Start
[exim.git] / doc / doc-txt / pcretest.txt
CommitLineData
495ae4b0
PH
1This file contains the PCRE man page that described the pcretest program. Note
2that not all of the features of PCRE are available in the limited version that
3is built with Exim.
4-------------------------------------------------------------------------------
5
6PCRETEST(1) PCRETEST(1)
7
8
9
10NAME
11 pcretest - a program for testing Perl-compatible regular expressions.
12
13SYNOPSIS
14
15 pcretest [-C] [-d] [-i] [-m] [-o osize] [-p] [-t] [source]
16 [destination]
17
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.
24
25
26OPTIONS
27
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.
31
32 -d Behave as if each regex had the /D (debug) modifier; the
33 internal form is output after compilation.
34
35 -i Behave as if each regex had the /I modifier; information
36 about the compiled pattern is given after compilation.
37
38 -m Output the size of each compiled pattern after it has been
39 compiled. This is equivalent to adding /M to each regular
40 expression. For compatibility with earlier versions of
41 pcretest, -s is a synonym for -m.
42
43 -o osize Set the number of elements in the output vector that is used
44 when calling pcre_exec() to be osize. The default value is
45 45, which is enough for 14 capturing subexpressions. The vec-
46 tor size can be changed for individual matching calls by
47 including \O in the data line (see below).
48
49 -p Behave as if each regex has /P modifier; the POSIX wrapper
50 API is used to call PCRE. None of the other options has any
51 effect when -p is set.
52
53 -t Run each compile, study, and match many times with a timer,
54 and output resulting time per compile or match (in millisec-
55 onds). Do not set -m with -t, because you will then get the
56 size output a zillion times, and the timing will be dis-
57 torted.
58
59
60DESCRIPTION
61
62 If pcretest is given two filename arguments, it reads from the first
63 and writes to the second. If it is given only one filename argument, it
64 reads from that file and writes to stdout. Otherwise, it reads from
65 stdin and writes to stdout, and prompts for each line of input, using
66 "re>" to prompt for regular expressions, and "data>" to prompt for data
67 lines.
68
69 The program handles any number of sets of input on a single input file.
70 Each set starts with a regular expression, and continues with any num-
71 ber of data lines to be matched against the pattern.
72
73 Each data line is matched separately and independently. If you want to
74 do multiple-line matches, you have to use the \n escape sequence in a
75 single line of input to encode the newline characters. The maximum
76 length of data line is 30,000 characters.
77
78 An empty line signals the end of the data lines, at which point a new
79 regular expression is read. The regular expressions are given enclosed
80 in any non-alphanumeric delimiters other than backslash, for example
81
82 /(a|bc)x+yz/
83
84 White space before the initial delimiter is ignored. A regular expres-
85 sion may be continued over several input lines, in which case the new-
86 line characters are included within it. It is possible to include the
87 delimiter within the pattern by escaping it, for example
88
89 /abc\/def/
90
91 If you do so, the escape and the delimiter form part of the pattern,
92 but since delimiters are always non-alphanumeric, this does not affect
93 its interpretation. If the terminating delimiter is immediately fol-
94 lowed by a backslash, for example,
95
96 /abc/\
97
98 then a backslash is added to the end of the pattern. This is done to
99 provide a way of testing the error condition that arises if a pattern
100 finishes with a backslash, because
101
102 /abc\/
103
104 is interpreted as the first line of a pattern that starts with "abc/",
105 causing pcretest to read the next line as a continuation of the regular
106 expression.
107
108
109PATTERN MODIFIERS
110
111 A pattern may be followed by any number of modifiers, which are mostly
112 single characters. Following Perl usage, these are referred to below
113 as, for example, "the /i modifier", even though the delimiter of the
114 pattern need not always be a slash, and no slash is used when writing
115 modifiers. Whitespace may appear between the final pattern delimiter
116 and the first modifier, and between the modifiers themselves.
117
118 The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
119 PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when pcre_com-
120 pile() is called. These four modifier letters have the same effect as
121 they do in Perl. For example:
122
123 /caseless/i
124
125 The following table shows additional modifiers for setting PCRE options
126 that do not correspond to anything in Perl:
127
128 /A PCRE_ANCHORED
129 /C PCRE_AUTO_CALLOUT
130 /E PCRE_DOLLAR_ENDONLY
131 /N PCRE_NO_AUTO_CAPTURE
132 /U PCRE_UNGREEDY
133 /X PCRE_EXTRA
134
135 Searching for all possible matches within each subject string can be
136 requested by the /g or /G modifier. After finding a match, PCRE is
137 called again to search the remainder of the subject string. The differ-
138 ence between /g and /G is that the former uses the startoffset argument
139 to pcre_exec() to start searching at a new point within the entire
140 string (which is in effect what Perl does), whereas the latter passes
141 over a shortened substring. This makes a difference to the matching
142 process if the pattern begins with a lookbehind assertion (including \b
143 or \B).
144
145 If any call to pcre_exec() in a /g or /G sequence matches an empty
146 string, the next call is done with the PCRE_NOTEMPTY and PCRE_ANCHORED
147 flags set in order to search for another, non-empty, match at the same
148 point. If this second match fails, the start offset is advanced by
149 one, and the normal match is retried. This imitates the way Perl han-
150 dles such cases when using the /g modifier or the split() function.
151
152 There are yet more modifiers for controlling the way pcretest operates.
153
154 The /+ modifier requests that as well as outputting the substring that
155 matched the entire pattern, pcretest should in addition output the
156 remainder of the subject string. This is useful for tests where the
157 subject contains multiple copies of the same substring.
158
159 The /L modifier must be followed directly by the name of a locale, for
160 example,
161
162 /pattern/Lfr_FR
163
164 For this reason, it must be the last modifier. The given locale is set,
165 pcre_maketables() is called to build a set of character tables for the
166 locale, and this is then passed to pcre_compile() when compiling the
167 regular expression. Without an /L modifier, NULL is passed as the
168 tables pointer; that is, /L applies only to the expression on which it
169 appears.
170
171 The /I modifier requests that pcretest output information about the
172 compiled pattern (whether it is anchored, has a fixed first character,
173 and so on). It does this by calling pcre_fullinfo() after compiling a
174 pattern. If the pattern is studied, the results of that are also out-
175 put.
176
177 The /D modifier is a PCRE debugging feature, which also assumes /I. It
178 causes the internal form of compiled regular expressions to be output
179 after compilation. If the pattern was studied, the information returned
180 is also output.
181
182 The /F modifier causes pcretest to flip the byte order of the fields in
183 the compiled pattern that contain 2-byte and 4-byte numbers. This
184 facility is for testing the feature in PCRE that allows it to execute
185 patterns that were compiled on a host with a different endianness. This
186 feature is not available when the POSIX interface to PCRE is being
187 used, that is, when the /P pattern modifier is specified. See also the
188 section about saving and reloading compiled patterns below.
189
190 The /S modifier causes pcre_study() to be called after the expression
191 has been compiled, and the results used when the expression is matched.
192
193 The /M modifier causes the size of memory block used to hold the com-
194 piled pattern to be output.
195
196 The /P modifier causes pcretest to call PCRE via the POSIX wrapper API
197 rather than its native API. When this is done, all other modifiers
198 except /i, /m, and /+ are ignored. REG_ICASE is set if /i is present,
199 and REG_NEWLINE is set if /m is present. The wrapper functions force
200 PCRE_DOLLAR_ENDONLY always, and PCRE_DOTALL unless REG_NEWLINE is set.
201
202 The /8 modifier causes pcretest to call PCRE with the PCRE_UTF8 option
203 set. This turns on support for UTF-8 character handling in PCRE, pro-
204 vided that it was compiled with this support enabled. This modifier
205 also causes any non-printing characters in output strings to be printed
206 using the \x{hh...} notation if they are valid UTF-8 sequences.
207
208 If the /? modifier is used with /8, it causes pcretest to call
209 pcre_compile() with the PCRE_NO_UTF8_CHECK option, to suppress the
210 checking of the string for UTF-8 validity.
211
212
213DATA LINES
214
215 Before each data line is passed to pcre_exec(), leading and trailing
216 whitespace is removed, and it is then scanned for \ escapes. Some of
217 these are pretty esoteric features, intended for checking out some of
218 the more complicated features of PCRE. If you are just testing "ordi-
219 nary" regular expressions, you probably don't need any of these. The
220 following escapes are recognized:
221
222 \a alarm (= BEL)
223 \b backspace
224 \e escape
225 \f formfeed
226 \n newline
227 \r carriage return
228 \t tab
229 \v vertical tab
230 \nnn octal character (up to 3 octal digits)
231 \xhh hexadecimal character (up to 2 hex digits)
232 \x{hh...} hexadecimal character, any number of digits
233 in UTF-8 mode
234 \A pass the PCRE_ANCHORED option to pcre_exec()
235 \B pass the PCRE_NOTBOL option to pcre_exec()
236 \Cdd call pcre_copy_substring() for substring dd
237 after a successful match (number less than 32)
238 \Cname call pcre_copy_named_substring() for substring
239 "name" after a successful match (name termin-
240 ated by next non alphanumeric character)
241 \C+ show the current captured substrings at callout
242 time
243 \C- do not supply a callout function
244 \C!n return 1 instead of 0 when callout number n is
245 reached
246 \C!n!m return 1 instead of 0 when callout number n is
247 reached for the nth time
248 \C*n pass the number n (may be negative) as callout
249 data; this is used as the callout return value
250 \Gdd call pcre_get_substring() for substring dd
251 after a successful match (number less than 32)
252 \Gname call pcre_get_named_substring() for substring
253 "name" after a successful match (name termin-
254 ated by next non-alphanumeric character)
255 \L call pcre_get_substringlist() after a
256 successful match
257 \M discover the minimum MATCH_LIMIT setting
258 \N pass the PCRE_NOTEMPTY option to pcre_exec()
259 \Odd set the size of the output vector passed to
260 pcre_exec() to dd (any number of digits)
261 \P pass the PCRE_PARTIAL option to pcre_exec()
262 \S output details of memory get/free calls during matching
263 \Z pass the PCRE_NOTEOL option to pcre_exec()
264 \? pass the PCRE_NO_UTF8_CHECK option to
265 pcre_exec()
266 \>dd start the match at offset dd (any number of digits);
267 this sets the startoffset argument for pcre_exec()
268
269 A backslash followed by anything else just escapes the anything else.
270 If the very last character is a backslash, it is ignored. This gives a
271 way of passing an empty line as data, since a real empty line termi-
272 nates the data input.
273
274 If \M is present, pcretest calls pcre_exec() several times, with dif-
275 ferent values in the match_limit field of the pcre_extra data struc-
276 ture, until it finds the minimum number that is needed for pcre_exec()
277 to complete. This number is a measure of the amount of recursion and
278 backtracking that takes place, and checking it out can be instructive.
279 For most simple matches, the number is quite small, but for patterns
280 with very large numbers of matching possibilities, it can become large
281 very quickly with increasing length of subject string.
282
283 When \O is used, the value specified may be higher or lower than the
284 size set by the -O command line option (or defaulted to 45); \O applies
285 only to the call of pcre_exec() for the line in which it appears.
286
287 If the /P modifier was present on the pattern, causing the POSIX wrap-
288 per API to be used, only \B and \Z have any effect, causing REG_NOTBOL
289 and REG_NOTEOL to be passed to regexec() respectively.
290
291 The use of \x{hh...} to represent UTF-8 characters is not dependent on
292 the use of the /8 modifier on the pattern. It is recognized always.
293 There may be any number of hexadecimal digits inside the braces. The
294 result is from one to six bytes, encoded according to the UTF-8 rules.
295
296
297OUTPUT FROM PCRETEST
298
299 When a match succeeds, pcretest outputs the list of captured substrings
300 that pcre_exec() returns, starting with number 0 for the string that
301 matched the whole pattern. Otherwise, it outputs "No match" or "Partial
302 match" when pcre_exec() returns PCRE_ERROR_NOMATCH or PCRE_ERROR_PAR-
303 TIAL, respectively, and otherwise the PCRE negative error number. Here
304 is an example of an interactive pcretest run.
305
306 $ pcretest
307 PCRE version 5.00 07-Sep-2004
308
309 re> /^abc(\d+)/
310 data> abc123
311 0: abc123
312 1: 123
313 data> xyz
314 No match
315
316 If the strings contain any non-printing characters, they are output as
317 \0x escapes, or as \x{...} escapes if the /8 modifier was present on
318 the pattern. If the pattern has the /+ modifier, the output for sub-
319 string 0 is followed by the the rest of the subject string, identified
320 by "0+" like this:
321
322 re> /cat/+
323 data> cataract
324 0: cat
325 0+ aract
326
327 If the pattern has the /g or /G modifier, the results of successive
328 matching attempts are output in sequence, like this:
329
330 re> /\Bi(\w\w)/g
331 data> Mississippi
332 0: iss
333 1: ss
334 0: iss
335 1: ss
336 0: ipp
337 1: pp
338
339 "No match" is output only if the first match attempt fails.
340
341 If any of the sequences \C, \G, or \L are present in a data line that
342 is successfully matched, the substrings extracted by the convenience
343 functions are output with C, G, or L after the string number instead of
344 a colon. This is in addition to the normal full list. The string length
345 (that is, the return from the extraction function) is given in paren-
346 theses after each string for \C and \G.
347
348 Note that while patterns can be continued over several lines (a plain
349 ">" prompt is used for continuations), data lines may not. However new-
350 lines can be included in data by means of the \n escape.
351
352
353CALLOUTS
354
355 If the pattern contains any callout requests, pcretest's callout func-
356 tion is called during matching. By default, it displays the callout
357 number, the start and current positions in the text at the callout
358 time, and the next pattern item to be tested. For example, the output
359
360 --->pqrabcdef
361 0 ^ ^ \d
362
363 indicates that callout number 0 occurred for a match attempt starting
364 at the fourth character of the subject string, when the pointer was at
365 the seventh character of the data, and when the next pattern item was
366 \d. Just one circumflex is output if the start and current positions
367 are the same.
368
369 Callouts numbered 255 are assumed to be automatic callouts, inserted as
370 a result of the /C pattern modifier. In this case, instead of showing
371 the callout number, the offset in the pattern, preceded by a plus, is
372 output. For example:
373
374 re> /\d?[A-E]\*/C
375 data> E*
376 --->E*
377 +0 ^ \d?
378 +3 ^ [A-E]
379 +8 ^^ \*
380 +10 ^ ^
381 0: E*
382
383 The callout function in pcretest returns zero (carry on matching) by
384 default, but you can use an \C item in a data line (as described above)
385 to change this.
386
387 Inserting callouts can be helpful when using pcretest to check compli-
388 cated regular expressions. For further information about callouts, see
389 the pcrecallout documentation.
390
391
392SAVING AND RELOADING COMPILED PATTERNS
393
394 The facilities described in this section are not available when the
395 POSIX inteface to PCRE is being used, that is, when the /P pattern mod-
396 ifier is specified.
397
398 When the POSIX interface is not in use, you can cause pcretest to write
399 a compiled pattern to a file, by following the modifiers with > and a
400 file name. For example:
401
402 /pattern/im >/some/file
403
404 See the pcreprecompile documentation for a discussion about saving and
405 re-using compiled patterns.
406
407 The data that is written is binary. The first eight bytes are the
408 length of the compiled pattern data followed by the length of the
409 optional study data, each written as four bytes in big-endian order
410 (most significant byte first). If there is no study data (either the
411 pattern was not studied, or studying did not return any data), the sec-
412 ond length is zero. The lengths are followed by an exact copy of the
413 compiled pattern. If there is additional study data, this follows imme-
414 diately after the compiled pattern. After writing the file, pcretest
415 expects to read a new pattern.
416
417 A saved pattern can be reloaded into pcretest by specifing < and a file
418 name instead of a pattern. The name of the file must not contain a <
419 character, as otherwise pcretest will interpret the line as a pattern
420 delimited by < characters. For example:
421
422 re> </some/file
423 Compiled regex loaded from /some/file
424 No study data
425
426 When the pattern has been loaded, pcretest proceeds to read data lines
427 in the usual way.
428
429 You can copy a file written by pcretest to a different host and reload
430 it there, even if the new host has opposite endianness to the one on
431 which the pattern was compiled. For example, you can compile on an i86
432 machine and run on a SPARC machine.
433
434 File names for saving and reloading can be absolute or relative, but
435 note that the shell facility of expanding a file name that starts with
436 a tilde (~) is not available.
437
438 The ability to save and reload files in pcretest is intended for test-
439 ing and experimentation. It is not intended for production use because
440 only a single pattern can be written to a file. Furthermore, there is
441 no facility for supplying custom character tables for use with a
442 reloaded pattern. If the original pattern was compiled with custom
443 tables, an attempt to match a subject string using a reloaded pattern
444 is likely to cause pcretest to crash. Finally, if you attempt to load
445 a file that is not in the correct format, the result is undefined.
446
447
448AUTHOR
449
450 Philip Hazel <ph10@cam.ac.uk>
451 University Computing Service,
452 Cambridge CB2 3QG, England.
453
454Last updated: 10 September 2004
455Copyright (c) 1997-2004 University of Cambridge.