ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 87 - (show annotations)
Sat Feb 24 21:41:21 2007 UTC (8 years, 7 months ago) by nigel
File MIME type: text/plain
File size: 25220 byte(s)
Error occurred while calculating annotation data.
Load pcre-6.5 into code/trunk.
5 pcretest - a program for testing Perl-compatible regular expressions.
10 pcretest [-C] [-d] [-dfa] [-i] [-m] [-o osize] [-p] [-t] [source]
11 [destination]
13 pcretest was written as a test program for the PCRE regular expression
14 library itself, but it can also be used for experimenting with regular
15 expressions. This document describes the features of the test program;
16 for details of the regular expressions themselves, see the pcrepattern
17 documentation. For details of the PCRE library function calls and their
18 options, see the pcreapi documentation.
23 -C Output the version number of the PCRE library, and all avail-
24 able information about the optional features that are
25 included, and then exit.
27 -d Behave as if each regex has the /D (debug) modifier; the
28 internal form is output after compilation.
30 -dfa Behave as if each data line contains the \D escape sequence;
31 this causes the alternative matching function,
32 pcre_dfa_exec(), to be used instead of the standard
33 pcre_exec() function (more detail is given below).
35 -i Behave as if each regex has the /I modifier; information
36 about the compiled pattern is given after compilation.
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.
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).
49 -p Behave as if each regex has the /P modifier; the POSIX wrap-
50 per API is used to call PCRE. None of the other options has
51 any effect when -p is set.
53 -q Do not output the version number of pcretest at the start of
54 execution.
56 -t Run each compile, study, and match many times with a timer,
57 and output resulting time per compile or match (in millisec-
58 onds). Do not set -m with -t, because you will then get the
59 size output a zillion times, and the timing will be dis-
60 torted.
65 If pcretest is given two filename arguments, it reads from the first
66 and writes to the second. If it is given only one filename argument, it
67 reads from that file and writes to stdout. Otherwise, it reads from
68 stdin and writes to stdout, and prompts for each line of input, using
69 "re>" to prompt for regular expressions, and "data>" to prompt for data
70 lines.
72 The program handles any number of sets of input on a single input file.
73 Each set starts with a regular expression, and continues with any num-
74 ber of data lines to be matched against the pattern.
76 Each data line is matched separately and independently. If you want to
77 do multiple-line matches, you have to use the \n escape sequence in a
78 single line of input to encode the newline characters. The maximum
79 length of data line is 30,000 characters.
81 An empty line signals the end of the data lines, at which point a new
82 regular expression is read. The regular expressions are given enclosed
83 in any non-alphanumeric delimiters other than backslash, for example
85 /(a|bc)x+yz/
87 White space before the initial delimiter is ignored. A regular expres-
88 sion may be continued over several input lines, in which case the new-
89 line characters are included within it. It is possible to include the
90 delimiter within the pattern by escaping it, for example
92 /abc\/def/
94 If you do so, the escape and the delimiter form part of the pattern,
95 but since delimiters are always non-alphanumeric, this does not affect
96 its interpretation. If the terminating delimiter is immediately fol-
97 lowed by a backslash, for example,
99 /abc/\
101 then a backslash is added to the end of the pattern. This is done to
102 provide a way of testing the error condition that arises if a pattern
103 finishes with a backslash, because
105 /abc\/
107 is interpreted as the first line of a pattern that starts with "abc/",
108 causing pcretest to read the next line as a continuation of the regular
109 expression.
114 A pattern may be followed by any number of modifiers, which are mostly
115 single characters. Following Perl usage, these are referred to below
116 as, for example, "the /i modifier", even though the delimiter of the
117 pattern need not always be a slash, and no slash is used when writing
118 modifiers. Whitespace may appear between the final pattern delimiter
119 and the first modifier, and between the modifiers themselves.
121 The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
122 PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when pcre_com-
123 pile() is called. These four modifier letters have the same effect as
124 they do in Perl. For example:
126 /caseless/i
128 The following table shows additional modifiers for setting PCRE options
129 that do not correspond to anything in Perl:
139 Searching for all possible matches within each subject string can be
140 requested by the /g or /G modifier. After finding a match, PCRE is
141 called again to search the remainder of the subject string. The differ-
142 ence between /g and /G is that the former uses the startoffset argument
143 to pcre_exec() to start searching at a new point within the entire
144 string (which is in effect what Perl does), whereas the latter passes
145 over a shortened substring. This makes a difference to the matching
146 process if the pattern begins with a lookbehind assertion (including \b
147 or \B).
149 If any call to pcre_exec() in a /g or /G sequence matches an empty
150 string, the next call is done with the PCRE_NOTEMPTY and PCRE_ANCHORED
151 flags set in order to search for another, non-empty, match at the same
152 point. If this second match fails, the start offset is advanced by
153 one, and the normal match is retried. This imitates the way Perl han-
154 dles such cases when using the /g modifier or the split() function.
156 There are yet more modifiers for controlling the way pcretest operates.
158 The /+ modifier requests that as well as outputting the substring that
159 matched the entire pattern, pcretest should in addition output the
160 remainder of the subject string. This is useful for tests where the
161 subject contains multiple copies of the same substring.
163 The /L modifier must be followed directly by the name of a locale, for
164 example,
166 /pattern/Lfr_FR
168 For this reason, it must be the last modifier. The given locale is set,
169 pcre_maketables() is called to build a set of character tables for the
170 locale, and this is then passed to pcre_compile() when compiling the
171 regular expression. Without an /L modifier, NULL is passed as the
172 tables pointer; that is, /L applies only to the expression on which it
173 appears.
175 The /I modifier requests that pcretest output information about the
176 compiled pattern (whether it is anchored, has a fixed first character,
177 and so on). It does this by calling pcre_fullinfo() after compiling a
178 pattern. If the pattern is studied, the results of that are also out-
179 put.
181 The /D modifier is a PCRE debugging feature, which also assumes /I. It
182 causes the internal form of compiled regular expressions to be output
183 after compilation. If the pattern was studied, the information returned
184 is also output.
186 The /F modifier causes pcretest to flip the byte order of the fields in
187 the compiled pattern that contain 2-byte and 4-byte numbers. This
188 facility is for testing the feature in PCRE that allows it to execute
189 patterns that were compiled on a host with a different endianness. This
190 feature is not available when the POSIX interface to PCRE is being
191 used, that is, when the /P pattern modifier is specified. See also the
192 section about saving and reloading compiled patterns below.
194 The /S modifier causes pcre_study() to be called after the expression
195 has been compiled, and the results used when the expression is matched.
197 The /M modifier causes the size of memory block used to hold the com-
198 piled pattern to be output.
200 The /P modifier causes pcretest to call PCRE via the POSIX wrapper API
201 rather than its native API. When this is done, all other modifiers
202 except /i, /m, and /+ are ignored. REG_ICASE is set if /i is present,
203 and REG_NEWLINE is set if /m is present. The wrapper functions force
204 PCRE_DOLLAR_ENDONLY always, and PCRE_DOTALL unless REG_NEWLINE is set.
206 The /8 modifier causes pcretest to call PCRE with the PCRE_UTF8 option
207 set. This turns on support for UTF-8 character handling in PCRE, pro-
208 vided that it was compiled with this support enabled. This modifier
209 also causes any non-printing characters in output strings to be printed
210 using the \x{hh...} notation if they are valid UTF-8 sequences.
212 If the /? modifier is used with /8, it causes pcretest to call
213 pcre_compile() with the PCRE_NO_UTF8_CHECK option, to suppress the
214 checking of the string for UTF-8 validity.
219 Before each data line is passed to pcre_exec(), leading and trailing
220 whitespace is removed, and it is then scanned for \ escapes. Some of
221 these are pretty esoteric features, intended for checking out some of
222 the more complicated features of PCRE. If you are just testing "ordi-
223 nary" regular expressions, you probably don't need any of these. The
224 following escapes are recognized:
226 \a alarm (= BEL)
227 \b backspace
228 \e escape
229 \f formfeed
230 \n newline
231 \r carriage return
232 \t tab
233 \v vertical tab
234 \nnn octal character (up to 3 octal digits)
235 \xhh hexadecimal character (up to 2 hex digits)
236 \x{hh...} hexadecimal character, any number of digits
237 in UTF-8 mode
238 \A pass the PCRE_ANCHORED option to pcre_exec()
239 \B pass the PCRE_NOTBOL option to pcre_exec()
240 \Cdd call pcre_copy_substring() for substring dd
241 after a successful match (number less than 32)
242 \Cname call pcre_copy_named_substring() for substring
243 "name" after a successful match (name termin-
244 ated by next non alphanumeric character)
245 \C+ show the current captured substrings at callout
246 time
247 \C- do not supply a callout function
248 \C!n return 1 instead of 0 when callout number n is
249 reached
250 \C!n!m return 1 instead of 0 when callout number n is
251 reached for the nth time
252 \C*n pass the number n (may be negative) as callout
253 data; this is used as the callout return value
254 \D use the pcre_dfa_exec() match function
255 \F only shortest match for pcre_dfa_exec()
256 \Gdd call pcre_get_substring() for substring dd
257 after a successful match (number less than 32)
258 \Gname call pcre_get_named_substring() for substring
259 "name" after a successful match (name termin-
260 ated by next non-alphanumeric character)
261 \L call pcre_get_substringlist() after a
262 successful match
263 \M discover the minimum MATCH_LIMIT and
265 \N pass the PCRE_NOTEMPTY option to pcre_exec()
266 \Odd set the size of the output vector passed to
267 pcre_exec() to dd (any number of digits)
268 \P pass the PCRE_PARTIAL option to pcre_exec()
269 or pcre_dfa_exec()
270 \R pass the PCRE_DFA_RESTART option to pcre_dfa_exec()
271 \S output details of memory get/free calls during matching
272 \Z pass the PCRE_NOTEOL option to pcre_exec()
273 \? pass the PCRE_NO_UTF8_CHECK option to
274 pcre_exec()
275 \>dd start the match at offset dd (any number of digits);
276 this sets the startoffset argument for pcre_exec()
278 A backslash followed by anything else just escapes the anything else.
279 If the very last character is a backslash, it is ignored. This gives a
280 way of passing an empty line as data, since a real empty line termi-
281 nates the data input.
283 If \M is present, pcretest calls pcre_exec() several times, with dif-
284 ferent values in the match_limit and match_limit_recursion fields of
285 the pcre_extra data structure, until it finds the minimum numbers for
286 each parameter that allow pcre_exec() to complete. The match_limit num-
287 ber is a measure of the amount of backtracking that takes place, and
288 checking it out can be instructive. For most simple matches, the number
289 is quite small, but for patterns with very large numbers of matching
290 possibilities, it can become large very quickly with increasing length
291 of subject string. The match_limit_recursion number is a measure of how
292 much stack (or, if PCRE is compiled with NO_RECURSE, how much heap)
293 memory is needed to complete the match attempt.
295 When \O is used, the value specified may be higher or lower than the
296 size set by the -O command line option (or defaulted to 45); \O applies
297 only to the call of pcre_exec() for the line in which it appears.
299 If the /P modifier was present on the pattern, causing the POSIX wrap-
300 per API to be used, the only option-setting sequences that have any
301 effect are \B and \Z, causing REG_NOTBOL and REG_NOTEOL, respectively,
302 to be passed to regexec().
304 The use of \x{hh...} to represent UTF-8 characters is not dependent on
305 the use of the /8 modifier on the pattern. It is recognized always.
306 There may be any number of hexadecimal digits inside the braces. The
307 result is from one to six bytes, encoded according to the UTF-8 rules.
312 By default, pcretest uses the standard PCRE matching function,
313 pcre_exec() to match each data line. From release 6.0, PCRE supports an
314 alternative matching function, pcre_dfa_test(), which operates in a
315 different way, and has some restrictions. The differences between the
316 two functions are described in the pcrematching documentation.
318 If a data line contains the \D escape sequence, or if the command line
319 contains the -dfa option, the alternative matching function is called.
320 This function finds all possible matches at a given point. If, however,
321 the \F escape sequence is present in the data line, it stops after the
322 first match is found. This is always the shortest possible match.
327 This section describes the output when the normal matching function,
328 pcre_exec(), is being used.
330 When a match succeeds, pcretest outputs the list of captured substrings
331 that pcre_exec() returns, starting with number 0 for the string that
332 matched the whole pattern. Otherwise, it outputs "No match" or "Partial
333 match" when pcre_exec() returns PCRE_ERROR_NOMATCH or PCRE_ERROR_PAR-
334 TIAL, respectively, and otherwise the PCRE negative error number. Here
335 is an example of an interactive pcretest run.
337 $ pcretest
338 PCRE version 5.00 07-Sep-2004
340 re> /^abc(\d+)/
341 data> abc123
342 0: abc123
343 1: 123
344 data> xyz
345 No match
347 If the strings contain any non-printing characters, they are output as
348 \0x escapes, or as \x{...} escapes if the /8 modifier was present on
349 the pattern. If the pattern has the /+ modifier, the output for sub-
350 string 0 is followed by the the rest of the subject string, identified
351 by "0+" like this:
353 re> /cat/+
354 data> cataract
355 0: cat
356 0+ aract
358 If the pattern has the /g or /G modifier, the results of successive
359 matching attempts are output in sequence, like this:
361 re> /\Bi(\w\w)/g
362 data> Mississippi
363 0: iss
364 1: ss
365 0: iss
366 1: ss
367 0: ipp
368 1: pp
370 "No match" is output only if the first match attempt fails.
372 If any of the sequences \C, \G, or \L are present in a data line that
373 is successfully matched, the substrings extracted by the convenience
374 functions are output with C, G, or L after the string number instead of
375 a colon. This is in addition to the normal full list. The string length
376 (that is, the return from the extraction function) is given in paren-
377 theses after each string for \C and \G.
379 Note that while patterns can be continued over several lines (a plain
380 ">" prompt is used for continuations), data lines may not. However new-
381 lines can be included in data by means of the \n escape.
386 When the alternative matching function, pcre_dfa_exec(), is used (by
387 means of the \D escape sequence or the -dfa command line option), the
388 output consists of a list of all the matches that start at the first
389 point in the subject where there is at least one match. For example:
391 re> /(tang|tangerine|tan)/
392 data> yellow tangerine\D
393 0: tangerine
394 1: tang
395 2: tan
397 (Using the normal matching function on this data finds only "tang".)
398 The longest matching string is always given first (and numbered zero).
400 If /gP is present on the pattern, the search for further matches
401 resumes at the end of the longest match. For example:
403 re> /(tang|tangerine|tan)/g
404 data> yellow tangerine and tangy sultana\D
405 0: tangerine
406 1: tang
407 2: tan
408 0: tang
409 1: tan
410 0: tan
412 Since the matching function does not support substring capture, the
413 escape sequences that are concerned with captured substrings are not
414 relevant.
419 When the alternative matching function has given the PCRE_ERROR_PARTIAL
420 return, indicating that the subject partially matched the pattern, you
421 can restart the match with additional subject data by means of the \R
422 escape sequence. For example:
424 re> /^?(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)$/
425 data> 23ja\P\D
426 Partial match: 23ja
427 data> n05\R\D
428 0: n05
430 For further information about partial matching, see the pcrepartial
431 documentation.
436 If the pattern contains any callout requests, pcretest's callout func-
437 tion is called during matching. This works with both matching func-
438 tions. By default, the called function displays the callout number, the
439 start and current positions in the text at the callout time, and the
440 next pattern item to be tested. For example, the output
442 --->pqrabcdef
443 0 ^ ^ \d
445 indicates that callout number 0 occurred for a match attempt starting
446 at the fourth character of the subject string, when the pointer was at
447 the seventh character of the data, and when the next pattern item was
448 \d. Just one circumflex is output if the start and current positions
449 are the same.
451 Callouts numbered 255 are assumed to be automatic callouts, inserted as
452 a result of the /C pattern modifier. In this case, instead of showing
453 the callout number, the offset in the pattern, preceded by a plus, is
454 output. For example:
456 re> /\d?[A-E]\*/C
457 data> E*
458 --->E*
459 +0 ^ \d?
460 +3 ^ [A-E]
461 +8 ^^ \*
462 +10 ^ ^
463 0: E*
465 The callout function in pcretest returns zero (carry on matching) by
466 default, but you can use a \C item in a data line (as described above)
467 to change this.
469 Inserting callouts can be helpful when using pcretest to check compli-
470 cated regular expressions. For further information about callouts, see
471 the pcrecallout documentation.
476 The facilities described in this section are not available when the
477 POSIX inteface to PCRE is being used, that is, when the /P pattern mod-
478 ifier is specified.
480 When the POSIX interface is not in use, you can cause pcretest to write
481 a compiled pattern to a file, by following the modifiers with > and a
482 file name. For example:
484 /pattern/im >/some/file
486 See the pcreprecompile documentation for a discussion about saving and
487 re-using compiled patterns.
489 The data that is written is binary. The first eight bytes are the
490 length of the compiled pattern data followed by the length of the
491 optional study data, each written as four bytes in big-endian order
492 (most significant byte first). If there is no study data (either the
493 pattern was not studied, or studying did not return any data), the sec-
494 ond length is zero. The lengths are followed by an exact copy of the
495 compiled pattern. If there is additional study data, this follows imme-
496 diately after the compiled pattern. After writing the file, pcretest
497 expects to read a new pattern.
499 A saved pattern can be reloaded into pcretest by specifing < and a file
500 name instead of a pattern. The name of the file must not contain a <
501 character, as otherwise pcretest will interpret the line as a pattern
502 delimited by < characters. For example:
504 re> </some/file
505 Compiled regex loaded from /some/file
506 No study data
508 When the pattern has been loaded, pcretest proceeds to read data lines
509 in the usual way.
511 You can copy a file written by pcretest to a different host and reload
512 it there, even if the new host has opposite endianness to the one on
513 which the pattern was compiled. For example, you can compile on an i86
514 machine and run on a SPARC machine.
516 File names for saving and reloading can be absolute or relative, but
517 note that the shell facility of expanding a file name that starts with
518 a tilde (~) is not available.
520 The ability to save and reload files in pcretest is intended for test-
521 ing and experimentation. It is not intended for production use because
522 only a single pattern can be written to a file. Furthermore, there is
523 no facility for supplying custom character tables for use with a
524 reloaded pattern. If the original pattern was compiled with custom
525 tables, an attempt to match a subject string using a reloaded pattern
526 is likely to cause pcretest to crash. Finally, if you attempt to load
527 a file that is not in the correct format, the result is undefined.
532 Philip Hazel
533 University Computing Service,
534 Cambridge CB2 3QG, England.
536 Last updated: 18 January 2006
537 Copyright (c) 1997-2006 University of Cambridge.

  ViewVC Help
Powered by ViewVC 1.1.5