/[pcre]/code/trunk/doc/pcretest.txt
ViewVC logotype

Contents of /code/trunk/doc/pcretest.txt

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1703 - (show annotations)
Tue Jun 13 16:53:16 2017 UTC (2 years, 4 months ago) by ph10
File MIME type: text/plain
File size: 54466 byte(s)
Error occurred while calculating annotation data.
File tidies for 8.41-RC1 release.
1 PCRETEST(1) General Commands Manual PCRETEST(1)
2
3
4
5 NAME
6 pcretest - a program for testing Perl-compatible regular expressions.
7
8 SYNOPSIS
9
10 pcretest [options] [input file [output file]]
11
12 pcretest was written as a test program for the PCRE regular expression
13 library itself, but it can also be used for experimenting with regular
14 expressions. This document describes the features of the test program;
15 for details of the regular expressions themselves, see the pcrepattern
16 documentation. For details of the PCRE library function calls and their
17 options, see the pcreapi , pcre16 and pcre32 documentation.
18
19 The input for pcretest is a sequence of regular expression patterns and
20 strings to be matched, as described below. The output shows the result
21 of each match. Options on the command line and the patterns control
22 PCRE options and exactly what is output.
23
24 As PCRE has evolved, it has acquired many different features, and as a
25 result, pcretest now has rather a lot of obscure options for testing
26 every possible feature. Some of these options are specifically designed
27 for use in conjunction with the test script and data files that are
28 distributed as part of PCRE, and are unlikely to be of use otherwise.
29 They are all documented here, but without much justification.
30
31
32 INPUT DATA FORMAT
33
34 Input to pcretest is processed line by line, either by calling the C
35 library's fgets() function, or via the libreadline library (see below).
36 In Unix-like environments, fgets() treats any bytes other than newline
37 as data characters. However, in some Windows environments character 26
38 (hex 1A) causes an immediate end of file, and no further data is read.
39 For maximum portability, therefore, it is safest to use only ASCII
40 characters in pcretest input files.
41
42 The input is processed using using C's string functions, so must not
43 contain binary zeroes, even though in Unix-like environments, fgets()
44 treats any bytes other than newline as data characters.
45
46
47 PCRE's 8-BIT, 16-BIT AND 32-BIT LIBRARIES
48
49 From release 8.30, two separate PCRE libraries can be built. The origi-
50 nal one supports 8-bit character strings, whereas the newer 16-bit
51 library supports character strings encoded in 16-bit units. From
52 release 8.32, a third library can be built, supporting character
53 strings encoded in 32-bit units. The pcretest program can be used to
54 test all three libraries. However, it is itself still an 8-bit program,
55 reading 8-bit input and writing 8-bit output. When testing the 16-bit
56 or 32-bit library, the patterns and data strings are converted to 16-
57 or 32-bit format before being passed to the PCRE library functions.
58 Results are converted to 8-bit for output.
59
60 References to functions and structures of the form pcre[16|32]_xx below
61 mean "pcre_xx when using the 8-bit library, pcre16_xx when using the
62 16-bit library, or pcre32_xx when using the 32-bit library".
63
64
65 COMMAND LINE OPTIONS
66
67 -8 If both the 8-bit library has been built, this option causes
68 the 8-bit library to be used (which is the default); if the
69 8-bit library has not been built, this option causes an
70 error.
71
72 -16 If both the 8-bit or the 32-bit, and the 16-bit libraries
73 have been built, this option causes the 16-bit library to be
74 used. If only the 16-bit library has been built, this is the
75 default (so has no effect). If only the 8-bit or the 32-bit
76 library has been built, this option causes an error.
77
78 -32 If both the 8-bit or the 16-bit, and the 32-bit libraries
79 have been built, this option causes the 32-bit library to be
80 used. If only the 32-bit library has been built, this is the
81 default (so has no effect). If only the 8-bit or the 16-bit
82 library has been built, this option causes an error.
83
84 -b Behave as if each pattern has the /B (show byte code) modi-
85 fier; the internal form is output after compilation.
86
87 -C Output the version number of the PCRE library, and all avail-
88 able information about the optional features that are
89 included, and then exit with zero exit code. All other
90 options are ignored.
91
92 -C option Output information about a specific build-time option, then
93 exit. This functionality is intended for use in scripts such
94 as RunTest. The following options output the value and set
95 the exit code as indicated:
96
97 ebcdic-nl the code for LF (= NL) in an EBCDIC environment:
98 0x15 or 0x25
99 0 if used in an ASCII environment
100 exit code is always 0
101 linksize the configured internal link size (2, 3, or 4)
102 exit code is set to the link size
103 newline the default newline setting:
104 CR, LF, CRLF, ANYCRLF, or ANY
105 exit code is always 0
106 bsr the default setting for what \R matches:
107 ANYCRLF or ANY
108 exit code is always 0
109
110 The following options output 1 for true or 0 for false, and
111 set the exit code to the same value:
112
113 ebcdic compiled for an EBCDIC environment
114 jit just-in-time support is available
115 pcre16 the 16-bit library was built
116 pcre32 the 32-bit library was built
117 pcre8 the 8-bit library was built
118 ucp Unicode property support is available
119 utf UTF-8 and/or UTF-16 and/or UTF-32 support
120 is available
121
122 If an unknown option is given, an error message is output;
123 the exit code is 0.
124
125 -d Behave as if each pattern has the /D (debug) modifier; the
126 internal form and information about the compiled pattern is
127 output after compilation; -d is equivalent to -b -i.
128
129 -dfa Behave as if each data line contains the \D escape sequence;
130 this causes the alternative matching function,
131 pcre[16|32]_dfa_exec(), to be used instead of the standard
132 pcre[16|32]_exec() function (more detail is given below).
133
134 -help Output a brief summary these options and then exit.
135
136 -i Behave as if each pattern has the /I modifier; information
137 about the compiled pattern is given after compilation.
138
139 -M Behave as if each data line contains the \M escape sequence;
140 this causes PCRE to discover the minimum MATCH_LIMIT and
141 MATCH_LIMIT_RECURSION settings by calling pcre[16|32]_exec()
142 repeatedly with different limits.
143
144 -m Output the size of each compiled pattern after it has been
145 compiled. This is equivalent to adding /M to each regular
146 expression. The size is given in bytes for both libraries.
147
148 -O Behave as if each pattern has the /O modifier, that is dis-
149 able auto-possessification for all patterns.
150
151 -o osize Set the number of elements in the output vector that is used
152 when calling pcre[16|32]_exec() or pcre[16|32]_dfa_exec() to
153 be osize. The default value is 45, which is enough for 14
154 capturing subexpressions for pcre[16|32]_exec() or 22 differ-
155 ent matches for pcre[16|32]_dfa_exec(). The vector size can
156 be changed for individual matching calls by including \O in
157 the data line (see below).
158
159 -p Behave as if each pattern has the /P modifier; the POSIX
160 wrapper API is used to call PCRE. None of the other options
161 has any effect when -p is set. This option can be used only
162 with the 8-bit library.
163
164 -q Do not output the version number of pcretest at the start of
165 execution.
166
167 -S size On Unix-like systems, set the size of the run-time stack to
168 size megabytes.
169
170 -s or -s+ Behave as if each pattern has the /S modifier; in other
171 words, force each pattern to be studied. If -s+ is used, all
172 the JIT compile options are passed to pcre[16|32]_study(),
173 causing just-in-time optimization to be set up if it is
174 available, for both full and partial matching. Specific JIT
175 compile options can be selected by following -s+ with a digit
176 in the range 1 to 7, which selects the JIT compile modes as
177 follows:
178
179 1 normal match only
180 2 soft partial match only
181 3 normal match and soft partial match
182 4 hard partial match only
183 6 soft and hard partial match
184 7 all three modes (default)
185
186 If -s++ is used instead of -s+ (with or without a following
187 digit), the text "(JIT)" is added to the first output line
188 after a match or no match when JIT-compiled code was actually
189 used.
190
191 Note that there are pattern options that can override -s,
192 either specifying no studying at all, or suppressing JIT com-
193 pilation.
194
195 If the /I or /D option is present on a pattern (requesting
196 output about the compiled pattern), information about the
197 result of studying is not included when studying is caused
198 only by -s and neither -i nor -d is present on the command
199 line. This behaviour means that the output from tests that
200 are run with and without -s should be identical, except when
201 options that output information about the actual running of a
202 match are set.
203
204 The -M, -t, and -tm options, which give information about
205 resources used, are likely to produce different output with
206 and without -s. Output may also differ if the /C option is
207 present on an individual pattern. This uses callouts to trace
208 the the matching process, and this may be different between
209 studied and non-studied patterns. If the pattern contains
210 (*MARK) items there may also be differences, for the same
211 reason. The -s command line option can be overridden for spe-
212 cific patterns that should never be studied (see the /S pat-
213 tern modifier below).
214
215 -t Run each compile, study, and match many times with a timer,
216 and output the resulting times per compile, study, or match
217 (in milliseconds). Do not set -m with -t, because you will
218 then get the size output a zillion times, and the timing will
219 be distorted. You can control the number of iterations that
220 are used for timing by following -t with a number (as a sepa-
221 rate item on the command line). For example, "-t 1000" iter-
222 ates 1000 times. The default is to iterate 500000 times.
223
224 -tm This is like -t except that it times only the matching phase,
225 not the compile or study phases.
226
227 -T -TM These behave like -t and -tm, but in addition, at the end of
228 a run, the total times for all compiles, studies, and matches
229 are output.
230
231
232 DESCRIPTION
233
234 If pcretest is given two filename arguments, it reads from the first
235 and writes to the second. If it is given only one filename argument, it
236 reads from that file and writes to stdout. Otherwise, it reads from
237 stdin and writes to stdout, and prompts for each line of input, using
238 "re>" to prompt for regular expressions, and "data>" to prompt for data
239 lines.
240
241 When pcretest is built, a configuration option can specify that it
242 should be linked with the libreadline library. When this is done, if
243 the input is from a terminal, it is read using the readline() function.
244 This provides line-editing and history facilities. The output from the
245 -help option states whether or not readline() will be used.
246
247 The program handles any number of sets of input on a single input file.
248 Each set starts with a regular expression, and continues with any num-
249 ber of data lines to be matched against that pattern.
250
251 Each data line is matched separately and independently. If you want to
252 do multi-line matches, you have to use the \n escape sequence (or \r or
253 \r\n, etc., depending on the newline setting) in a single line of input
254 to encode the newline sequences. There is no limit on the length of
255 data lines; the input buffer is automatically extended if it is too
256 small.
257
258 An empty line signals the end of the data lines, at which point a new
259 regular expression is read. The regular expressions are given enclosed
260 in any non-alphanumeric delimiters other than backslash, for example:
261
262 /(a|bc)x+yz/
263
264 White space before the initial delimiter is ignored. A regular expres-
265 sion may be continued over several input lines, in which case the new-
266 line characters are included within it. It is possible to include the
267 delimiter within the pattern by escaping it, for example
268
269 /abc\/def/
270
271 If you do so, the escape and the delimiter form part of the pattern,
272 but since delimiters are always non-alphanumeric, this does not affect
273 its interpretation. If the terminating delimiter is immediately fol-
274 lowed by a backslash, for example,
275
276 /abc/\
277
278 then a backslash is added to the end of the pattern. This is done to
279 provide a way of testing the error condition that arises if a pattern
280 finishes with a backslash, because
281
282 /abc\/
283
284 is interpreted as the first line of a pattern that starts with "abc/",
285 causing pcretest to read the next line as a continuation of the regular
286 expression.
287
288
289 PATTERN MODIFIERS
290
291 A pattern may be followed by any number of modifiers, which are mostly
292 single characters, though some of these can be qualified by further
293 characters. Following Perl usage, these are referred to below as, for
294 example, "the /i modifier", even though the delimiter of the pattern
295 need not always be a slash, and no slash is used when writing modi-
296 fiers. White space may appear between the final pattern delimiter and
297 the first modifier, and between the modifiers themselves. For refer-
298 ence, here is a complete list of modifiers. They fall into several
299 groups that are described in detail in the following sections.
300
301 /8 set UTF mode
302 /9 set PCRE_NEVER_UTF (locks out UTF mode)
303 /? disable UTF validity check
304 /+ show remainder of subject after match
305 /= show all captures (not just those that are set)
306
307 /A set PCRE_ANCHORED
308 /B show compiled code
309 /C set PCRE_AUTO_CALLOUT
310 /D same as /B plus /I
311 /E set PCRE_DOLLAR_ENDONLY
312 /F flip byte order in compiled pattern
313 /f set PCRE_FIRSTLINE
314 /G find all matches (shorten string)
315 /g find all matches (use startoffset)
316 /I show information about pattern
317 /i set PCRE_CASELESS
318 /J set PCRE_DUPNAMES
319 /K show backtracking control names
320 /L set locale
321 /M show compiled memory size
322 /m set PCRE_MULTILINE
323 /N set PCRE_NO_AUTO_CAPTURE
324 /O set PCRE_NO_AUTO_POSSESS
325 /P use the POSIX wrapper
326 /Q test external stack check function
327 /S study the pattern after compilation
328 /s set PCRE_DOTALL
329 /T select character tables
330 /U set PCRE_UNGREEDY
331 /W set PCRE_UCP
332 /X set PCRE_EXTRA
333 /x set PCRE_EXTENDED
334 /Y set PCRE_NO_START_OPTIMIZE
335 /Z don't show lengths in /B output
336
337 /<any> set PCRE_NEWLINE_ANY
338 /<anycrlf> set PCRE_NEWLINE_ANYCRLF
339 /<cr> set PCRE_NEWLINE_CR
340 /<crlf> set PCRE_NEWLINE_CRLF
341 /<lf> set PCRE_NEWLINE_LF
342 /<bsr_anycrlf> set PCRE_BSR_ANYCRLF
343 /<bsr_unicode> set PCRE_BSR_UNICODE
344 /<JS> set PCRE_JAVASCRIPT_COMPAT
345
346
347 Perl-compatible modifiers
348
349 The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
350 PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when
351 pcre[16|32]_compile() is called. These four modifier letters have the
352 same effect as they do in Perl. For example:
353
354 /caseless/i
355
356
357 Modifiers for other PCRE options
358
359 The following table shows additional modifiers for setting PCRE com-
360 pile-time options that do not correspond to anything in Perl:
361
362 /8 PCRE_UTF8 ) when using the 8-bit
363 /? PCRE_NO_UTF8_CHECK ) library
364
365 /8 PCRE_UTF16 ) when using the 16-bit
366 /? PCRE_NO_UTF16_CHECK ) library
367
368 /8 PCRE_UTF32 ) when using the 32-bit
369 /? PCRE_NO_UTF32_CHECK ) library
370
371 /9 PCRE_NEVER_UTF
372 /A PCRE_ANCHORED
373 /C PCRE_AUTO_CALLOUT
374 /E PCRE_DOLLAR_ENDONLY
375 /f PCRE_FIRSTLINE
376 /J PCRE_DUPNAMES
377 /N PCRE_NO_AUTO_CAPTURE
378 /O PCRE_NO_AUTO_POSSESS
379 /U PCRE_UNGREEDY
380 /W PCRE_UCP
381 /X PCRE_EXTRA
382 /Y PCRE_NO_START_OPTIMIZE
383 /<any> PCRE_NEWLINE_ANY
384 /<anycrlf> PCRE_NEWLINE_ANYCRLF
385 /<cr> PCRE_NEWLINE_CR
386 /<crlf> PCRE_NEWLINE_CRLF
387 /<lf> PCRE_NEWLINE_LF
388 /<bsr_anycrlf> PCRE_BSR_ANYCRLF
389 /<bsr_unicode> PCRE_BSR_UNICODE
390 /<JS> PCRE_JAVASCRIPT_COMPAT
391
392 The modifiers that are enclosed in angle brackets are literal strings
393 as shown, including the angle brackets, but the letters within can be
394 in either case. This example sets multiline matching with CRLF as the
395 line ending sequence:
396
397 /^abc/m<CRLF>
398
399 As well as turning on the PCRE_UTF8/16/32 option, the /8 modifier
400 causes all non-printing characters in output strings to be printed
401 using the \x{hh...} notation. Otherwise, those less than 0x100 are out-
402 put in hex without the curly brackets.
403
404 Full details of the PCRE options are given in the pcreapi documenta-
405 tion.
406
407 Finding all matches in a string
408
409 Searching for all possible matches within each subject string can be
410 requested by the /g or /G modifier. After finding a match, PCRE is
411 called again to search the remainder of the subject string. The differ-
412 ence between /g and /G is that the former uses the startoffset argument
413 to pcre[16|32]_exec() to start searching at a new point within the
414 entire string (which is in effect what Perl does), whereas the latter
415 passes over a shortened substring. This makes a difference to the
416 matching process if the pattern begins with a lookbehind assertion
417 (including \b or \B).
418
419 If any call to pcre[16|32]_exec() in a /g or /G sequence matches an
420 empty string, the next call is done with the PCRE_NOTEMPTY_ATSTART and
421 PCRE_ANCHORED flags set in order to search for another, non-empty,
422 match at the same point. If this second match fails, the start offset
423 is advanced, and the normal match is retried. This imitates the way
424 Perl handles such cases when using the /g modifier or the split() func-
425 tion. Normally, the start offset is advanced by one character, but if
426 the newline convention recognizes CRLF as a newline, and the current
427 character is CR followed by LF, an advance of two is used.
428
429 Other modifiers
430
431 There are yet more modifiers for controlling the way pcretest operates.
432
433 The /+ modifier requests that as well as outputting the substring that
434 matched the entire pattern, pcretest should in addition output the
435 remainder of the subject string. This is useful for tests where the
436 subject contains multiple copies of the same substring. If the + modi-
437 fier appears twice, the same action is taken for captured substrings.
438 In each case the remainder is output on the following line with a plus
439 character following the capture number. Note that this modifier must
440 not immediately follow the /S modifier because /S+ and /S++ have other
441 meanings.
442
443 The /= modifier requests that the values of all potential captured
444 parentheses be output after a match. By default, only those up to the
445 highest one actually used in the match are output (corresponding to the
446 return code from pcre[16|32]_exec()). Values in the offsets vector cor-
447 responding to higher numbers should be set to -1, and these are output
448 as "<unset>". This modifier gives a way of checking that this is hap-
449 pening.
450
451 The /B modifier is a debugging feature. It requests that pcretest out-
452 put a representation of the compiled code after compilation. Normally
453 this information contains length and offset values; however, if /Z is
454 also present, this data is replaced by spaces. This is a special fea-
455 ture for use in the automatic test scripts; it ensures that the same
456 output is generated for different internal link sizes.
457
458 The /D modifier is a PCRE debugging feature, and is equivalent to /BI,
459 that is, both the /B and the /I modifiers.
460
461 The /F modifier causes pcretest to flip the byte order of the 2-byte
462 and 4-byte fields in the compiled pattern. This facility is for testing
463 the feature in PCRE that allows it to execute patterns that were com-
464 piled on a host with a different endianness. This feature is not avail-
465 able when the POSIX interface to PCRE is being used, that is, when the
466 /P pattern modifier is specified. See also the section about saving and
467 reloading compiled patterns below.
468
469 The /I modifier requests that pcretest output information about the
470 compiled pattern (whether it is anchored, has a fixed first character,
471 and so on). It does this by calling pcre[16|32]_fullinfo() after com-
472 piling a pattern. If the pattern is studied, the results of that are
473 also output. In this output, the word "char" means a non-UTF character,
474 that is, the value of a single data item (8-bit, 16-bit, or 32-bit,
475 depending on the library that is being tested).
476
477 The /K modifier requests pcretest to show names from backtracking con-
478 trol verbs that are returned from calls to pcre[16|32]_exec(). It
479 causes pcretest to create a pcre[16|32]_extra block if one has not
480 already been created by a call to pcre[16|32]_study(), and to set the
481 PCRE_EXTRA_MARK flag and the mark field within it, every time that
482 pcre[16|32]_exec() is called. If the variable that the mark field
483 points to is non-NULL for a match, non-match, or partial match,
484 pcretest prints the string to which it points. For a match, this is
485 shown on a line by itself, tagged with "MK:". For a non-match it is
486 added to the message.
487
488 The /L modifier must be followed directly by the name of a locale, for
489 example,
490
491 /pattern/Lfr_FR
492
493 For this reason, it must be the last modifier. The given locale is set,
494 pcre[16|32]_maketables() is called to build a set of character tables
495 for the locale, and this is then passed to pcre[16|32]_compile() when
496 compiling the regular expression. Without an /L (or /T) modifier, NULL
497 is passed as the tables pointer; that is, /L applies only to the
498 expression on which it appears.
499
500 The /M modifier causes the size in bytes of the memory block used to
501 hold the compiled pattern to be output. This does not include the size
502 of the pcre[16|32] block; it is just the actual compiled data. If the
503 pattern is successfully studied with the PCRE_STUDY_JIT_COMPILE option,
504 the size of the JIT compiled code is also output.
505
506 The /Q modifier is used to test the use of pcre_stack_guard. It must be
507 followed by '0' or '1', specifying the return code to be given from an
508 external function that is passed to PCRE and used for stack checking
509 during compilation (see the pcreapi documentation for details).
510
511 The /S modifier causes pcre[16|32]_study() to be called after the
512 expression has been compiled, and the results used when the expression
513 is matched. There are a number of qualifying characters that may follow
514 /S. They may appear in any order.
515
516 If /S is followed by an exclamation mark, pcre[16|32]_study() is called
517 with the PCRE_STUDY_EXTRA_NEEDED option, causing it always to return a
518 pcre_extra block, even when studying discovers no useful information.
519
520 If /S is followed by a second S character, it suppresses studying, even
521 if it was requested externally by the -s command line option. This
522 makes it possible to specify that certain patterns are always studied,
523 and others are never studied, independently of -s. This feature is used
524 in the test files in a few cases where the output is different when the
525 pattern is studied.
526
527 If the /S modifier is followed by a + character, the call to
528 pcre[16|32]_study() is made with all the JIT study options, requesting
529 just-in-time optimization support if it is available, for both normal
530 and partial matching. If you want to restrict the JIT compiling modes,
531 you can follow /S+ with a digit in the range 1 to 7:
532
533 1 normal match only
534 2 soft partial match only
535 3 normal match and soft partial match
536 4 hard partial match only
537 6 soft and hard partial match
538 7 all three modes (default)
539
540 If /S++ is used instead of /S+ (with or without a following digit), the
541 text "(JIT)" is added to the first output line after a match or no
542 match when JIT-compiled code was actually used.
543
544 Note that there is also an independent /+ modifier; it must not be
545 given immediately after /S or /S+ because this will be misinterpreted.
546
547 If JIT studying is successful, the compiled JIT code will automatically
548 be used when pcre[16|32]_exec() is run, except when incompatible run-
549 time options are specified. For more details, see the pcrejit documen-
550 tation. See also the \J escape sequence below for a way of setting the
551 size of the JIT stack.
552
553 Finally, if /S is followed by a minus character, JIT compilation is
554 suppressed, even if it was requested externally by the -s command line
555 option. This makes it possible to specify that JIT is never to be used
556 for certain patterns.
557
558 The /T modifier must be followed by a single digit. It causes a spe-
559 cific set of built-in character tables to be passed to pcre[16|32]_com-
560 pile(). It is used in the standard PCRE tests to check behaviour with
561 different character tables. The digit specifies the tables as follows:
562
563 0 the default ASCII tables, as distributed in
564 pcre_chartables.c.dist
565 1 a set of tables defining ISO 8859 characters
566
567 In table 1, some characters whose codes are greater than 128 are iden-
568 tified as letters, digits, spaces, etc.
569
570 Using the POSIX wrapper API
571
572 The /P modifier causes pcretest to call PCRE via the POSIX wrapper API
573 rather than its native API. This supports only the 8-bit library. When
574 /P is set, the following modifiers set options for the regcomp() func-
575 tion:
576
577 /i REG_ICASE
578 /m REG_NEWLINE
579 /N REG_NOSUB
580 /s REG_DOTALL )
581 /U REG_UNGREEDY ) These options are not part of
582 /W REG_UCP ) the POSIX standard
583 /8 REG_UTF8 )
584
585 The /+ modifier works as described above. All other modifiers are
586 ignored.
587
588 Locking out certain modifiers
589
590 PCRE can be compiled with or without support for certain features such
591 as UTF-8/16/32 or Unicode properties. Accordingly, the standard tests
592 are split up into a number of different files that are selected for
593 running depending on which features are available. When updating the
594 tests, it is all too easy to put a new test into the wrong file by mis-
595 take; for example, to put a test that requires UTF support into a file
596 that is used when it is not available. To help detect such mistakes as
597 early as possible, there is a facility for locking out specific modi-
598 fiers. If an input line for pcretest starts with the string "< forbid "
599 the following sequence of characters is taken as a list of forbidden
600 modifiers. For example, in the test files that must not use UTF or Uni-
601 code property support, this line appears:
602
603 < forbid 8W
604
605 This locks out the /8 and /W modifiers. An immediate error is given if
606 they are subsequently encountered. If the character string contains <
607 but not >, all the multi-character modifiers that begin with < are
608 locked out. Otherwise, such modifiers must be explicitly listed, for
609 example:
610
611 < forbid <JS><cr>
612
613 There must be a single space between < and "forbid" for this feature to
614 be recognised. If there is not, the line is interpreted either as a
615 request to re-load a pre-compiled pattern (see "SAVING AND RELOADING
616 COMPILED PATTERNS" below) or, if there is a another < character, as a
617 pattern that uses < as its delimiter.
618
619
620 DATA LINES
621
622 Before each data line is passed to pcre[16|32]_exec(), leading and
623 trailing white space is removed, and it is then scanned for \ escapes.
624 Some of these are pretty esoteric features, intended for checking out
625 some of the more complicated features of PCRE. If you are just testing
626 "ordinary" regular expressions, you probably don't need any of these.
627 The following escapes are recognized:
628
629 \a alarm (BEL, \x07)
630 \b backspace (\x08)
631 \e escape (\x27)
632 \f form feed (\x0c)
633 \n newline (\x0a)
634 \qdd set the PCRE_MATCH_LIMIT limit to dd
635 (any number of digits)
636 \r carriage return (\x0d)
637 \t tab (\x09)
638 \v vertical tab (\x0b)
639 \nnn octal character (up to 3 octal digits); always
640 a byte unless > 255 in UTF-8 or 16-bit or 32-bit mode
641 \o{dd...} octal character (any number of octal digits}
642 \xhh hexadecimal byte (up to 2 hex digits)
643 \x{hh...} hexadecimal character (any number of hex digits)
644 \A pass the PCRE_ANCHORED option to pcre[16|32]_exec()
645 or pcre[16|32]_dfa_exec()
646 \B pass the PCRE_NOTBOL option to pcre[16|32]_exec()
647 or pcre[16|32]_dfa_exec()
648 \Cdd call pcre[16|32]_copy_substring() for substring dd
649 after a successful match (number less than 32)
650 \Cname call pcre[16|32]_copy_named_substring() for substring
651 "name" after a successful match (name termin-
652 ated by next non alphanumeric character)
653 \C+ show the current captured substrings at callout
654 time
655 \C- do not supply a callout function
656 \C!n return 1 instead of 0 when callout number n is
657 reached
658 \C!n!m return 1 instead of 0 when callout number n is
659 reached for the nth time
660 \C*n pass the number n (may be negative) as callout
661 data; this is used as the callout return value
662 \D use the pcre[16|32]_dfa_exec() match function
663 \F only shortest match for pcre[16|32]_dfa_exec()
664 \Gdd call pcre[16|32]_get_substring() for substring dd
665 after a successful match (number less than 32)
666 \Gname call pcre[16|32]_get_named_substring() for substring
667 "name" after a successful match (name termin-
668 ated by next non-alphanumeric character)
669 \Jdd set up a JIT stack of dd kilobytes maximum (any
670 number of digits)
671 \L call pcre[16|32]_get_substringlist() after a
672 successful match
673 \M discover the minimum MATCH_LIMIT and
674 MATCH_LIMIT_RECURSION settings
675 \N pass the PCRE_NOTEMPTY option to pcre[16|32]_exec()
676 or pcre[16|32]_dfa_exec(); if used twice, pass the
677 PCRE_NOTEMPTY_ATSTART option
678 \Odd set the size of the output vector passed to
679 pcre[16|32]_exec() to dd (any number of digits)
680 \P pass the PCRE_PARTIAL_SOFT option to pcre[16|32]_exec()
681 or pcre[16|32]_dfa_exec(); if used twice, pass the
682 PCRE_PARTIAL_HARD option
683 \Qdd set the PCRE_MATCH_LIMIT_RECURSION limit to dd
684 (any number of digits)
685 \R pass the PCRE_DFA_RESTART option to pcre[16|32]_dfa_exec()
686 \S output details of memory get/free calls during matching
687 \Y pass the PCRE_NO_START_OPTIMIZE option to
688 pcre[16|32]_exec()
689 or pcre[16|32]_dfa_exec()
690 \Z pass the PCRE_NOTEOL option to pcre[16|32]_exec()
691 or pcre[16|32]_dfa_exec()
692 \? pass the PCRE_NO_UTF[8|16|32]_CHECK option to
693 pcre[16|32]_exec() or pcre[16|32]_dfa_exec()
694 \>dd start the match at offset dd (optional "-"; then
695 any number of digits); this sets the startoffset
696 argument for pcre[16|32]_exec() or
697 pcre[16|32]_dfa_exec()
698 \<cr> pass the PCRE_NEWLINE_CR option to pcre[16|32]_exec()
699 or pcre[16|32]_dfa_exec()
700 \<lf> pass the PCRE_NEWLINE_LF option to pcre[16|32]_exec()
701 or pcre[16|32]_dfa_exec()
702 \<crlf> pass the PCRE_NEWLINE_CRLF option to pcre[16|32]_exec()
703 or pcre[16|32]_dfa_exec()
704 \<anycrlf> pass the PCRE_NEWLINE_ANYCRLF option to pcre[16|32]_exec()
705 or pcre[16|32]_dfa_exec()
706 \<any> pass the PCRE_NEWLINE_ANY option to pcre[16|32]_exec()
707 or pcre[16|32]_dfa_exec()
708
709 The use of \x{hh...} is not dependent on the use of the /8 modifier on
710 the pattern. It is recognized always. There may be any number of hexa-
711 decimal digits inside the braces; invalid values provoke error mes-
712 sages.
713
714 Note that \xhh specifies one byte rather than one character in UTF-8
715 mode; this makes it possible to construct invalid UTF-8 sequences for
716 testing purposes. On the other hand, \x{hh} is interpreted as a UTF-8
717 character in UTF-8 mode, generating more than one byte if the value is
718 greater than 127. When testing the 8-bit library not in UTF-8 mode,
719 \x{hh} generates one byte for values less than 256, and causes an error
720 for greater values.
721
722 In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it
723 possible to construct invalid UTF-16 sequences for testing purposes.
724
725 In UTF-32 mode, all 4- to 8-digit \x{...} values are accepted. This
726 makes it possible to construct invalid UTF-32 sequences for testing
727 purposes.
728
729 The escapes that specify line ending sequences are literal strings,
730 exactly as shown. No more than one newline setting should be present in
731 any data line.
732
733 A backslash followed by anything else just escapes the anything else.
734 If the very last character is a backslash, it is ignored. This gives a
735 way of passing an empty line as data, since a real empty line termi-
736 nates the data input.
737
738 The \J escape provides a way of setting the maximum stack size that is
739 used by the just-in-time optimization code. It is ignored if JIT opti-
740 mization is not being used. Providing a stack that is larger than the
741 default 32K is necessary only for very complicated patterns.
742
743 If \M is present, pcretest calls pcre[16|32]_exec() several times, with
744 different values in the match_limit and match_limit_recursion fields of
745 the pcre[16|32]_extra data structure, until it finds the minimum num-
746 bers for each parameter that allow pcre[16|32]_exec() to complete with-
747 out error. Because this is testing a specific feature of the normal
748 interpretive pcre[16|32]_exec() execution, the use of any JIT optimiza-
749 tion that might have been set up by the /S+ qualifier of -s+ option is
750 disabled.
751
752 The match_limit number is a measure of the amount of backtracking that
753 takes place, and checking it out can be instructive. For most simple
754 matches, the number is quite small, but for patterns with very large
755 numbers of matching possibilities, it can become large very quickly
756 with increasing length of subject string. The match_limit_recursion
757 number is a measure of how much stack (or, if PCRE is compiled with
758 NO_RECURSE, how much heap) memory is needed to complete the match
759 attempt.
760
761 When \O is used, the value specified may be higher or lower than the
762 size set by the -O command line option (or defaulted to 45); \O applies
763 only to the call of pcre[16|32]_exec() for the line in which it
764 appears.
765
766 If the /P modifier was present on the pattern, causing the POSIX wrap-
767 per API to be used, the only option-setting sequences that have any
768 effect are \B, \N, and \Z, causing REG_NOTBOL, REG_NOTEMPTY, and
769 REG_NOTEOL, respectively, to be passed to regexec().
770
771
772 THE ALTERNATIVE MATCHING FUNCTION
773
774 By default, pcretest uses the standard PCRE matching function,
775 pcre[16|32]_exec() to match each data line. PCRE also supports an
776 alternative matching function, pcre[16|32]_dfa_test(), which operates
777 in a different way, and has some restrictions. The differences between
778 the two functions are described in the pcrematching documentation.
779
780 If a data line contains the \D escape sequence, or if the command line
781 contains the -dfa option, the alternative matching function is used.
782 This function finds all possible matches at a given point. If, however,
783 the \F escape sequence is present in the data line, it stops after the
784 first match is found. This is always the shortest possible match.
785
786
787 DEFAULT OUTPUT FROM PCRETEST
788
789 This section describes the output when the normal matching function,
790 pcre[16|32]_exec(), is being used.
791
792 When a match succeeds, pcretest outputs the list of captured substrings
793 that pcre[16|32]_exec() returns, starting with number 0 for the string
794 that matched the whole pattern. Otherwise, it outputs "No match" when
795 the return is PCRE_ERROR_NOMATCH, and "Partial match:" followed by the
796 partially matching substring when pcre[16|32]_exec() returns
797 PCRE_ERROR_PARTIAL. (Note that this is the entire substring that was
798 inspected during the partial match; it may include characters before
799 the actual match start if a lookbehind assertion, \K, \b, or \B was
800 involved.) For any other return, pcretest outputs the PCRE negative
801 error number and a short descriptive phrase. If the error is a failed
802 UTF string check, the offset of the start of the failing character and
803 the reason code are also output, provided that the size of the output
804 vector is at least two. Here is an example of an interactive pcretest
805 run.
806
807 $ pcretest
808 PCRE version 8.13 2011-04-30
809
810 re> /^abc(\d+)/
811 data> abc123
812 0: abc123
813 1: 123
814 data> xyz
815 No match
816
817 Unset capturing substrings that are not followed by one that is set are
818 not returned by pcre[16|32]_exec(), and are not shown by pcretest. In
819 the following example, there are two capturing substrings, but when the
820 first data line is matched, the second, unset substring is not shown.
821 An "internal" unset substring is shown as "<unset>", as for the second
822 data line.
823
824 re> /(a)|(b)/
825 data> a
826 0: a
827 1: a
828 data> b
829 0: b
830 1: <unset>
831 2: b
832
833 If the strings contain any non-printing characters, they are output as
834 \xhh escapes if the value is less than 256 and UTF mode is not set.
835 Otherwise they are output as \x{hh...} escapes. See below for the defi-
836 nition of non-printing characters. If the pattern has the /+ modifier,
837 the output for substring 0 is followed by the the rest of the subject
838 string, identified by "0+" like this:
839
840 re> /cat/+
841 data> cataract
842 0: cat
843 0+ aract
844
845 If the pattern has the /g or /G modifier, the results of successive
846 matching attempts are output in sequence, like this:
847
848 re> /\Bi(\w\w)/g
849 data> Mississippi
850 0: iss
851 1: ss
852 0: iss
853 1: ss
854 0: ipp
855 1: pp
856
857 "No match" is output only if the first match attempt fails. Here is an
858 example of a failure message (the offset 4 that is specified by \>4 is
859 past the end of the subject string):
860
861 re> /xyz/
862 data> xyz\>4
863 Error -24 (bad offset value)
864
865 If any of the sequences \C, \G, or \L are present in a data line that
866 is successfully matched, the substrings extracted by the convenience
867 functions are output with C, G, or L after the string number instead of
868 a colon. This is in addition to the normal full list. The string length
869 (that is, the return from the extraction function) is given in paren-
870 theses after each string for \C and \G.
871
872 Note that whereas patterns can be continued over several lines (a plain
873 ">" prompt is used for continuations), data lines may not. However new-
874 lines can be included in data by means of the \n escape (or \r, \r\n,
875 etc., depending on the newline sequence setting).
876
877
878 OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
879
880 When the alternative matching function, pcre[16|32]_dfa_exec(), is used
881 (by means of the \D escape sequence or the -dfa command line option),
882 the output consists of a list of all the matches that start at the
883 first point in the subject where there is at least one match. For exam-
884 ple:
885
886 re> /(tang|tangerine|tan)/
887 data> yellow tangerine\D
888 0: tangerine
889 1: tang
890 2: tan
891
892 (Using the normal matching function on this data finds only "tang".)
893 The longest matching string is always given first (and numbered zero).
894 After a PCRE_ERROR_PARTIAL return, the output is "Partial match:", fol-
895 lowed by the partially matching substring. (Note that this is the
896 entire substring that was inspected during the partial match; it may
897 include characters before the actual match start if a lookbehind asser-
898 tion, \K, \b, or \B was involved.)
899
900 If /g is present on the pattern, the search for further matches resumes
901 at the end of the longest match. For example:
902
903 re> /(tang|tangerine|tan)/g
904 data> yellow tangerine and tangy sultana\D
905 0: tangerine
906 1: tang
907 2: tan
908 0: tang
909 1: tan
910 0: tan
911
912 Since the matching function does not support substring capture, the
913 escape sequences that are concerned with captured substrings are not
914 relevant.
915
916
917 RESTARTING AFTER A PARTIAL MATCH
918
919 When the alternative matching function has given the PCRE_ERROR_PARTIAL
920 return, indicating that the subject partially matched the pattern, you
921 can restart the match with additional subject data by means of the \R
922 escape sequence. For example:
923
924 re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
925 data> 23ja\P\D
926 Partial match: 23ja
927 data> n05\R\D
928 0: n05
929
930 For further information about partial matching, see the pcrepartial
931 documentation.
932
933
934 CALLOUTS
935
936 If the pattern contains any callout requests, pcretest's callout func-
937 tion is called during matching. This works with both matching func-
938 tions. By default, the called function displays the callout number, the
939 start and current positions in the text at the callout time, and the
940 next pattern item to be tested. For example:
941
942 --->pqrabcdef
943 0 ^ ^ \d
944
945 This output indicates that callout number 0 occurred for a match
946 attempt starting at the fourth character of the subject string, when
947 the pointer was at the seventh character of the data, and when the next
948 pattern item was \d. Just one circumflex is output if the start and
949 current positions are the same.
950
951 Callouts numbered 255 are assumed to be automatic callouts, inserted as
952 a result of the /C pattern modifier. In this case, instead of showing
953 the callout number, the offset in the pattern, preceded by a plus, is
954 output. For example:
955
956 re> /\d?[A-E]\*/C
957 data> E*
958 --->E*
959 +0 ^ \d?
960 +3 ^ [A-E]
961 +8 ^^ \*
962 +10 ^ ^
963 0: E*
964
965 If a pattern contains (*MARK) items, an additional line is output when-
966 ever a change of latest mark is passed to the callout function. For
967 example:
968
969 re> /a(*MARK:X)bc/C
970 data> abc
971 --->abc
972 +0 ^ a
973 +1 ^^ (*MARK:X)
974 +10 ^^ b
975 Latest Mark: X
976 +11 ^ ^ c
977 +12 ^ ^
978 0: abc
979
980 The mark changes between matching "a" and "b", but stays the same for
981 the rest of the match, so nothing more is output. If, as a result of
982 backtracking, the mark reverts to being unset, the text "<unset>" is
983 output.
984
985 The callout function in pcretest returns zero (carry on matching) by
986 default, but you can use a \C item in a data line (as described above)
987 to change this and other parameters of the callout.
988
989 Inserting callouts can be helpful when using pcretest to check compli-
990 cated regular expressions. For further information about callouts, see
991 the pcrecallout documentation.
992
993
994 NON-PRINTING CHARACTERS
995
996 When pcretest is outputting text in the compiled version of a pattern,
997 bytes other than 32-126 are always treated as non-printing characters
998 are are therefore shown as hex escapes.
999
1000 When pcretest is outputting text that is a matched part of a subject
1001 string, it behaves in the same way, unless a different locale has been
1002 set for the pattern (using the /L modifier). In this case, the
1003 isprint() function to distinguish printing and non-printing characters.
1004
1005
1006 SAVING AND RELOADING COMPILED PATTERNS
1007
1008 The facilities described in this section are not available when the
1009 POSIX interface to PCRE is being used, that is, when the /P pattern
1010 modifier is specified.
1011
1012 When the POSIX interface is not in use, you can cause pcretest to write
1013 a compiled pattern to a file, by following the modifiers with > and a
1014 file name. For example:
1015
1016 /pattern/im >/some/file
1017
1018 See the pcreprecompile documentation for a discussion about saving and
1019 re-using compiled patterns. Note that if the pattern was successfully
1020 studied with JIT optimization, the JIT data cannot be saved.
1021
1022 The data that is written is binary. The first eight bytes are the
1023 length of the compiled pattern data followed by the length of the
1024 optional study data, each written as four bytes in big-endian order
1025 (most significant byte first). If there is no study data (either the
1026 pattern was not studied, or studying did not return any data), the sec-
1027 ond length is zero. The lengths are followed by an exact copy of the
1028 compiled pattern. If there is additional study data, this (excluding
1029 any JIT data) follows immediately after the compiled pattern. After
1030 writing the file, pcretest expects to read a new pattern.
1031
1032 A saved pattern can be reloaded into pcretest by specifying < and a
1033 file name instead of a pattern. There must be no space between < and
1034 the file name, which must not contain a < character, as otherwise
1035 pcretest will interpret the line as a pattern delimited by < charac-
1036 ters. For example:
1037
1038 re> </some/file
1039 Compiled pattern loaded from /some/file
1040 No study data
1041
1042 If the pattern was previously studied with the JIT optimization, the
1043 JIT information cannot be saved and restored, and so is lost. When the
1044 pattern has been loaded, pcretest proceeds to read data lines in the
1045 usual way.
1046
1047 You can copy a file written by pcretest to a different host and reload
1048 it there, even if the new host has opposite endianness to the one on
1049 which the pattern was compiled. For example, you can compile on an i86
1050 machine and run on a SPARC machine. When a pattern is reloaded on a
1051 host with different endianness, the confirmation message is changed to:
1052
1053 Compiled pattern (byte-inverted) loaded from /some/file
1054
1055 The test suite contains some saved pre-compiled patterns with different
1056 endianness. These are reloaded using "<!" instead of just "<". This
1057 suppresses the "(byte-inverted)" text so that the output is the same on
1058 all hosts. It also forces debugging output once the pattern has been
1059 reloaded.
1060
1061 File names for saving and reloading can be absolute or relative, but
1062 note that the shell facility of expanding a file name that starts with
1063 a tilde (~) is not available.
1064
1065 The ability to save and reload files in pcretest is intended for test-
1066 ing and experimentation. It is not intended for production use because
1067 only a single pattern can be written to a file. Furthermore, there is
1068 no facility for supplying custom character tables for use with a
1069 reloaded pattern. If the original pattern was compiled with custom
1070 tables, an attempt to match a subject string using a reloaded pattern
1071 is likely to cause pcretest to crash. Finally, if you attempt to load
1072 a file that is not in the correct format, the result is undefined.
1073
1074
1075 SEE ALSO
1076
1077 pcre(3), pcre16(3), pcre32(3), pcreapi(3), pcrecallout(3), pcrejit,
1078 pcrematching(3), pcrepartial(d), pcrepattern(3), pcreprecompile(3).
1079
1080
1081 AUTHOR
1082
1083 Philip Hazel
1084 University Computing Service
1085 Cambridge CB2 3QH, England.
1086
1087
1088 REVISION
1089
1090 Last updated: 23 February 2017
1091 Copyright (c) 1997-2017 University of Cambridge.

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5