / WORD_BIT] &= ~ (1 << (bit % WORD_BIT));
  return;
}


static void or_bits(REP_SET *to,REP_SET *from)
{
  reg1 uint i;
  for (i=0 ; i < to->size_of_bits ; i++)
    to->bits[i]|=from->bits[i];
  return;
}

static void copy_bits(REP_SET *to,REP_SET *from)
{
  memcpy((uchar*) to->bits,(uchar*) from->bits,
	 (size_t) (sizeof(uint) * to->size_of_bits));
}

static int cmp_bits(REP_SET *set1,REP_SET *set2)
{
  return memcmp(set1->bits, set2->bits,
                sizeof(uint) * set1->size_of_bits);
}


	/* Get next set bit from set. */

static int get_next_bit(REP_SET *set,uint lastpos)
{
  uint pos,*start,*end,bits;

  start=set->bits+ ((lastpos+1) / WORD_BIT);
  end=set->bits + set->size_of_bits;
  bits=start[0] & ~((1 << ((lastpos+1) % WORD_BIT)) -1);

  while (! bits && ++start < end)
    bits=start[0];
  if (!bits)
    return 0;
  pos=(uint) (start-set->bits)*WORD_BIT;
  while (! (bits & 1))
  {
    bits>>=1;
    pos++;
  }
  return pos;
}

	/* find if there is a same set in sets. If there is, use it and
	   free given set, else put in given set in sets and return it's
	   position */

static short find_set(REP_SETS *sets,REP_SET *find)
{
  uint i;
  for (i=0 ; i < sets->count-1 ; i++)
  {
    if (!cmp_bits(sets->set+i,find))
    {
      free_last_set(sets);
      return (short) i;
    }
  }
  return (short) i;			/* return new position */
}


/*
  find if there is a found_set with same table_offset & found_offset
  If there is return offset to it, else add new offset and return pos.
  Pos returned is -offset-2 in found_set_structure because it's is
  saved in set->next and set->next[] >= 0 points to next set and
  set->next[] == -1 is reserved for end without replaces.
*/

static short find_found(FOUND_SET *found_set,uint table_offset,
                        int found_offset)
{
  int i;
  for (i=0 ; (uint) i < found_sets ; i++)
    if (found_set[i].table_offset == table_offset &&
	found_set[i].found_offset == found_offset)
      return (short) (-i-2);
  found_set[i].table_offset=table_offset;
  found_set[i].found_offset=found_offset;
  found_sets++;
  return (short) (-i-2);			/* return new position */
}

	/* Return 1 if regexp starts with \b or ends with \b*/

static uint start_at_word(char * pos)
{
  return (((!memcmp(pos,"\\b",2) && pos[2]) || !memcmp(pos,"\\^",2)) ? 1 : 0);
}

static uint end_of_word(char * pos)
{
  char * end=strend(pos);
  return ((end > pos+2 && !memcmp(end-2,"\\b",2)) ||
	  (end >= pos+2 && !memcmp(end-2,"\\$",2))) ?
	    1 : 0;
}


static uint replace_len(char * str)
{
  uint len=0;
  while (*str)
  {
    if (str[0] == '\\' && str[1])
      str++;
    str++;
    len++;
  }
  return len;
}


	/* The actual loop */

static uint replace_strings(REPLACE *rep, char **start, uint *max_length,
                            char *from)
{
  reg1 REPLACE *rep_pos;
  reg2 REPLACE_STRING *rep_str;
  char *to, *end, *pos, *new;

  end=(to= *start) + *max_length-1;
  rep_pos=rep+1;
  for(;;)
  {
    while (!rep_pos->found)
    {
      rep_pos= rep_pos->next[(uchar) *from];
      if (to == end)
      {
	(*max_length)+=8192;
	if (!(new=my_realloc(PSI_NOT_INSTRUMENTED, *start,*max_length,MYF(MY_WME))))
	  return (uint) -1;
	to=new+(to - *start);
	end=(*start=new)+ *max_length-1;
      }
      *to++= *from++;
    }
    if (!(rep_str = ((REPLACE_STRING*) rep_pos))->replace_string)
      return (uint) (to - *start)-1;
    updated=1;			/* Some char * is replaced */
    to-=rep_str->to_offset;
    for (pos=rep_str->replace_string; *pos ; pos++)
    {
      if (to == end)
      {
	(*max_length)*=2;
	if (!(new=my_realloc(PSI_NOT_INSTRUMENTED, *start,*max_length,MYF(MY_WME))))
	  return (uint) -1;
	to=new+(to - *start);
	end=(*start=new)+ *max_length-1;
      }
      *to++= *pos;
    }
    if (!*(from-=rep_str->from_offset) && rep_pos->found != 2)
      return (uint) (to - *start);
    rep_pos=rep;
  }
}

static char *buffer;		/* The buffer itself, grown as needed. */
static int bufbytes;		/* Number of bytes in the buffer. */
static int bufread,my_eof;		/* Number of bytes to get with each read(). */
static uint bufalloc;
static char *out_buff;
static uint out_length;

static int initialize_buffer()
{
  bufread = 8192;
  bufalloc = bufread + bufread / 2;
  if (!(buffer = my_malloc(PSI_NOT_INSTRUMENTED, bufalloc+1, MYF(MY_WME))))
    return 1;
  bufbytes=my_eof=0;
  out_length=bufread;
  if (!(out_buff=my_malloc(PSI_NOT_INSTRUMENTED, out_length, MYF(MY_WME))))
    return(1);
  return 0;
}

static void reset_buffer()
{
  bufbytes=my_eof=0;
}

static void free_buffer()
{
  my_free(buffer);
  my_free(out_buff);
}


/*
  Fill the buffer retaining the last n bytes at the beginning of the
  newly filled buffer (for backward context).  Returns the number of new
  bytes read from disk.
*/

static int fill_buffer_retaining(File fd, int n)
{
  int i;

  /* See if we need to grow the buffer. */
  if ((int) bufalloc - n <= bufread)
  {
    while ((int) bufalloc - n <= bufread)
    {
      bufalloc *= 2;
      bufread *= 2;
    }
    buffer = my_realloc(PSI_NOT_INSTRUMENTED, buffer, bufalloc+1, MYF(MY_WME));
    if (! buffer)
      return(-1);
  }

  /* Shift stuff down. */
  bmove(buffer,buffer+bufbytes-n,(uint) n);
  bufbytes = n;

  if (my_eof)
    return 0;

  /* Read in new stuff. */
  if ((i=(int) my_read(fd, (uchar*) buffer + bufbytes,
                       (size_t) bufread, MYF(MY_WME))) < 0)
    return -1;

  /* Kludge to pretend every nonempty file ends with a newline. */
  if (i == 0 && bufbytes > 0 && buffer[bufbytes - 1] != '\n')
  {
    my_eof = i = 1;
    buffer[bufbytes] = '\n';
  }

  bufbytes += i;
  return i;
}

	/* Return 0 if convert is ok */
	/* Global variable update is set if something was changed */

static int convert_pipe(REPLACE *rep, FILE *in, FILE *out)
{
  int retain,error;
  uint length;
  char save_char,*end_of_line,*start_of_line;
  DBUG_ENTER("convert_pipe");

  updated=retain=0;
  reset_buffer();

  while ((error=fill_buffer_retaining(my_fileno(in),retain)) > 0)
  {
    end_of_line=buffer ;
    buffer[bufbytes]=0;			/* Sentinel  */
    for (;;)
    {
      start_of_line=end_of_line;
      while (end_of_line[0] != '\n' && end_of_line[0])
	end_of_line++;
      if (end_of_line == buffer+bufbytes)
      {
	retain= (int) (end_of_line - start_of_line);
	break;				/* No end of line, read more */
      }
      save_char=end_of_line[0];
      end_of_line[0]=0;
      end_of_line++;
      if ((length=replace_strings(rep,&out_buff,&out_length,start_of_line)) ==
	  (uint) -1)
	return 1;
      if (!my_eof)
	out_buff[length++]=save_char;	/* Don't write added newline */
      if (my_fwrite(out, (uchar*) out_buff, length, MYF(MY_WME | MY_NABP)))
	DBUG_RETURN(1);
    }
  }
  DBUG_RETURN(error);
}


static int convert_file(REPLACE *rep, char * name)
{
  int error;
  FILE *in,*out;
  char dir_buff[FN_REFLEN], tempname[FN_REFLEN], *org_name = name;
#ifdef HAVE_READLINK
  char link_name[FN_REFLEN];
#endif
  File temp_file;
  size_t dir_buff_length;
  DBUG_ENTER("convert_file");

  /* check if name is a symlink */
#ifdef HAVE_READLINK  
  org_name= (!my_disable_symlinks && 
             !my_readlink(link_name, name, MYF(0))) ? link_name : name;
#endif
  if (!(in= my_fopen(org_name,O_RDONLY,MYF(MY_WME))))
    DBUG_RETURN(1);
  dirname_part(dir_buff, org_name, &dir_buff_length);
  if ((temp_file= create_temp_file(tempname, dir_buff, "PR", 0,
                                   MYF(MY_WME))) < 0)
  {
    my_fclose(in,MYF(0));
    DBUG_RETURN(1);
  }    
  if (!(out= my_fdopen(temp_file, tempname, O_WRONLY, MYF(MY_WME))))
  {
    my_fclose(in,MYF(0));
    DBUG_RETURN(1);
  }

  error=convert_pipe(rep,in,out);
  my_fclose(in,MYF(0)); my_fclose(out,MYF(0));

  if (updated && ! error)
    my_redel(org_name, tempname, 0, MYF(MY_WME | MY_LINK_WARNING));
  else
    my_delete(tempname,MYF(MY_WME));
  if (!silent && ! error)
  {
    if (updated)
      printf("%s converted\n",name);
    else if (verbose)
      printf("%s left unchanged\n",name);
  }
  DBUG_RETURN(error);
}