summaryrefslogtreecommitdiff
path: root/doc/pcretest.txt
blob: 55de5024431dc4a787ba4c0c4a152a4bf9dfc32c (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
PCRETEST(1)                 General Commands Manual                PCRETEST(1)



NAME
       pcretest - a program for testing Perl-compatible regular expressions.

SYNOPSIS

       pcretest [options] [input file [output file]]

       pcretest  was written as a test program for the PCRE regular expression
       library itself, but it can also be used for experimenting with  regular
       expressions.  This document describes the features of the test program;
       for details of the regular expressions themselves, see the  pcrepattern
       documentation. For details of the PCRE library function calls and their
       options, see the pcreapi , pcre16 and pcre32 documentation.

       The input for pcretest is a sequence of regular expression patterns and
       strings  to be matched, as described below. The output shows the result
       of each match. Options on the command line  and  the  patterns  control
       PCRE options and exactly what is output.

       As  PCRE has evolved, it has acquired many different features, and as a
       result, pcretest now has rather a lot of obscure  options  for  testing
       every possible feature. Some of these options are specifically designed
       for use in conjunction with the test script and  data  files  that  are
       distributed  as  part of PCRE, and are unlikely to be of use otherwise.
       They are all documented here, but without much justification.


INPUT DATA FORMAT

       Input to pcretest is processed line by line, either by  calling  the  C
       library's fgets() function, or via the libreadline library (see below).
       In Unix-like environments, fgets() treats any bytes other than  newline
       as  data characters. However, in some Windows environments character 26
       (hex 1A) causes an immediate end of file, and no further data is  read.
       For  maximum  portability,  therefore,  it  is safest to use only ASCII
       characters in pcretest input files.


PCRE's 8-BIT, 16-BIT AND 32-BIT LIBRARIES

       From release 8.30, two separate PCRE libraries can be built. The origi-
       nal  one  supports  8-bit  character  strings, whereas the newer 16-bit
       library supports  character  strings  encoded  in  16-bit  units.  From
       release  8.32,  a  third  library  can  be  built, supporting character
       strings encoded in 32-bit units. The pcretest program can  be  used  to
       test all three libraries. However, it is itself still an 8-bit program,
       reading 8-bit input and writing 8-bit output.  When testing the  16-bit
       or  32-bit  library, the patterns and data strings are converted to 16-
       or 32-bit format before being passed to  the  PCRE  library  functions.
       Results are converted to 8-bit for output.

       References to functions and structures of the form pcre[16|32]_xx below
       mean "pcre_xx when using the 8-bit library, pcre16_xx  when  using  the
       16-bit library, or pcre32_xx when using the 32-bit library".


COMMAND LINE OPTIONS

       -8        If  both the 8-bit library has been built, this option causes
                 the 8-bit library to be used (which is the default);  if  the
                 8-bit  library  has  not  been  built,  this option causes an
                 error.

       -16       If both the 8-bit or the 32-bit,  and  the  16-bit  libraries
                 have  been built, this option causes the 16-bit library to be
                 used. If only the 16-bit library has been built, this is  the
                 default  (so  has no effect). If only the 8-bit or the 32-bit
                 library has been built, this option causes an error.

       -32       If both the 8-bit or the 16-bit,  and  the  32-bit  libraries
                 have  been built, this option causes the 32-bit library to be
                 used. If only the 32-bit library has been built, this is  the
                 default  (so  has no effect). If only the 8-bit or the 16-bit
                 library has been built, this option causes an error.

       -b        Behave as if each pattern has the /B (show byte  code)  modi-
                 fier; the internal form is output after compilation.

       -C        Output the version number of the PCRE library, and all avail-
                 able  information  about  the  optional  features  that   are
                 included,  and  then  exit  with  zero  exit  code. All other
                 options are ignored.

       -C option Output information about a specific build-time  option,  then
                 exit.  This functionality is intended for use in scripts such
                 as RunTest. The following options output the  value  and  set
                 the exit code as indicated:

                   ebcdic-nl  the code for LF (= NL) in an EBCDIC environment:
                                0x15 or 0x25
                                0 if used in an ASCII environment
                                exit code is always 0
                   linksize   the configured internal link size (2, 3, or 4)
                                exit code is set to the link size
                   newline    the default newline setting:
                                CR, LF, CRLF, ANYCRLF, or ANY
                                exit code is always 0
                   bsr        the default setting for what \R matches:
                                ANYCRLF or ANY
                                exit code is always 0

                 The  following  options output 1 for true or 0 for false, and
                 set the exit code to the same value:

                   ebcdic     compiled for an EBCDIC environment
                   jit        just-in-time support is available
                   pcre16     the 16-bit library was built
                   pcre32     the 32-bit library was built
                   pcre8      the 8-bit library was built
                   ucp        Unicode property support is available
                   utf        UTF-8 and/or UTF-16 and/or UTF-32 support
                                is available

                 If an unknown option is given, an error  message  is  output;
                 the exit code is 0.

       -d        Behave  as  if  each pattern has the /D (debug) modifier; the
                 internal form and information about the compiled  pattern  is
                 output after compilation; -d is equivalent to -b -i.

       -dfa      Behave  as if each data line contains the \D escape sequence;
                 this    causes    the    alternative    matching    function,
                 pcre[16|32]_dfa_exec(),  to  be  used instead of the standard
                 pcre[16|32]_exec() function (more detail is given below).

       -help     Output a brief summary these options and then exit.

       -i        Behave as if each pattern has the  /I  modifier;  information
                 about the compiled pattern is given after compilation.

       -M        Behave  as if each data line contains the \M escape sequence;
                 this causes PCRE to  discover  the  minimum  MATCH_LIMIT  and
                 MATCH_LIMIT_RECURSION  settings by calling pcre[16|32]_exec()
                 repeatedly with different limits.

       -m        Output the size of each compiled pattern after  it  has  been
                 compiled.  This  is  equivalent  to adding /M to each regular
                 expression. The size is given in bytes for both libraries.

       -O        Behave as if each pattern has the /O modifier, that  is  dis-
                 able auto-possessification for all patterns.

       -o osize  Set  the number of elements in the output vector that is used
                 when calling pcre[16|32]_exec() or pcre[16|32]_dfa_exec()  to
                 be  osize.  The  default  value is 45, which is enough for 14
                 capturing subexpressions for pcre[16|32]_exec() or 22 differ-
                 ent  matches for pcre[16|32]_dfa_exec().  The vector size can
                 be changed for individual matching calls by including  \O  in
                 the data line (see below).

       -p        Behave  as  if  each  pattern  has the /P modifier; the POSIX
                 wrapper API is used to call PCRE. None of the  other  options
                 has  any  effect when -p is set. This option can be used only
                 with the 8-bit library.

       -q        Do not output the version number of pcretest at the start  of
                 execution.

       -S size   On  Unix-like  systems, set the size of the run-time stack to
                 size megabytes.

       -s or -s+ Behave as if each pattern  has  the  /S  modifier;  in  other
                 words,  force each pattern to be studied. If -s+ is used, all
                 the JIT compile options are  passed  to  pcre[16|32]_study(),
                 causing  just-in-time  optimization  to  be  set  up if it is
                 available, for both full and partial matching.  Specific  JIT
                 compile options can be selected by following -s+ with a digit
                 in the range 1 to 7, which selects the JIT compile  modes  as
                 follows:

                   1  normal match only
                   2  soft partial match only
                   3  normal match and soft partial match
                   4  hard partial match only
                   6  soft and hard partial match
                   7  all three modes (default)

                 If  -s++  is used instead of -s+ (with or without a following
                 digit), the text "(JIT)" is added to the  first  output  line
                 after a match or no match when JIT-compiled code was actually
                 used.

                 Note that there are pattern options  that  can  override  -s,
                 either specifying no studying at all, or suppressing JIT com-
                 pilation.

                 If the /I or /D option is present on  a  pattern  (requesting
                 output  about  the  compiled  pattern), information about the
                 result of studying is not included when  studying  is  caused
                 only  by  -s  and neither -i nor -d is present on the command
                 line. This behaviour means that the output  from  tests  that
                 are  run with and without -s should be identical, except when
                 options that output information about the actual running of a
                 match are set.

                 The  -M,  -t,  and  -tm options, which give information about
                 resources used, are likely to produce different  output  with
                 and  without  -s.  Output may also differ if the /C option is
                 present on an individual pattern. This uses callouts to trace
                 the  the  matching process, and this may be different between
                 studied and non-studied patterns.  If  the  pattern  contains
                 (*MARK)  items  there  may  also be differences, for the same
                 reason. The -s command line option can be overridden for spe-
                 cific  patterns that should never be studied (see the /S pat-
                 tern modifier below).

       -t        Run each compile, study, and match many times with  a  timer,
                 and  output  the resulting times per compile, study, or match
                 (in milliseconds). Do not set -m with -t,  because  you  will
                 then get the size output a zillion times, and the timing will
                 be distorted. You can control the number of  iterations  that
                 are used for timing by following -t with a number (as a sepa-
                 rate item on the command line). For example, "-t 1000"  iter-
                 ates 1000 times.  The default is to iterate 500000 times.

       -tm       This is like -t except that it times only the matching phase,
                 not the compile or study phases.

       -T -TM    These behave like -t and -tm, but in addition, at the end  of
                 a run, the total times for all compiles, studies, and matches
                 are output.


DESCRIPTION

       If pcretest is given two filename arguments, it reads  from  the  first
       and writes to the second. If it is given only one filename argument, it
       reads from that file and writes to stdout.  Otherwise,  it  reads  from
       stdin  and  writes to stdout, and prompts for each line of input, using
       "re>" to prompt for regular expressions, and "data>" to prompt for data
       lines.

       When  pcretest  is  built,  a  configuration option can specify that it
       should be linked with the libreadline library. When this  is  done,  if
       the input is from a terminal, it is read using the readline() function.
       This provides line-editing and history facilities. The output from  the
       -help option states whether or not readline() will be used.

       The program handles any number of sets of input on a single input file.
       Each set starts with a regular expression, and continues with any  num-
       ber of data lines to be matched against that pattern.

       Each  data line is matched separately and independently. If you want to
       do multi-line matches, you have to use the \n escape sequence (or \r or
       \r\n, etc., depending on the newline setting) in a single line of input
       to encode the newline sequences. There is no limit  on  the  length  of
       data  lines;  the  input  buffer is automatically extended if it is too
       small.

       An empty line signals the end of the data lines, at which point  a  new
       regular  expression is read. The regular expressions are given enclosed
       in any non-alphanumeric delimiters other than backslash, for example:

         /(a|bc)x+yz/

       White space before the initial delimiter is ignored. A regular  expres-
       sion  may be continued over several input lines, in which case the new-
       line characters are included within it. It is possible to  include  the
       delimiter within the pattern by escaping it, for example

         /abc\/def/

       If  you  do  so, the escape and the delimiter form part of the pattern,
       but since delimiters are always non-alphanumeric, this does not  affect
       its  interpretation.   If the terminating delimiter is immediately fol-
       lowed by a backslash, for example,

         /abc/\

       then a backslash is added to the end of the pattern. This  is  done  to
       provide  a  way of testing the error condition that arises if a pattern
       finishes with a backslash, because

         /abc\/

       is interpreted as the first line of a pattern that starts with  "abc/",
       causing pcretest to read the next line as a continuation of the regular
       expression.


PATTERN MODIFIERS

       A pattern may be followed by any number of modifiers, which are  mostly
       single  characters,  though  some  of these can be qualified by further
       characters.  Following Perl usage, these are referred to below as,  for
       example,  "the  /i  modifier", even though the delimiter of the pattern
       need not always be a slash, and no slash is  used  when  writing  modi-
       fiers.  White  space may appear between the final pattern delimiter and
       the first modifier, and between the modifiers  themselves.  For  refer-
       ence,  here  is  a  complete  list of modifiers. They fall into several
       groups that are described in detail in the following sections.

         /8              set UTF mode
         /9              set PCRE_NEVER_UTF (locks out UTF mode)
         /?              disable UTF validity check
         /+              show remainder of subject after match
         /=              show all captures (not just those that are set)

         /A              set PCRE_ANCHORED
         /B              show compiled code
         /C              set PCRE_AUTO_CALLOUT
         /D              same as /B plus /I
         /E              set PCRE_DOLLAR_ENDONLY
         /F              flip byte order in compiled pattern
         /f              set PCRE_FIRSTLINE
         /G              find all matches (shorten string)
         /g              find all matches (use startoffset)
         /I              show information about pattern
         /i              set PCRE_CASELESS
         /J              set PCRE_DUPNAMES
         /K              show backtracking control names
         /L              set locale
         /M              show compiled memory size
         /m              set PCRE_MULTILINE
         /N              set PCRE_NO_AUTO_CAPTURE
         /O              set PCRE_NO_AUTO_POSSESS
         /P              use the POSIX wrapper
         /Q              test external stack check function
         /S              study the pattern after compilation
         /s              set PCRE_DOTALL
         /T              select character tables
         /U              set PCRE_UNGREEDY
         /W              set PCRE_UCP
         /X              set PCRE_EXTRA
         /x              set PCRE_EXTENDED
         /Y              set PCRE_NO_START_OPTIMIZE
         /Z              don't show lengths in /B output

         /<any>          set PCRE_NEWLINE_ANY
         /<anycrlf>      set PCRE_NEWLINE_ANYCRLF
         /<cr>           set PCRE_NEWLINE_CR
         /<crlf>         set PCRE_NEWLINE_CRLF
         /<lf>           set PCRE_NEWLINE_LF
         /<bsr_anycrlf>  set PCRE_BSR_ANYCRLF
         /<bsr_unicode>  set PCRE_BSR_UNICODE
         /<JS>           set PCRE_JAVASCRIPT_COMPAT


   Perl-compatible modifiers

       The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
       PCRE_DOTALL,    or    PCRE_EXTENDED    options,    respectively,   when
       pcre[16|32]_compile() is called. These four modifier letters  have  the
       same effect as they do in Perl. For example:

         /caseless/i


   Modifiers for other PCRE options

       The  following  table  shows additional modifiers for setting PCRE com-
       pile-time options that do not correspond to anything in Perl:

         /8              PCRE_UTF8           ) when using the 8-bit
         /?              PCRE_NO_UTF8_CHECK  )   library

         /8              PCRE_UTF16          ) when using the 16-bit
         /?              PCRE_NO_UTF16_CHECK )   library

         /8              PCRE_UTF32          ) when using the 32-bit
         /?              PCRE_NO_UTF32_CHECK )   library

         /9              PCRE_NEVER_UTF
         /A              PCRE_ANCHORED
         /C              PCRE_AUTO_CALLOUT
         /E              PCRE_DOLLAR_ENDONLY
         /f              PCRE_FIRSTLINE
         /J              PCRE_DUPNAMES
         /N              PCRE_NO_AUTO_CAPTURE
         /O              PCRE_NO_AUTO_POSSESS
         /U              PCRE_UNGREEDY
         /W              PCRE_UCP
         /X              PCRE_EXTRA
         /Y              PCRE_NO_START_OPTIMIZE
         /<any>          PCRE_NEWLINE_ANY
         /<anycrlf>      PCRE_NEWLINE_ANYCRLF
         /<cr>           PCRE_NEWLINE_CR
         /<crlf>         PCRE_NEWLINE_CRLF
         /<lf>           PCRE_NEWLINE_LF
         /<bsr_anycrlf>  PCRE_BSR_ANYCRLF
         /<bsr_unicode>  PCRE_BSR_UNICODE
         /<JS>           PCRE_JAVASCRIPT_COMPAT

       The modifiers that are enclosed in angle brackets are  literal  strings
       as  shown,  including the angle brackets, but the letters within can be
       in either case.  This example sets multiline matching with CRLF as  the
       line ending sequence:

         /^abc/m<CRLF>

       As  well  as  turning  on  the  PCRE_UTF8/16/32 option, the /8 modifier
       causes all non-printing characters in  output  strings  to  be  printed
       using the \x{hh...} notation. Otherwise, those less than 0x100 are out-
       put in hex without the curly brackets.

       Full details of the PCRE options are given in  the  pcreapi  documenta-
       tion.

   Finding all matches in a string

       Searching  for  all  possible matches within each subject string can be
       requested by the /g or /G modifier. After  finding  a  match,  PCRE  is
       called again to search the remainder of the subject string. The differ-
       ence between /g and /G is that the former uses the startoffset argument
       to  pcre[16|32]_exec()  to  start  searching  at a new point within the
       entire string (which is in effect what Perl does), whereas  the  latter
       passes  over  a  shortened  substring.  This  makes a difference to the
       matching process if the pattern  begins  with  a  lookbehind  assertion
       (including \b or \B).

       If  any  call  to  pcre[16|32]_exec() in a /g or /G sequence matches an
       empty string, the next call is done with the PCRE_NOTEMPTY_ATSTART  and
       PCRE_ANCHORED  flags  set  in  order  to search for another, non-empty,
       match at the same point. If this second match fails, the  start  offset
       is  advanced,  and  the  normal match is retried. This imitates the way
       Perl handles such cases when using the /g modifier or the split() func-
       tion.  Normally,  the start offset is advanced by one character, but if
       the newline convention recognizes CRLF as a newline,  and  the  current
       character is CR followed by LF, an advance of two is used.

   Other modifiers

       There are yet more modifiers for controlling the way pcretest operates.

       The  /+ modifier requests that as well as outputting the substring that
       matched the entire pattern, pcretest  should  in  addition  output  the
       remainder  of  the  subject  string. This is useful for tests where the
       subject contains multiple copies of the same substring. If the +  modi-
       fier  appears  twice, the same action is taken for captured substrings.
       In each case the remainder is output on the following line with a  plus
       character  following  the  capture number. Note that this modifier must
       not immediately follow the /S modifier because /S+ and /S++ have  other
       meanings.

       The  /=  modifier  requests  that  the values of all potential captured
       parentheses be output after a match. By default, only those up  to  the
       highest one actually used in the match are output (corresponding to the
       return code from pcre[16|32]_exec()). Values in the offsets vector cor-
       responding  to higher numbers should be set to -1, and these are output
       as "<unset>". This modifier gives a way of checking that this  is  hap-
       pening.

       The  /B modifier is a debugging feature. It requests that pcretest out-
       put a representation of the compiled code after  compilation.  Normally
       this  information  contains length and offset values; however, if /Z is
       also present, this data is replaced by spaces. This is a  special  fea-
       ture  for  use  in the automatic test scripts; it ensures that the same
       output is generated for different internal link sizes.

       The /D modifier is a PCRE debugging feature, and is equivalent to  /BI,
       that is, both the /B and the /I modifiers.

       The  /F  modifier  causes pcretest to flip the byte order of the 2-byte
       and 4-byte fields in the compiled pattern. This facility is for testing
       the  feature  in PCRE that allows it to execute patterns that were com-
       piled on a host with a different endianness. This feature is not avail-
       able  when the POSIX interface to PCRE is being used, that is, when the
       /P pattern modifier is specified. See also the section about saving and
       reloading compiled patterns below.

       The  /I  modifier  requests  that pcretest output information about the
       compiled pattern (whether it is anchored, has a fixed first  character,
       and  so  on). It does this by calling pcre[16|32]_fullinfo() after com-
       piling a pattern. If the pattern is studied, the results  of  that  are
       also output. In this output, the word "char" means a non-UTF character,
       that is, the value of a single data item  (8-bit,  16-bit,  or  32-bit,
       depending on the library that is being tested).

       The  /K modifier requests pcretest to show names from backtracking con-
       trol verbs that are  returned  from  calls  to  pcre[16|32]_exec().  It
       causes  pcretest  to  create  a  pcre[16|32]_extra block if one has not
       already been created by a call to pcre[16|32]_study(), and to  set  the
       PCRE_EXTRA_MARK  flag  and  the  mark  field within it, every time that
       pcre[16|32]_exec() is called. If  the  variable  that  the  mark  field
       points  to  is  non-NULL  for  a  match,  non-match,  or partial match,
       pcretest prints the string to which it points. For  a  match,  this  is
       shown  on  a  line  by itself, tagged with "MK:". For a non-match it is
       added to the message.

       The /L modifier must be followed directly by the name of a locale,  for
       example,

         /pattern/Lfr_FR

       For this reason, it must be the last modifier. The given locale is set,
       pcre[16|32]_maketables() is called to build a set of  character  tables
       for  the  locale, and this is then passed to pcre[16|32]_compile() when
       compiling the regular expression. Without an /L (or /T) modifier,  NULL
       is  passed  as  the  tables  pointer;  that  is, /L applies only to the
       expression on which it appears.

       The /M modifier causes the size in bytes of the memory  block  used  to
       hold  the compiled pattern to be output. This does not include the size
       of the pcre[16|32] block; it is just the actual compiled data.  If  the
       pattern is successfully studied with the PCRE_STUDY_JIT_COMPILE option,
       the size of the JIT compiled code is also output.

       The /Q modifier is used to test the use of pcre_stack_guard. It must be
       followed  by '0' or '1', specifying the return code to be given from an
       external function that is passed to PCRE and used  for  stack  checking
       during compilation (see the pcreapi documentation for details).

       The  /S  modifier  causes  pcre[16|32]_study()  to  be called after the
       expression has been compiled, and the results used when the  expression
       is matched. There are a number of qualifying characters that may follow
       /S.  They may appear in any order.

       If /S is followed by an exclamation mark, pcre[16|32]_study() is called
       with  the PCRE_STUDY_EXTRA_NEEDED option, causing it always to return a
       pcre_extra block, even when studying discovers no useful information.

       If /S is followed by a second S character, it suppresses studying, even
       if  it  was  requested  externally  by the -s command line option. This
       makes it possible to specify that certain patterns are always  studied,
       and others are never studied, independently of -s. This feature is used
       in the test files in a few cases where the output is different when the
       pattern is studied.

       If  the  /S  modifier  is  followed  by  a  +  character,  the  call to
       pcre[16|32]_study() is made with all the JIT study options,  requesting
       just-in-time  optimization  support if it is available, for both normal
       and partial matching. If you want to restrict the JIT compiling  modes,
       you can follow /S+ with a digit in the range 1 to 7:

         1  normal match only
         2  soft partial match only
         3  normal match and soft partial match
         4  hard partial match only
         6  soft and hard partial match
         7  all three modes (default)

       If /S++ is used instead of /S+ (with or without a following digit), the
       text "(JIT)" is added to the first output line  after  a  match  or  no
       match when JIT-compiled code was actually used.

       Note  that  there  is  also  an independent /+ modifier; it must not be
       given immediately after /S or /S+ because this will be misinterpreted.

       If JIT studying is successful, the compiled JIT code will automatically
       be  used  when pcre[16|32]_exec() is run, except when incompatible run-
       time options are specified. For more details, see the pcrejit  documen-
       tation.  See also the \J escape sequence below for a way of setting the
       size of the JIT stack.

       Finally, if /S is followed by a minus  character,  JIT  compilation  is
       suppressed,  even if it was requested externally by the -s command line
       option. This makes it possible to specify that JIT is never to be  used
       for certain patterns.

       The  /T  modifier  must be followed by a single digit. It causes a spe-
       cific set of built-in character tables to be passed to pcre[16|32]_com-
       pile().  It  is used in the standard PCRE tests to check behaviour with
       different character tables. The digit specifies the tables as follows:

         0   the default ASCII tables, as distributed in
               pcre_chartables.c.dist
         1   a set of tables defining ISO 8859 characters

       In table 1, some characters whose codes are greater than 128 are  iden-
       tified as letters, digits, spaces, etc.

   Using the POSIX wrapper API

       The  /P modifier causes pcretest to call PCRE via the POSIX wrapper API
       rather than its native API. This supports only the 8-bit library.  When
       /P  is set, the following modifiers set options for the regcomp() func-
       tion:

         /i    REG_ICASE
         /m    REG_NEWLINE
         /N    REG_NOSUB
         /s    REG_DOTALL     )
         /U    REG_UNGREEDY   ) These options are not part of
         /W    REG_UCP        )   the POSIX standard
         /8    REG_UTF8       )

       The /+ modifier works as  described  above.  All  other  modifiers  are
       ignored.

   Locking out certain modifiers

       PCRE  can be compiled with or without support for certain features such
       as UTF-8/16/32 or Unicode properties. Accordingly, the  standard  tests
       are  split  up  into  a number of different files that are selected for
       running depending on which features are available.  When  updating  the
       tests, it is all too easy to put a new test into the wrong file by mis-
       take; for example, to put a test that requires UTF support into a  file
       that  is used when it is not available. To help detect such mistakes as
       early as possible, there is a facility for locking out  specific  modi-
       fiers. If an input line for pcretest starts with the string "< forbid "
       the following sequence of characters is taken as a  list  of  forbidden
       modifiers. For example, in the test files that must not use UTF or Uni-
       code property support, this line appears:

         < forbid 8W

       This locks out the /8 and /W modifiers. An immediate error is given  if
       they  are  subsequently encountered. If the character string contains <
       but not >, all the multi-character modifiers  that  begin  with  <  are
       locked  out.  Otherwise,  such modifiers must be explicitly listed, for
       example:

         < forbid <JS><cr>

       There must be a single space between < and "forbid" for this feature to
       be  recognised.  If  there  is not, the line is interpreted either as a
       request to re-load a pre-compiled pattern (see  "SAVING  AND  RELOADING
       COMPILED  PATTERNS"  below) or, if there is a another < character, as a
       pattern that uses < as its delimiter.


DATA LINES

       Before each data line is  passed  to  pcre[16|32]_exec(),  leading  and
       trailing  white space is removed, and it is then scanned for \ escapes.
       Some of these are pretty esoteric features, intended for  checking  out
       some  of the more complicated features of PCRE. If you are just testing
       "ordinary" regular expressions, you probably don't need any  of  these.
       The following escapes are recognized:

         \a         alarm (BEL, \x07)
         \b         backspace (\x08)
         \e         escape (\x27)
         \f         form feed (\x0c)
         \n         newline (\x0a)
         \qdd       set the PCRE_MATCH_LIMIT limit to dd
                      (any number of digits)
         \r         carriage return (\x0d)
         \t         tab (\x09)
         \v         vertical tab (\x0b)
         \nnn       octal character (up to 3 octal digits); always
                      a byte unless > 255 in UTF-8 or 16-bit or 32-bit mode
         \o{dd...}  octal character (any number of octal digits}
         \xhh       hexadecimal byte (up to 2 hex digits)
         \x{hh...}  hexadecimal character (any number of hex digits)
         \A         pass the PCRE_ANCHORED option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \B         pass the PCRE_NOTBOL option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \Cdd       call pcre[16|32]_copy_substring() for substring dd
                      after a successful match (number less than 32)
         \Cname     call pcre[16|32]_copy_named_substring() for substring
                      "name" after a successful match (name termin-
                      ated by next non alphanumeric character)
         \C+        show the current captured substrings at callout
                      time
         \C-        do not supply a callout function
         \C!n       return 1 instead of 0 when callout number n is
                      reached
         \C!n!m     return 1 instead of 0 when callout number n is
                      reached for the nth time
         \C*n       pass the number n (may be negative) as callout
                      data; this is used as the callout return value
         \D         use the pcre[16|32]_dfa_exec() match function
         \F         only shortest match for pcre[16|32]_dfa_exec()
         \Gdd       call pcre[16|32]_get_substring() for substring dd
                      after a successful match (number less than 32)
         \Gname     call pcre[16|32]_get_named_substring() for substring
                      "name" after a successful match (name termin-
                      ated by next non-alphanumeric character)
         \Jdd       set up a JIT stack of dd kilobytes maximum (any
                      number of digits)
         \L         call pcre[16|32]_get_substringlist() after a
                      successful match
         \M         discover the minimum MATCH_LIMIT and
                      MATCH_LIMIT_RECURSION settings
         \N         pass the PCRE_NOTEMPTY option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec(); if used twice, pass the
                      PCRE_NOTEMPTY_ATSTART option
         \Odd       set the size of the output vector passed to
                      pcre[16|32]_exec() to dd (any number of digits)
         \P         pass the PCRE_PARTIAL_SOFT option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec(); if used twice, pass the
                      PCRE_PARTIAL_HARD option
         \Qdd       set the PCRE_MATCH_LIMIT_RECURSION limit to dd
                      (any number of digits)
         \R         pass the PCRE_DFA_RESTART option to pcre[16|32]_dfa_exec()
         \S         output details of memory get/free calls during matching
         \Y             pass     the    PCRE_NO_START_OPTIMIZE    option    to
       pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \Z         pass the PCRE_NOTEOL option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \?         pass the PCRE_NO_UTF[8|16|32]_CHECK option to
                      pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
         \>dd       start the match at offset dd (optional "-"; then
                      any number of digits); this sets the startoffset
                      argument        for        pcre[16|32]_exec()         or
       pcre[16|32]_dfa_exec()
         \<cr>      pass the PCRE_NEWLINE_CR option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \<lf>      pass the PCRE_NEWLINE_LF option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \<crlf>    pass the PCRE_NEWLINE_CRLF option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \<anycrlf> pass the PCRE_NEWLINE_ANYCRLF option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()
         \<any>     pass the PCRE_NEWLINE_ANY option to pcre[16|32]_exec()
                      or pcre[16|32]_dfa_exec()

       The  use of \x{hh...} is not dependent on the use of the /8 modifier on
       the pattern. It is recognized always. There may be any number of  hexa-
       decimal  digits  inside  the  braces; invalid values provoke error mes-
       sages.

       Note that \xhh specifies one byte rather than one  character  in  UTF-8
       mode;  this  makes it possible to construct invalid UTF-8 sequences for
       testing purposes. On the other hand, \x{hh} is interpreted as  a  UTF-8
       character  in UTF-8 mode, generating more than one byte if the value is
       greater than 127.  When testing the 8-bit library not  in  UTF-8  mode,
       \x{hh} generates one byte for values less than 256, and causes an error
       for greater values.

       In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it
       possible to construct invalid UTF-16 sequences for testing purposes.

       In  UTF-32  mode,  all  4- to 8-digit \x{...} values are accepted. This
       makes it possible to construct invalid  UTF-32  sequences  for  testing
       purposes.

       The  escapes  that  specify  line ending sequences are literal strings,
       exactly as shown. No more than one newline setting should be present in
       any data line.

       A  backslash  followed by anything else just escapes the anything else.
       If the very last character is a backslash, it is ignored. This gives  a
       way  of  passing  an empty line as data, since a real empty line termi-
       nates the data input.

       The \J escape provides a way of setting the maximum stack size that  is
       used  by the just-in-time optimization code. It is ignored if JIT opti-
       mization is not being used. Providing a stack that is larger  than  the
       default 32K is necessary only for very complicated patterns.

       If \M is present, pcretest calls pcre[16|32]_exec() several times, with
       different values in the match_limit and match_limit_recursion fields of
       the  pcre[16|32]_extra  data structure, until it finds the minimum num-
       bers for each parameter that allow pcre[16|32]_exec() to complete with-
       out  error.  Because  this  is testing a specific feature of the normal
       interpretive pcre[16|32]_exec() execution, the use of any JIT optimiza-
       tion  that might have been set up by the /S+ qualifier of -s+ option is
       disabled.

       The match_limit number is a measure of the amount of backtracking  that
       takes  place,  and  checking it out can be instructive. For most simple
       matches, the number is quite small, but for patterns  with  very  large
       numbers  of  matching  possibilities,  it can become large very quickly
       with increasing length of  subject  string.  The  match_limit_recursion
       number  is  a  measure  of how much stack (or, if PCRE is compiled with
       NO_RECURSE, how much heap) memory  is  needed  to  complete  the  match
       attempt.

       When  \O  is  used, the value specified may be higher or lower than the
       size set by the -O command line option (or defaulted to 45); \O applies
       only  to  the  call  of  pcre[16|32]_exec()  for  the  line in which it
       appears.

       If the /P modifier was present on the pattern, causing the POSIX  wrap-
       per  API  to  be  used, the only option-setting sequences that have any
       effect are \B,  \N,  and  \Z,  causing  REG_NOTBOL,  REG_NOTEMPTY,  and
       REG_NOTEOL, respectively, to be passed to regexec().


THE ALTERNATIVE MATCHING FUNCTION

       By   default,  pcretest  uses  the  standard  PCRE  matching  function,
       pcre[16|32]_exec() to match each  data  line.  PCRE  also  supports  an
       alternative  matching  function, pcre[16|32]_dfa_test(), which operates
       in a different way, and has some restrictions. The differences  between
       the two functions are described in the pcrematching documentation.

       If  a data line contains the \D escape sequence, or if the command line
       contains the -dfa option, the alternative matching  function  is  used.
       This function finds all possible matches at a given point. If, however,
       the \F escape sequence is present in the data line, it stops after  the
       first match is found. This is always the shortest possible match.


DEFAULT OUTPUT FROM PCRETEST

       This  section  describes  the output when the normal matching function,
       pcre[16|32]_exec(), is being used.

       When a match succeeds, pcretest outputs the list of captured substrings
       that  pcre[16|32]_exec() returns, starting with number 0 for the string
       that matched the whole pattern. Otherwise, it outputs "No  match"  when
       the  return is PCRE_ERROR_NOMATCH, and "Partial match:" followed by the
       partially   matching   substring   when   pcre[16|32]_exec()    returns
       PCRE_ERROR_PARTIAL.  (Note  that  this is the entire substring that was
       inspected during the partial match; it may  include  characters  before
       the  actual  match  start  if a lookbehind assertion, \K, \b, or \B was
       involved.) For any other return, pcretest  outputs  the  PCRE  negative
       error  number  and a short descriptive phrase. If the error is a failed
       UTF string check, the offset of the start of the failing character  and
       the  reason  code are also output, provided that the size of the output
       vector is at least two. Here is an example of an  interactive  pcretest
       run.

         $ pcretest
         PCRE version 8.13 2011-04-30

           re> /^abc(\d+)/
         data> abc123
          0: abc123
          1: 123
         data> xyz
         No match

       Unset capturing substrings that are not followed by one that is set are
       not returned by pcre[16|32]_exec(), and are not shown by  pcretest.  In
       the following example, there are two capturing substrings, but when the
       first data line is matched, the second, unset substring is  not  shown.
       An  "internal" unset substring is shown as "<unset>", as for the second
       data line.

           re> /(a)|(b)/
         data> a
          0: a
          1: a
         data> b
          0: b
          1: <unset>
          2: b

       If the strings contain any non-printing characters, they are output  as
       \xhh  escapes  if  the  value is less than 256 and UTF mode is not set.
       Otherwise they are output as \x{hh...} escapes. See below for the defi-
       nition  of non-printing characters. If the pattern has the /+ modifier,
       the output for substring 0 is followed by the the rest of  the  subject
       string, identified by "0+" like this:

           re> /cat/+
         data> cataract
          0: cat
          0+ aract

       If  the  pattern  has  the /g or /G modifier, the results of successive
       matching attempts are output in sequence, like this:

           re> /\Bi(\w\w)/g
         data> Mississippi
          0: iss
          1: ss
          0: iss
          1: ss
          0: ipp
          1: pp

       "No match" is output only if the first match attempt fails. Here is  an
       example  of a failure message (the offset 4 that is specified by \>4 is
       past the end of the subject string):

           re> /xyz/
         data> xyz\>4
         Error -24 (bad offset value)

       If any of the sequences \C, \G, or \L are present in a data  line  that
       is  successfully  matched,  the substrings extracted by the convenience
       functions are output with C, G, or L after the string number instead of
       a colon. This is in addition to the normal full list. The string length
       (that is, the return from the extraction function) is given  in  paren-
       theses after each string for \C and \G.

       Note that whereas patterns can be continued over several lines (a plain
       ">" prompt is used for continuations), data lines may not. However new-
       lines  can  be included in data by means of the \n escape (or \r, \r\n,
       etc., depending on the newline sequence setting).


OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION

       When the alternative matching function, pcre[16|32]_dfa_exec(), is used
       (by  means  of the \D escape sequence or the -dfa command line option),
       the output consists of a list of all the  matches  that  start  at  the
       first point in the subject where there is at least one match. For exam-
       ple:

           re> /(tang|tangerine|tan)/
         data> yellow tangerine\D
          0: tangerine
          1: tang
          2: tan

       (Using the normal matching function on this data  finds  only  "tang".)
       The  longest matching string is always given first (and numbered zero).
       After a PCRE_ERROR_PARTIAL return, the output is "Partial match:", fol-
       lowed  by  the  partially  matching  substring.  (Note that this is the
       entire substring that was inspected during the partial  match;  it  may
       include characters before the actual match start if a lookbehind asser-
       tion, \K, \b, or \B was involved.)

       If /g is present on the pattern, the search for further matches resumes
       at the end of the longest match. For example:

           re> /(tang|tangerine|tan)/g
         data> yellow tangerine and tangy sultana\D
          0: tangerine
          1: tang
          2: tan
          0: tang
          1: tan
          0: tan

       Since  the  matching  function  does not support substring capture, the
       escape sequences that are concerned with captured  substrings  are  not
       relevant.


RESTARTING AFTER A PARTIAL MATCH

       When the alternative matching function has given the PCRE_ERROR_PARTIAL
       return, indicating that the subject partially matched the pattern,  you
       can  restart  the match with additional subject data by means of the \R
       escape sequence. For example:

           re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
         data> 23ja\P\D
         Partial match: 23ja
         data> n05\R\D
          0: n05

       For further information about partial  matching,  see  the  pcrepartial
       documentation.


CALLOUTS

       If  the pattern contains any callout requests, pcretest's callout func-
       tion is called during matching. This works  with  both  matching  func-
       tions. By default, the called function displays the callout number, the
       start and current positions in the text at the callout  time,  and  the
       next pattern item to be tested. For example:

         --->pqrabcdef
           0    ^  ^     \d

       This  output  indicates  that  callout  number  0  occurred for a match
       attempt starting at the fourth character of the  subject  string,  when
       the pointer was at the seventh character of the data, and when the next
       pattern item was \d. Just one circumflex is output  if  the  start  and
       current positions are the same.

       Callouts numbered 255 are assumed to be automatic callouts, inserted as
       a result of the /C pattern modifier. In this case, instead  of  showing
       the  callout  number, the offset in the pattern, preceded by a plus, is
       output. For example:

           re> /\d?[A-E]\*/C
         data> E*
         --->E*
          +0 ^      \d?
          +3 ^      [A-E]
          +8 ^^     \*
         +10 ^ ^
          0: E*

       If a pattern contains (*MARK) items, an additional line is output when-
       ever  a  change  of  latest mark is passed to the callout function. For
       example:

           re> /a(*MARK:X)bc/C
         data> abc
         --->abc
          +0 ^       a
          +1 ^^      (*MARK:X)
         +10 ^^      b
         Latest Mark: X
         +11 ^ ^     c
         +12 ^  ^
          0: abc

       The mark changes between matching "a" and "b", but stays the  same  for
       the  rest  of  the match, so nothing more is output. If, as a result of
       backtracking, the mark reverts to being unset, the  text  "<unset>"  is
       output.

       The  callout  function  in pcretest returns zero (carry on matching) by
       default, but you can use a \C item in a data line (as described  above)
       to change this and other parameters of the callout.

       Inserting  callouts can be helpful when using pcretest to check compli-
       cated regular expressions. For further information about callouts,  see
       the pcrecallout documentation.


NON-PRINTING CHARACTERS

       When  pcretest is outputting text in the compiled version of a pattern,
       bytes other than 32-126 are always treated as  non-printing  characters
       are are therefore shown as hex escapes.

       When  pcretest  is  outputting text that is a matched part of a subject
       string, it behaves in the same way, unless a different locale has  been
       set  for  the  pattern  (using  the  /L  modifier).  In  this case, the
       isprint() function to distinguish printing and non-printing characters.


SAVING AND RELOADING COMPILED PATTERNS

       The facilities described in this section are  not  available  when  the
       POSIX  interface  to  PCRE  is being used, that is, when the /P pattern
       modifier is specified.

       When the POSIX interface is not in use, you can cause pcretest to write
       a  compiled  pattern to a file, by following the modifiers with > and a
       file name.  For example:

         /pattern/im >/some/file

       See the pcreprecompile documentation for a discussion about saving  and
       re-using  compiled patterns.  Note that if the pattern was successfully
       studied with JIT optimization, the JIT data cannot be saved.

       The data that is written is binary.  The  first  eight  bytes  are  the
       length  of  the  compiled  pattern  data  followed by the length of the
       optional study data, each written as four  bytes  in  big-endian  order
       (most  significant  byte  first). If there is no study data (either the
       pattern was not studied, or studying did not return any data), the sec-
       ond  length  is  zero. The lengths are followed by an exact copy of the
       compiled pattern. If there is additional study  data,  this  (excluding
       any  JIT  data)  follows  immediately after the compiled pattern. After
       writing the file, pcretest expects to read a new pattern.

       A saved pattern can be reloaded into pcretest by  specifying  <  and  a
       file  name  instead  of a pattern. There must be no space between < and
       the file name, which must not  contain  a  <  character,  as  otherwise
       pcretest  will  interpret  the line as a pattern delimited by < charac-
       ters. For example:

          re> </some/file
         Compiled pattern loaded from /some/file
         No study data

       If the pattern was previously studied with the  JIT  optimization,  the
       JIT  information cannot be saved and restored, and so is lost. When the
       pattern has been loaded, pcretest proceeds to read data  lines  in  the
       usual way.

       You  can copy a file written by pcretest to a different host and reload
       it there, even if the new host has opposite endianness to  the  one  on
       which  the pattern was compiled. For example, you can compile on an i86
       machine and run on a SPARC machine. When a pattern  is  reloaded  on  a
       host with different endianness, the confirmation message is changed to:

         Compiled pattern (byte-inverted) loaded from /some/file

       The test suite contains some saved pre-compiled patterns with different
       endianness. These are reloaded using "<!" instead  of  just  "<".  This
       suppresses the "(byte-inverted)" text so that the output is the same on
       all hosts. It also forces debugging output once the  pattern  has  been
       reloaded.

       File  names  for  saving and reloading can be absolute or relative, but
       note that the shell facility of expanding a file name that starts  with
       a tilde (~) is not available.

       The  ability to save and reload files in pcretest is intended for test-
       ing and experimentation. It is not intended for production use  because
       only  a  single pattern can be written to a file. Furthermore, there is
       no facility for supplying  custom  character  tables  for  use  with  a
       reloaded  pattern.  If  the  original  pattern was compiled with custom
       tables, an attempt to match a subject string using a  reloaded  pattern
       is  likely to cause pcretest to crash.  Finally, if you attempt to load
       a file that is not in the correct format, the result is undefined.


SEE ALSO

       pcre(3), pcre16(3),  pcre32(3),  pcreapi(3),  pcrecallout(3),  pcrejit,
       pcrematching(3), pcrepartial(d), pcrepattern(3), pcreprecompile(3).


AUTHOR

       Philip Hazel
       University Computing Service
       Cambridge CB2 3QH, England.


REVISION

       Last updated: 09 February 2014
       Copyright (c) 1997-2014 University of Cambridge.