1 This file contains the PCRE man page that described the pcretest program. Note
2 that not all of the features of PCRE are available in the limited version that
4 -------------------------------------------------------------------------------
6 PCRETEST(1) PCRETEST(1)
11 pcretest - a program for testing Perl-compatible regular expressions.
15 pcretest [-C] [-d] [-dfa] [-i] [-m] [-o osize] [-p] [-t] [source]
18 pcretest was written as a test program for the PCRE regular expression
19 library itself, but it can also be used for experimenting with regular
20 expressions. This document describes the features of the test program;
21 for details of the regular expressions themselves, see the pcrepattern
22 documentation. For details of the PCRE library function calls and their
23 options, see the pcreapi documentation.
28 -C Output the version number of the PCRE library, and all avail-
29 able information about the optional features that are
30 included, and then exit.
32 -d Behave as if each regex has the /D (debug) modifier; the
33 internal form is output after compilation.
35 -dfa Behave as if each data line contains the \D escape sequence;
36 this causes the alternative matching function,
37 pcre_dfa_exec(), to be used instead of the standard
38 pcre_exec() function (more detail is given below).
40 -i Behave as if each regex has the /I modifier; information
41 about the compiled pattern is given after compilation.
43 -m Output the size of each compiled pattern after it has been
44 compiled. This is equivalent to adding /M to each regular
45 expression. For compatibility with earlier versions of
46 pcretest, -s is a synonym for -m.
48 -o osize Set the number of elements in the output vector that is used
49 when calling pcre_exec() to be osize. The default value is
50 45, which is enough for 14 capturing subexpressions. The vec-
51 tor size can be changed for individual matching calls by
52 including \O in the data line (see below).
54 -p Behave as if each regex has the /P modifier; the POSIX wrap-
55 per API is used to call PCRE. None of the other options has
56 any effect when -p is set.
58 -t Run each compile, study, and match many times with a timer,
59 and output resulting time per compile or match (in millisec-
60 onds). Do not set -m with -t, because you will then get the
61 size output a zillion times, and the timing will be dis-
67 If pcretest is given two filename arguments, it reads from the first
68 and writes to the second. If it is given only one filename argument, it
69 reads from that file and writes to stdout. Otherwise, it reads from
70 stdin and writes to stdout, and prompts for each line of input, using
71 "re>" to prompt for regular expressions, and "data>" to prompt for data
74 The program handles any number of sets of input on a single input file.
75 Each set starts with a regular expression, and continues with any num-
76 ber of data lines to be matched against the pattern.
78 Each data line is matched separately and independently. If you want to
79 do multiple-line matches, you have to use the \n escape sequence in a
80 single line of input to encode the newline characters. The maximum
81 length of data line is 30,000 characters.
83 An empty line signals the end of the data lines, at which point a new
84 regular expression is read. The regular expressions are given enclosed
85 in any non-alphanumeric delimiters other than backslash, for example
89 White space before the initial delimiter is ignored. A regular expres-
90 sion may be continued over several input lines, in which case the new-
91 line characters are included within it. It is possible to include the
92 delimiter within the pattern by escaping it, for example
96 If you do so, the escape and the delimiter form part of the pattern,
97 but since delimiters are always non-alphanumeric, this does not affect
98 its interpretation. If the terminating delimiter is immediately fol-
99 lowed by a backslash, for example,
103 then a backslash is added to the end of the pattern. This is done to
104 provide a way of testing the error condition that arises if a pattern
105 finishes with a backslash, because
109 is interpreted as the first line of a pattern that starts with "abc/",
110 causing pcretest to read the next line as a continuation of the regular
116 A pattern may be followed by any number of modifiers, which are mostly
117 single characters. Following Perl usage, these are referred to below
118 as, for example, "the /i modifier", even though the delimiter of the
119 pattern need not always be a slash, and no slash is used when writing
120 modifiers. Whitespace may appear between the final pattern delimiter
121 and the first modifier, and between the modifiers themselves.
123 The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
124 PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when pcre_com-
125 pile() is called. These four modifier letters have the same effect as
126 they do in Perl. For example:
130 The following table shows additional modifiers for setting PCRE options
131 that do not correspond to anything in Perl:
135 /E PCRE_DOLLAR_ENDONLY
137 /N PCRE_NO_AUTO_CAPTURE
141 Searching for all possible matches within each subject string can be
142 requested by the /g or /G modifier. After finding a match, PCRE is
143 called again to search the remainder of the subject string. The differ-
144 ence between /g and /G is that the former uses the startoffset argument
145 to pcre_exec() to start searching at a new point within the entire
146 string (which is in effect what Perl does), whereas the latter passes
147 over a shortened substring. This makes a difference to the matching
148 process if the pattern begins with a lookbehind assertion (including \b
151 If any call to pcre_exec() in a /g or /G sequence matches an empty
152 string, the next call is done with the PCRE_NOTEMPTY and PCRE_ANCHORED
153 flags set in order to search for another, non-empty, match at the same
154 point. If this second match fails, the start offset is advanced by
155 one, and the normal match is retried. This imitates the way Perl han-
156 dles such cases when using the /g modifier or the split() function.
158 There are yet more modifiers for controlling the way pcretest operates.
160 The /+ modifier requests that as well as outputting the substring that
161 matched the entire pattern, pcretest should in addition output the
162 remainder of the subject string. This is useful for tests where the
163 subject contains multiple copies of the same substring.
165 The /L modifier must be followed directly by the name of a locale, for
170 For this reason, it must be the last modifier. The given locale is set,
171 pcre_maketables() is called to build a set of character tables for the
172 locale, and this is then passed to pcre_compile() when compiling the
173 regular expression. Without an /L modifier, NULL is passed as the
174 tables pointer; that is, /L applies only to the expression on which it
177 The /I modifier requests that pcretest output information about the
178 compiled pattern (whether it is anchored, has a fixed first character,
179 and so on). It does this by calling pcre_fullinfo() after compiling a
180 pattern. If the pattern is studied, the results of that are also out-
183 The /D modifier is a PCRE debugging feature, which also assumes /I. It
184 causes the internal form of compiled regular expressions to be output
185 after compilation. If the pattern was studied, the information returned
188 The /F modifier causes pcretest to flip the byte order of the fields in
189 the compiled pattern that contain 2-byte and 4-byte numbers. This
190 facility is for testing the feature in PCRE that allows it to execute
191 patterns that were compiled on a host with a different endianness. This
192 feature is not available when the POSIX interface to PCRE is being
193 used, that is, when the /P pattern modifier is specified. See also the
194 section about saving and reloading compiled patterns below.
196 The /S modifier causes pcre_study() to be called after the expression
197 has been compiled, and the results used when the expression is matched.
199 The /M modifier causes the size of memory block used to hold the com-
200 piled pattern to be output.
202 The /P modifier causes pcretest to call PCRE via the POSIX wrapper API
203 rather than its native API. When this is done, all other modifiers
204 except /i, /m, and /+ are ignored. REG_ICASE is set if /i is present,
205 and REG_NEWLINE is set if /m is present. The wrapper functions force
206 PCRE_DOLLAR_ENDONLY always, and PCRE_DOTALL unless REG_NEWLINE is set.
208 The /8 modifier causes pcretest to call PCRE with the PCRE_UTF8 option
209 set. This turns on support for UTF-8 character handling in PCRE, pro-
210 vided that it was compiled with this support enabled. This modifier
211 also causes any non-printing characters in output strings to be printed
212 using the \x{hh...} notation if they are valid UTF-8 sequences.
214 If the /? modifier is used with /8, it causes pcretest to call
215 pcre_compile() with the PCRE_NO_UTF8_CHECK option, to suppress the
216 checking of the string for UTF-8 validity.
221 Before each data line is passed to pcre_exec(), leading and trailing
222 whitespace is removed, and it is then scanned for \ escapes. Some of
223 these are pretty esoteric features, intended for checking out some of
224 the more complicated features of PCRE. If you are just testing "ordi-
225 nary" regular expressions, you probably don't need any of these. The
226 following escapes are recognized:
236 \nnn octal character (up to 3 octal digits)
237 \xhh hexadecimal character (up to 2 hex digits)
238 \x{hh...} hexadecimal character, any number of digits
240 \A pass the PCRE_ANCHORED option to pcre_exec()
241 \B pass the PCRE_NOTBOL option to pcre_exec()
242 \Cdd call pcre_copy_substring() for substring dd
243 after a successful match (number less than 32)
244 \Cname call pcre_copy_named_substring() for substring
245 "name" after a successful match (name termin-
246 ated by next non alphanumeric character)
247 \C+ show the current captured substrings at callout
249 \C- do not supply a callout function
250 \C!n return 1 instead of 0 when callout number n is
252 \C!n!m return 1 instead of 0 when callout number n is
253 reached for the nth time
254 \C*n pass the number n (may be negative) as callout
255 data; this is used as the callout return value
256 \D use the pcre_dfa_exec() match function
257 \F only shortest match for pcre_dfa_exec()
258 \Gdd call pcre_get_substring() for substring dd
259 after a successful match (number less than 32)
260 \Gname call pcre_get_named_substring() for substring
261 "name" after a successful match (name termin-
262 ated by next non-alphanumeric character)
263 \L call pcre_get_substringlist() after a
265 \M discover the minimum MATCH_LIMIT setting
266 \N pass the PCRE_NOTEMPTY option to pcre_exec()
267 \Odd set the size of the output vector passed to
268 pcre_exec() to dd (any number of digits)
269 \P pass the PCRE_PARTIAL option to pcre_exec()
271 \R pass the PCRE_DFA_RESTART option to pcre_dfa_exec()
272 \S output details of memory get/free calls during matching
273 \Z pass the PCRE_NOTEOL option to pcre_exec()
274 \? pass the PCRE_NO_UTF8_CHECK option to
276 \>dd start the match at offset dd (any number of digits);
277 this sets the startoffset argument for pcre_exec()
279 A backslash followed by anything else just escapes the anything else.
280 If the very last character is a backslash, it is ignored. This gives a
281 way of passing an empty line as data, since a real empty line termi-
282 nates the data input.
284 If \M is present, pcretest calls pcre_exec() several times, with dif-
285 ferent values in the match_limit field of the pcre_extra data struc-
286 ture, until it finds the minimum number that is needed for pcre_exec()
287 to complete. This number is a measure of the amount of recursion and
288 backtracking that takes place, and checking it out can be instructive.
289 For most simple matches, the number is quite small, but for patterns
290 with very large numbers of matching possibilities, it can become large
291 very quickly with increasing length of subject string.
293 When \O is used, the value specified may be higher or lower than the
294 size set by the -O command line option (or defaulted to 45); \O applies
295 only to the call of pcre_exec() for the line in which it appears.
297 If the /P modifier was present on the pattern, causing the POSIX wrap-
298 per API to be used, only \B and \Z have any effect, causing REG_NOTBOL
299 and REG_NOTEOL to be passed to regexec() respectively.
301 The use of \x{hh...} to represent UTF-8 characters is not dependent on
302 the use of the /8 modifier on the pattern. It is recognized always.
303 There may be any number of hexadecimal digits inside the braces. The
304 result is from one to six bytes, encoded according to the UTF-8 rules.
307 THE ALTERNATIVE MATCHING FUNCTION
309 By default, pcretest uses the standard PCRE matching function,
310 pcre_exec() to match each data line. From release 6.0, PCRE supports an
311 alternative matching function, pcre_dfa_test(), which operates in a
312 different way, and has some restrictions. The differences between the
313 two functions are described in the pcrematching documentation.
315 If a data line contains the \D escape sequence, or if the command line
316 contains the -dfa option, the alternative matching function is called.
317 This function finds all possible matches at a given point. If, however,
318 the \F escape sequence is present in the data line, it stops after the
319 first match is found. This is always the shortest possible match.
322 DEFAULT OUTPUT FROM PCRETEST
324 This section describes the output when the normal matching function,
325 pcre_exec(), is being used.
327 When a match succeeds, pcretest outputs the list of captured substrings
328 that pcre_exec() returns, starting with number 0 for the string that
329 matched the whole pattern. Otherwise, it outputs "No match" or "Partial
330 match" when pcre_exec() returns PCRE_ERROR_NOMATCH or PCRE_ERROR_PAR-
331 TIAL, respectively, and otherwise the PCRE negative error number. Here
332 is an example of an interactive pcretest run.
335 PCRE version 5.00 07-Sep-2004
344 If the strings contain any non-printing characters, they are output as
345 \0x escapes, or as \x{...} escapes if the /8 modifier was present on
346 the pattern. If the pattern has the /+ modifier, the output for sub-
347 string 0 is followed by the the rest of the subject string, identified
355 If the pattern has the /g or /G modifier, the results of successive
356 matching attempts are output in sequence, like this:
367 "No match" is output only if the first match attempt fails.
369 If any of the sequences \C, \G, or \L are present in a data line that
370 is successfully matched, the substrings extracted by the convenience
371 functions are output with C, G, or L after the string number instead of
372 a colon. This is in addition to the normal full list. The string length
373 (that is, the return from the extraction function) is given in paren-
374 theses after each string for \C and \G.
376 Note that while patterns can be continued over several lines (a plain
377 ">" prompt is used for continuations), data lines may not. However new-
378 lines can be included in data by means of the \n escape.
381 OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
383 When the alternative matching function, pcre_dfa_exec(), is used (by
384 means of the \D escape sequence or the -dfa command line option), the
385 output consists of a list of all the matches that start at the first
386 point in the subject where there is at least one match. For example:
388 re> /(tang|tangerine|tan)/
389 data> yellow tangerine\D
394 (Using the normal matching function on this data finds only "tang".)
395 The longest matching string is always given first (and numbered zero).
397 If /gP is present on the pattern, the search for further matches
398 resumes at the end of the longest match. For example:
400 re> /(tang|tangerine|tan)/g
401 data> yellow tangerine and tangy sultana\D
409 Since the matching function does not support substring capture, the
410 escape sequences that are concerned with captured substrings are not
414 RESTARTING AFTER A PARTIAL MATCH
416 When the alternative matching function has given the PCRE_ERROR_PARTIAL
417 return, indicating that the subject partially matched the pattern, you
418 can restart the match with additional subject data by means of the \R
419 escape sequence. For example:
421 re> /^?(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)$/
427 For further information about partial matching, see the pcrepartial
433 If the pattern contains any callout requests, pcretest's callout func-
434 tion is called during matching. This works with both matching func-
435 tions. By default, the called function displays the callout number, the
436 start and current positions in the text at the callout time, and the
437 next pattern item to be tested. For example, the output
442 indicates that callout number 0 occurred for a match attempt starting
443 at the fourth character of the subject string, when the pointer was at
444 the seventh character of the data, and when the next pattern item was
445 \d. Just one circumflex is output if the start and current positions
448 Callouts numbered 255 are assumed to be automatic callouts, inserted as
449 a result of the /C pattern modifier. In this case, instead of showing
450 the callout number, the offset in the pattern, preceded by a plus, is
462 The callout function in pcretest returns zero (carry on matching) by
463 default, but you can use a \C item in a data line (as described above)
466 Inserting callouts can be helpful when using pcretest to check compli-
467 cated regular expressions. For further information about callouts, see
468 the pcrecallout documentation.
471 SAVING AND RELOADING COMPILED PATTERNS
473 The facilities described in this section are not available when the
474 POSIX inteface to PCRE is being used, that is, when the /P pattern mod-
477 When the POSIX interface is not in use, you can cause pcretest to write
478 a compiled pattern to a file, by following the modifiers with > and a
479 file name. For example:
481 /pattern/im >/some/file
483 See the pcreprecompile documentation for a discussion about saving and
484 re-using compiled patterns.
486 The data that is written is binary. The first eight bytes are the
487 length of the compiled pattern data followed by the length of the
488 optional study data, each written as four bytes in big-endian order
489 (most significant byte first). If there is no study data (either the
490 pattern was not studied, or studying did not return any data), the sec-
491 ond length is zero. The lengths are followed by an exact copy of the
492 compiled pattern. If there is additional study data, this follows imme-
493 diately after the compiled pattern. After writing the file, pcretest
494 expects to read a new pattern.
496 A saved pattern can be reloaded into pcretest by specifing < and a file
497 name instead of a pattern. The name of the file must not contain a <
498 character, as otherwise pcretest will interpret the line as a pattern
499 delimited by < characters. For example:
502 Compiled regex loaded from /some/file
505 When the pattern has been loaded, pcretest proceeds to read data lines
508 You can copy a file written by pcretest to a different host and reload
509 it there, even if the new host has opposite endianness to the one on
510 which the pattern was compiled. For example, you can compile on an i86
511 machine and run on a SPARC machine.
513 File names for saving and reloading can be absolute or relative, but
514 note that the shell facility of expanding a file name that starts with
515 a tilde (~) is not available.
517 The ability to save and reload files in pcretest is intended for test-
518 ing and experimentation. It is not intended for production use because
519 only a single pattern can be written to a file. Furthermore, there is
520 no facility for supplying custom character tables for use with a
521 reloaded pattern. If the original pattern was compiled with custom
522 tables, an attempt to match a subject string using a reloaded pattern
523 is likely to cause pcretest to crash. Finally, if you attempt to load
524 a file that is not in the correct format, the result is undefined.
530 University Computing Service,
531 Cambridge CB2 3QG, England.
533 Last updated: 28 February 2005
534 Copyright (c) 1997-2005 University of Cambridge.