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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 869 - (hide annotations) (download)
Sat Jan 14 11:16:23 2012 UTC (2 years, 11 months ago) by ph10
File MIME type: text/plain
File size: 43370 byte(s)
Bring HTML docs up to date.

1 nigel 73 PCRETEST(1) PCRETEST(1)
2 nigel 41
3    
4 nigel 73 NAME
5     pcretest - a program for testing Perl-compatible regular expressions.
6    
7 nigel 79
8 nigel 53 SYNOPSIS
9 nigel 41
10 ph10 654 pcretest [options] [input file [output file]]
11 nigel 75
12 nigel 73 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 nigel 75 documentation. For details of the PCRE library function calls and their
17 ph10 869 options, see the pcreapi and pcre16 documentation. The input for
18     pcretest is a sequence of regular expression patterns and strings to be
19     matched, as described below. The output shows the result of each match.
20     Options on the command line and the patterns control PCRE options and
21     exactly what is output.
22 nigel 41
23    
24 ph10 869 PCRE's 8-BIT and 16-BIT LIBRARIES
25    
26     From release 8.30, two separate PCRE libraries can be built. The origi-
27     nal one supports 8-bit character strings, whereas the newer 16-bit
28     library supports character strings encoded in 16-bit units. The
29     pcretest program can be used to test both libraries. However, it is
30     itself still an 8-bit program, reading 8-bit input and writing 8-bit
31     output. When testing the 16-bit library, the patterns and data strings
32     are converted to 16-bit format before being passed to the PCRE library
33     functions. Results are converted to 8-bit for output.
34    
35     References to functions and structures of the form pcre[16]_xx below
36     mean "pcre_xx when using the 8-bit library or pcre16_xx when using the
37     16-bit library".
38    
39    
40 ph10 654 COMMAND LINE OPTIONS
41 nigel 41
42 ph10 869 -16 If both the 8-bit and the 16-bit libraries have been built,
43     this option causes the 16-bit library to be used. If only the
44     16-bit library has been built, this is the default (so has no
45     effect). If only the 8-bit library has been built, this
46     option causes an error.
47    
48     -b Behave as if each pattern has the /B (show byte code) modi-
49 ph10 654 fier; the internal form is output after compilation.
50 nigel 93
51 nigel 73 -C Output the version number of the PCRE library, and all avail-
52 ph10 869 able information about the optional features that are
53     included, and then exit. All other options are ignored.
54 nigel 63
55 ph10 869 -C option Output information about a specific build-time option, then
56     exit. This functionality is intended for use in scripts such
57     as RunTest. The following options output the value indicated:
58    
59     linksize the internal link size (2, 3, or 4)
60     newline the default newline setting:
61     CR, LF, CRLF, ANYCRLF, or ANY
62    
63     The following options output 1 for true or zero for false:
64    
65     jit just-in-time support is available
66     pcre16 the 16-bit library was built
67     pcre8 the 8-bit library was built
68     ucp Unicode property support is available
69     utf UTF-8 and/or UTF-16 support is available
70    
71     -d Behave as if each pattern has the /D (debug) modifier; the
72     internal form and information about the compiled pattern is
73 nigel 93 output after compilation; -d is equivalent to -b -i.
74 nigel 41
75 ph10 869 -dfa Behave as if each data line contains the \D escape sequence;
76 nigel 77 this causes the alternative matching function,
77 ph10 869 pcre[16]_dfa_exec(), to be used instead of the standard
78     pcre[16]_exec() function (more detail is given below).
79 nigel 77
80 nigel 93 -help Output a brief summary these options and then exit.
81    
82 ph10 869 -i Behave as if each pattern has the /I modifier; information
83 nigel 73 about the compiled pattern is given after compilation.
84 nigel 41
85 ph10 869 -M Behave as if each data line contains the \M escape sequence;
86     this causes PCRE to discover the minimum MATCH_LIMIT and
87     MATCH_LIMIT_RECURSION settings by calling pcre[16]_exec()
88     repeatedly with different limits.
89 ph10 392
90 ph10 869 -m Output the size of each compiled pattern after it has been
91     compiled. This is equivalent to adding /M to each regular
92     expression. The size is given in bytes for both libraries.
93 nigel 41
94 ph10 869 -o osize Set the number of elements in the output vector that is used
95     when calling pcre[16]_exec() or pcre[16]_dfa_exec() to be
96     osize. The default value is 45, which is enough for 14 cap-
97     turing subexpressions for pcre[16]_exec() or 22 different
98     matches for pcre[16]_dfa_exec(). The vector size can be
99     changed for individual matching calls by including \O in the
100     data line (see below).
101 nigel 41
102 ph10 869 -p Behave as if each pattern has the /P modifier; the POSIX
103     wrapper API is used to call PCRE. None of the other options
104     has any effect when -p is set. This option can be used only
105     with the 8-bit library.
106 nigel 41
107 ph10 654 -q Do not output the version number of pcretest at the start of
108 nigel 87 execution.
109    
110 ph10 654 -S size On Unix-like systems, set the size of the run-time stack to
111 nigel 91 size megabytes.
112    
113 ph10 691 -s or -s+ Behave as if each pattern has the /S modifier; in other
114     words, force each pattern to be studied. If -s+ is used, the
115 ph10 869 PCRE_STUDY_JIT_COMPILE flag is passed to pcre[16]_study(),
116     causing just-in-time optimization to be set up if it is
117     available. If the /I or /D option is present on a pattern
118 ph10 691 (requesting output about the compiled pattern), information
119     about the result of studying is not included when studying is
120     caused only by -s and neither -i nor -d is present on the
121     command line. This behaviour means that the output from tests
122     that are run with and without -s should be identical, except
123     when options that output information about the actual running
124 ph10 869 of a match are set.
125 ph10 654
126 ph10 869 The -M, -t, and -tm options, which give information about
127     resources used, are likely to produce different output with
128     and without -s. Output may also differ if the /C option is
129     present on an individual pattern. This uses callouts to trace
130     the the matching process, and this may be different between
131     studied and non-studied patterns. If the pattern contains
132     (*MARK) items there may also be differences, for the same
133     reason. The -s command line option can be overridden for spe-
134     cific patterns that should never be studied (see the /S pat-
135     tern modifier below).
136    
137 ph10 654 -t Run each compile, study, and match many times with a timer,
138     and output resulting time per compile or match (in millisec-
139     onds). Do not set -m with -t, because you will then get the
140     size output a zillion times, and the timing will be dis-
141     torted. You can control the number of iterations that are
142     used for timing by following -t with a number (as a separate
143 nigel 93 item on the command line). For example, "-t 1000" would iter-
144     ate 1000 times. The default is to iterate 500000 times.
145 nigel 49
146 nigel 93 -tm This is like -t except that it times only the matching phase,
147     not the compile or study phases.
148 nigel 49
149 nigel 93
150 nigel 63 DESCRIPTION
151 nigel 41
152 ph10 654 If pcretest is given two filename arguments, it reads from the first
153 nigel 73 and writes to the second. If it is given only one filename argument, it
154 ph10 654 reads from that file and writes to stdout. Otherwise, it reads from
155     stdin and writes to stdout, and prompts for each line of input, using
156 nigel 73 "re>" to prompt for regular expressions, and "data>" to prompt for data
157     lines.
158 nigel 41
159 ph10 654 When pcretest is built, a configuration option can specify that it
160     should be linked with the libreadline library. When this is done, if
161 ph10 289 the input is from a terminal, it is read using the readline() function.
162 ph10 654 This provides line-editing and history facilities. The output from the
163 ph10 289 -help option states whether or not readline() will be used.
164    
165 nigel 73 The program handles any number of sets of input on a single input file.
166 ph10 654 Each set starts with a regular expression, and continues with any num-
167 nigel 73 ber of data lines to be matched against the pattern.
168 nigel 41
169 ph10 654 Each data line is matched separately and independently. If you want to
170 nigel 91 do multi-line matches, you have to use the \n escape sequence (or \r or
171 nigel 93 \r\n, etc., depending on the newline setting) in a single line of input
172 ph10 654 to encode the newline sequences. There is no limit on the length of
173     data lines; the input buffer is automatically extended if it is too
174 nigel 93 small.
175 nigel 63
176 ph10 654 An empty line signals the end of the data lines, at which point a new
177     regular expression is read. The regular expressions are given enclosed
178 nigel 91 in any non-alphanumeric delimiters other than backslash, for example:
179 nigel 63
180 nigel 73 /(a|bc)x+yz/
181 nigel 41
182 ph10 654 White space before the initial delimiter is ignored. A regular expres-
183     sion may be continued over several input lines, in which case the new-
184     line characters are included within it. It is possible to include the
185 nigel 73 delimiter within the pattern by escaping it, for example
186 nigel 41
187 nigel 73 /abc\/def/
188 nigel 41
189 ph10 654 If you do so, the escape and the delimiter form part of the pattern,
190     but since delimiters are always non-alphanumeric, this does not affect
191     its interpretation. If the terminating delimiter is immediately fol-
192 nigel 73 lowed by a backslash, for example,
193 nigel 41
194 nigel 73 /abc/\
195 nigel 41
196 ph10 654 then a backslash is added to the end of the pattern. This is done to
197     provide a way of testing the error condition that arises if a pattern
198 nigel 73 finishes with a backslash, because
199 nigel 49
200 nigel 73 /abc\/
201 nigel 49
202 ph10 654 is interpreted as the first line of a pattern that starts with "abc/",
203 nigel 73 causing pcretest to read the next line as a continuation of the regular
204     expression.
205 nigel 49
206 nigel 41
207 nigel 63 PATTERN MODIFIERS
208 nigel 41
209 ph10 654 A pattern may be followed by any number of modifiers, which are mostly
210     single characters. Following Perl usage, these are referred to below
211     as, for example, "the /i modifier", even though the delimiter of the
212     pattern need not always be a slash, and no slash is used when writing
213     modifiers. White space may appear between the final pattern delimiter
214 nigel 75 and the first modifier, and between the modifiers themselves.
215 nigel 41
216 nigel 75 The /i, /m, /s, and /x modifiers set the PCRE_CASELESS, PCRE_MULTILINE,
217 ph10 869 PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when pcre[16]_com-
218 ph10 654 pile() is called. These four modifier letters have the same effect as
219 nigel 75 they do in Perl. For example:
220    
221 nigel 73 /caseless/i
222 nigel 41
223 ph10 654 The following table shows additional modifiers for setting PCRE com-
224 ph10 535 pile-time options that do not correspond to anything in Perl:
225 nigel 41
226 ph10 869 /8 PCRE_UTF8 ) when using the 8-bit
227     /? PCRE_NO_UTF8_CHECK ) library
228    
229     /8 PCRE_UTF16 ) when using the 16-bit
230     /? PCRE_NO_UTF16_CHECK ) library
231    
232 ph10 231 /A PCRE_ANCHORED
233     /C PCRE_AUTO_CALLOUT
234     /E PCRE_DOLLAR_ENDONLY
235     /f PCRE_FIRSTLINE
236     /J PCRE_DUPNAMES
237     /N PCRE_NO_AUTO_CAPTURE
238     /U PCRE_UNGREEDY
239 ph10 518 /W PCRE_UCP
240 ph10 231 /X PCRE_EXTRA
241 ph10 579 /Y PCRE_NO_START_OPTIMIZE
242 ph10 345 /<JS> PCRE_JAVASCRIPT_COMPAT
243 ph10 231 /<cr> PCRE_NEWLINE_CR
244     /<lf> PCRE_NEWLINE_LF
245     /<crlf> PCRE_NEWLINE_CRLF
246     /<anycrlf> PCRE_NEWLINE_ANYCRLF
247     /<any> PCRE_NEWLINE_ANY
248     /<bsr_anycrlf> PCRE_BSR_ANYCRLF
249     /<bsr_unicode> PCRE_BSR_UNICODE
250 nigel 75
251 ph10 654 The modifiers that are enclosed in angle brackets are literal strings
252     as shown, including the angle brackets, but the letters within can be
253     in either case. This example sets multiline matching with CRLF as the
254     line ending sequence:
255 nigel 91
256 ph10 654 /^abc/m<CRLF>
257 nigel 93
258 ph10 869 As well as turning on the PCRE_UTF8/16 option, the /8 modifier causes
259     all non-printing characters in output strings to be printed using the
260     \x{hh...} notation. Otherwise, those less than 0x100 are output in hex
261     without the curly brackets.
262 nigel 93
263 ph10 869 Full details of the PCRE options are given in the pcreapi documenta-
264     tion.
265    
266 nigel 91 Finding all matches in a string
267    
268 ph10 869 Searching for all possible matches within each subject string can be
269     requested by the /g or /G modifier. After finding a match, PCRE is
270 nigel 73 called again to search the remainder of the subject string. The differ-
271     ence between /g and /G is that the former uses the startoffset argument
272 ph10 869 to pcre[16]_exec() to start searching at a new point within the entire
273     string (which is in effect what Perl does), whereas the latter passes
274     over a shortened substring. This makes a difference to the matching
275 nigel 73 process if the pattern begins with a lookbehind assertion (including \b
276     or \B).
277 nigel 49
278 ph10 869 If any call to pcre[16]_exec() in a /g or /G sequence matches an empty
279     string, the next call is done with the PCRE_NOTEMPTY_ATSTART and
280     PCRE_ANCHORED flags set in order to search for another, non-empty,
281     match at the same point. If this second match fails, the start offset
282     is advanced, and the normal match is retried. This imitates the way
283 ph10 567 Perl handles such cases when using the /g modifier or the split() func-
284 ph10 869 tion. Normally, the start offset is advanced by one character, but if
285     the newline convention recognizes CRLF as a newline, and the current
286 ph10 567 character is CR followed by LF, an advance of two is used.
287 nigel 49
288 nigel 91 Other modifiers
289    
290 nigel 75 There are yet more modifiers for controlling the way pcretest operates.
291 nigel 49
292 ph10 869 The /+ modifier requests that as well as outputting the substring that
293     matched the entire pattern, pcretest should in addition output the
294     remainder of the subject string. This is useful for tests where the
295     subject contains multiple copies of the same substring. If the + modi-
296     fier appears twice, the same action is taken for captured substrings.
297     In each case the remainder is output on the following line with a plus
298     character following the capture number. Note that this modifier must
299 ph10 691 not immediately follow the /S modifier because /S+ has another meaning.
300 nigel 41
301 ph10 869 The /= modifier requests that the values of all potential captured
302     parentheses be output after a match. By default, only those up to the
303     highest one actually used in the match are output (corresponding to the
304     return code from pcre[16]_exec()). Values in the offsets vector corre-
305     sponding to higher numbers should be set to -1, and these are output as
306     "<unset>". This modifier gives a way of checking that this is happen-
307     ing.
308 ph10 654
309 ph10 869 The /B modifier is a debugging feature. It requests that pcretest out-
310     put a representation of the compiled code after compilation. Normally
311     this information contains length and offset values; however, if /Z is
312     also present, this data is replaced by spaces. This is a special fea-
313     ture for use in the automatic test scripts; it ensures that the same
314 ph10 123 output is generated for different internal link sizes.
315 nigel 93
316 ph10 869 The /D modifier is a PCRE debugging feature, and is equivalent to /BI,
317 nigel 93 that is, both the /B and the /I modifiers.
318 nigel 41
319 ph10 869 The /F modifier causes pcretest to flip the byte order of the 2-byte
320     and 4-byte fields in the compiled pattern. This facility is for testing
321     the feature in PCRE that allows it to execute patterns that were com-
322     piled on a host with a different endianness. This feature is not avail-
323     able when the POSIX interface to PCRE is being used, that is, when the
324     /P pattern modifier is specified. See also the section about saving and
325     reloading compiled patterns below.
326 nigel 75
327 ph10 869 The /I modifier requests that pcretest output information about the
328     compiled pattern (whether it is anchored, has a fixed first character,
329     and so on). It does this by calling pcre[16]_fullinfo() after compiling
330     a pattern. If the pattern is studied, the results of that are also out-
331 ph10 512 put.
332 nigel 41
333 ph10 869 The /K modifier requests pcretest to show names from backtracking con-
334     trol verbs that are returned from calls to pcre[16]_exec(). It causes
335     pcretest to create a pcre[16]_extra block if one has not already been
336     created by a call to pcre[16]_study(), and to set the PCRE_EXTRA_MARK
337     flag and the mark field within it, every time that pcre[16]_exec() is
338     called. If the variable that the mark field points to is non-NULL for a
339     match, non-match, or partial match, pcretest prints the string to which
340     it points. For a match, this is shown on a line by itself, tagged with
341     "MK:". For a non-match it is added to the message.
342 ph10 512
343 ph10 869 The /L modifier must be followed directly by the name of a locale, for
344 ph10 512 example,
345    
346     /pattern/Lfr_FR
347    
348     For this reason, it must be the last modifier. The given locale is set,
349 ph10 869 pcre[16]_maketables() is called to build a set of character tables for
350     the locale, and this is then passed to pcre[16]_compile() when compil-
351     ing the regular expression. Without an /L (or /T) modifier, NULL is
352     passed as the tables pointer; that is, /L applies only to the expres-
353     sion on which it appears.
354 ph10 512
355 ph10 869 The /M modifier causes the size in bytes of the memory block used to
356     hold the compiled pattern to be output. This does not include the size
357     of the pcre[16] block; it is just the actual compiled data. If the pat-
358     tern is successfully studied with the PCRE_STUDY_JIT_COMPILE option,
359     the size of the JIT compiled code is also output.
360 nigel 41
361 ph10 869 If the /S modifier appears once, it causes pcre[16]_study() to be
362     called after the expression has been compiled, and the results used
363     when the expression is matched. If /S appears twice, it suppresses
364     studying, even if it was requested externally by the -s command line
365     option. This makes it possible to specify that certain patterns are
366     always studied, and others are never studied, independently of -s. This
367     feature is used in the test files in a few cases where the output is
368     different when the pattern is studied.
369 nigel 41
370 ph10 869 If the /S modifier is immediately followed by a + character, the call
371     to pcre[16]_study() is made with the PCRE_STUDY_JIT_COMPILE option,
372     requesting just-in-time optimization support if it is available. Note
373     that there is also a /+ modifier; it must not be given immediately
374     after /S because this will be misinterpreted. If JIT studying is suc-
375     cessful, it will automatically be used when pcre[16]_exec() is run,
376     except when incompatible run-time options are specified. These include
377     the partial matching options; a complete list is given in the pcrejit
378     documentation. See also the \J escape sequence below for a way of set-
379     ting the size of the JIT stack.
380 ph10 691
381 ph10 869 The /T modifier must be followed by a single digit. It causes a spe-
382     cific set of built-in character tables to be passed to pcre[16]_com-
383     pile(). It is used in the standard PCRE tests to check behaviour with
384     different character tables. The digit specifies the tables as follows:
385 ph10 545
386     0 the default ASCII tables, as distributed in
387     pcre_chartables.c.dist
388     1 a set of tables defining ISO 8859 characters
389    
390 ph10 869 In table 1, some characters whose codes are greater than 128 are iden-
391 ph10 545 tified as letters, digits, spaces, etc.
392    
393 ph10 535 Using the POSIX wrapper API
394 nigel 41
395 ph10 869 The /P modifier causes pcretest to call PCRE via the POSIX wrapper API
396     rather than its native API. This supports only the 8-bit library. When
397     /P is set, the following modifiers set options for the regcomp() func-
398     tion:
399 ph10 535
400     /i REG_ICASE
401     /m REG_NEWLINE
402     /N REG_NOSUB
403     /s REG_DOTALL )
404     /U REG_UNGREEDY ) These options are not part of
405     /W REG_UCP ) the POSIX standard
406     /8 REG_UTF8 )
407    
408 ph10 836 The /+ modifier works as described above. All other modifiers are
409 ph10 535 ignored.
410    
411    
412 nigel 53 DATA LINES
413 nigel 63
414 ph10 869 Before each data line is passed to pcre[16]_exec(), leading and trail-
415     ing white space is removed, and it is then scanned for \ escapes. Some
416     of these are pretty esoteric features, intended for checking out some
417     of the more complicated features of PCRE. If you are just testing
418     "ordinary" regular expressions, you probably don't need any of these.
419     The following escapes are recognized:
420 nigel 49
421 nigel 93 \a alarm (BEL, \x07)
422     \b backspace (\x08)
423     \e escape (\x27)
424 ph10 654 \f form feed (\x0c)
425 nigel 93 \n newline (\x0a)
426 nigel 91 \qdd set the PCRE_MATCH_LIMIT limit to dd
427     (any number of digits)
428 nigel 93 \r carriage return (\x0d)
429     \t tab (\x09)
430     \v vertical tab (\x0b)
431 ph10 869 \nnn octal character (up to 3 octal digits); always
432     a byte unless > 255 in UTF-8 or 16-bit mode
433 ph10 572 \xhh hexadecimal byte (up to 2 hex digits)
434 ph10 869 \x{hh...} hexadecimal character (any number of hex digits)
435     \A pass the PCRE_ANCHORED option to pcre[16]_exec()
436     or pcre[16]_dfa_exec()
437     \B pass the PCRE_NOTBOL option to pcre[16]_exec()
438     or pcre[16]_dfa_exec()
439     \Cdd call pcre[16]_copy_substring() for substring dd
440 nigel 75 after a successful match (number less than 32)
441 ph10 869 \Cname call pcre[16]_copy_named_substring() for substring
442 nigel 73 "name" after a successful match (name termin-
443     ated by next non alphanumeric character)
444     \C+ show the current captured substrings at callout
445     time
446     \C- do not supply a callout function
447     \C!n return 1 instead of 0 when callout number n is
448     reached
449     \C!n!m return 1 instead of 0 when callout number n is
450     reached for the nth time
451     \C*n pass the number n (may be negative) as callout
452 nigel 75 data; this is used as the callout return value
453 ph10 869 \D use the pcre[16]_dfa_exec() match function
454     \F only shortest match for pcre[16]_dfa_exec()
455     \Gdd call pcre[16]_get_substring() for substring dd
456 nigel 75 after a successful match (number less than 32)
457 ph10 869 \Gname call pcre[16]_get_named_substring() for substring
458 nigel 73 "name" after a successful match (name termin-
459     ated by next non-alphanumeric character)
460 ph10 691 \Jdd set up a JIT stack of dd kilobytes maximum (any
461     number of digits)
462 ph10 869 \L call pcre[16]_get_substringlist() after a
463 nigel 73 successful match
464 nigel 87 \M discover the minimum MATCH_LIMIT and
465     MATCH_LIMIT_RECURSION settings
466 ph10 869 \N pass the PCRE_NOTEMPTY option to pcre[16]_exec()
467     or pcre[16]_dfa_exec(); if used twice, pass the
468 ph10 453 PCRE_NOTEMPTY_ATSTART option
469 nigel 73 \Odd set the size of the output vector passed to
470 ph10 869 pcre[16]_exec() to dd (any number of digits)
471     \P pass the PCRE_PARTIAL_SOFT option to pcre[16]_exec()
472     or pcre[16]_dfa_exec(); if used twice, pass the
473 ph10 429 PCRE_PARTIAL_HARD option
474 nigel 91 \Qdd set the PCRE_MATCH_LIMIT_RECURSION limit to dd
475     (any number of digits)
476 ph10 869 \R pass the PCRE_DFA_RESTART option to pcre[16]_dfa_exec()
477 nigel 73 \S output details of memory get/free calls during matching
478 ph10 869 \Y pass the PCRE_NO_START_OPTIMIZE option to pcre[16]_exec()
479     or pcre[16]_dfa_exec()
480     \Z pass the PCRE_NOTEOL option to pcre[16]_exec()
481     or pcre[16]_dfa_exec()
482     \? pass the PCRE_NO_UTF[8|16]_CHECK option to
483     pcre[16]_exec() or pcre[16]_dfa_exec()
484 ph10 567 \>dd start the match at offset dd (optional "-"; then
485     any number of digits); this sets the startoffset
486 ph10 869 argument for pcre[16]_exec() or pcre[16]_dfa_exec()
487     \<cr> pass the PCRE_NEWLINE_CR option to pcre[16]_exec()
488     or pcre[16]_dfa_exec()
489     \<lf> pass the PCRE_NEWLINE_LF option to pcre[16]_exec()
490     or pcre[16]_dfa_exec()
491     \<crlf> pass the PCRE_NEWLINE_CRLF option to pcre[16]_exec()
492     or pcre[16]_dfa_exec()
493     \<anycrlf> pass the PCRE_NEWLINE_ANYCRLF option to pcre[16]_exec()
494     or pcre[16]_dfa_exec()
495     \<any> pass the PCRE_NEWLINE_ANY option to pcre[16]_exec()
496     or pcre[16]_dfa_exec()
497 nigel 71
498 ph10 869 The use of \x{hh...} is not dependent on the use of the /8 modifier on
499     the pattern. It is recognized always. There may be any number of hexa-
500     decimal digits inside the braces; invalid values provoke error mes-
501     sages.
502 ph10 572
503 ph10 869 Note that \xhh specifies one byte in UTF-8 mode; this makes it possible
504     to construct invalid UTF-8 sequences for testing purposes. On the other
505     hand, \x{hh} is interpreted as a UTF-8 character in UTF-8 mode, gener-
506     ating more than one byte if the value is greater than 127. When testing
507     the 8-bit library not in UTF-8 mode, \x{hh} generates one byte for val-
508     ues less than 256, and causes an error for greater values.
509    
510     In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it
511     possible to construct invalid UTF-16 sequences for testing purposes.
512    
513 ph10 836 The escapes that specify line ending sequences are literal strings,
514 nigel 93 exactly as shown. No more than one newline setting should be present in
515     any data line.
516 nigel 75
517 ph10 836 A backslash followed by anything else just escapes the anything else.
518     If the very last character is a backslash, it is ignored. This gives a
519     way of passing an empty line as data, since a real empty line termi-
520 nigel 93 nates the data input.
521    
522 ph10 836 The \J escape provides a way of setting the maximum stack size that is
523     used by the just-in-time optimization code. It is ignored if JIT opti-
524     mization is not being used. Providing a stack that is larger than the
525 ph10 691 default 32K is necessary only for very complicated patterns.
526 nigel 41
527 ph10 869 If \M is present, pcretest calls pcre[16]_exec() several times, with
528     different values in the match_limit and match_limit_recursion fields of
529     the pcre[16]_extra data structure, until it finds the minimum numbers
530     for each parameter that allow pcre[16]_exec() to complete without
531     error. Because this is testing a specific feature of the normal inter-
532     pretive pcre[16]_exec() execution, the use of any JIT optimization that
533     might have been set up by the /S+ qualifier of -s+ option is disabled.
534 ph10 691
535 ph10 836 The match_limit number is a measure of the amount of backtracking that
536     takes place, and checking it out can be instructive. For most simple
537     matches, the number is quite small, but for patterns with very large
538     numbers of matching possibilities, it can become large very quickly
539     with increasing length of subject string. The match_limit_recursion
540     number is a measure of how much stack (or, if PCRE is compiled with
541     NO_RECURSE, how much heap) memory is needed to complete the match
542 ph10 691 attempt.
543    
544 ph10 836 When \O is used, the value specified may be higher or lower than the
545 nigel 75 size set by the -O command line option (or defaulted to 45); \O applies
546 ph10 869 only to the call of pcre[16]_exec() for the line in which it appears.
547 nigel 63
548 ph10 836 If the /P modifier was present on the pattern, causing the POSIX wrap-
549     per API to be used, the only option-setting sequences that have any
550     effect are \B, \N, and \Z, causing REG_NOTBOL, REG_NOTEMPTY, and
551 ph10 535 REG_NOTEOL, respectively, to be passed to regexec().
552 nigel 41
553    
554 nigel 77 THE ALTERNATIVE MATCHING FUNCTION
555 nigel 53
556 ph10 869 By default, pcretest uses the standard PCRE matching function,
557     pcre[16]_exec() to match each data line. PCRE also supports an alterna-
558     tive matching function, pcre[16]_dfa_test(), which operates in a dif-
559     ferent way, and has some restrictions. The differences between the two
560     functions are described in the pcrematching documentation.
561 nigel 77
562 ph10 869 If a data line contains the \D escape sequence, or if the command line
563     contains the -dfa option, the alternative matching function is used.
564 nigel 77 This function finds all possible matches at a given point. If, however,
565 ph10 869 the \F escape sequence is present in the data line, it stops after the
566 nigel 77 first match is found. This is always the shortest possible match.
567    
568    
569     DEFAULT OUTPUT FROM PCRETEST
570    
571 ph10 869 This section describes the output when the normal matching function,
572     pcre[16]_exec(), is being used.
573 nigel 77
574 nigel 73 When a match succeeds, pcretest outputs the list of captured substrings
575 ph10 869 that pcre[16]_exec() returns, starting with number 0 for the string
576     that matched the whole pattern. Otherwise, it outputs "No match" when
577     the return is PCRE_ERROR_NOMATCH, and "Partial match:" followed by the
578     partially matching substring when pcre[16]_exec() returns
579     PCRE_ERROR_PARTIAL. (Note that this is the entire substring that was
580     inspected during the partial match; it may include characters before
581     the actual match start if a lookbehind assertion, \K, \b, or \B was
582     involved.) For any other return, pcretest outputs the PCRE negative
583     error number and a short descriptive phrase. If the error is a failed
584     UTF string check, the offset of the start of the failing character and
585     the reason code are also output, provided that the size of the output
586     vector is at least two. Here is an example of an interactive pcretest
587     run.
588 nigel 53
589 nigel 73 $ pcretest
590 ph10 654 PCRE version 8.13 2011-04-30
591 nigel 53
592 nigel 73 re> /^abc(\d+)/
593     data> abc123
594     0: abc123
595     1: 123
596     data> xyz
597     No match
598 nigel 53
599 ph10 654 Unset capturing substrings that are not followed by one that is set are
600 ph10 869 not returned by pcre[16]_exec(), and are not shown by pcretest. In the
601     following example, there are two capturing substrings, but when the
602     first data line is matched, the second, unset substring is not shown.
603     An "internal" unset substring is shown as "<unset>", as for the second
604 ph10 654 data line.
605 ph10 286
606     re> /(a)|(b)/
607     data> a
608     0: a
609     1: a
610     data> b
611     0: b
612     1: <unset>
613     2: b
614    
615 ph10 836 If the strings contain any non-printing characters, they are output as
616 ph10 869 \xhh escapes if the value is less than 256 and UTF mode is not set.
617     Otherwise they are output as \x{hh...} escapes. See below for the defi-
618     nition of non-printing characters. If the pattern has the /+ modifier,
619     the output for substring 0 is followed by the the rest of the subject
620     string, identified by "0+" like this:
621 nigel 53
622 nigel 73 re> /cat/+
623     data> cataract
624     0: cat
625     0+ aract
626 nigel 53
627 ph10 836 If the pattern has the /g or /G modifier, the results of successive
628 nigel 73 matching attempts are output in sequence, like this:
629 nigel 53
630 nigel 73 re> /\Bi(\w\w)/g
631     data> Mississippi
632     0: iss
633     1: ss
634     0: iss
635     1: ss
636     0: ipp
637     1: pp
638 nigel 53
639 ph10 836 "No match" is output only if the first match attempt fails. Here is an
640     example of a failure message (the offset 4 that is specified by \>4 is
641 ph10 654 past the end of the subject string):
642 nigel 53
643 ph10 654 re> /xyz/
644     data> xyz\>4
645     Error -24 (bad offset value)
646    
647 ph10 836 If any of the sequences \C, \G, or \L are present in a data line that
648     is successfully matched, the substrings extracted by the convenience
649 nigel 73 functions are output with C, G, or L after the string number instead of
650     a colon. This is in addition to the normal full list. The string length
651 ph10 836 (that is, the return from the extraction function) is given in paren-
652 nigel 73 theses after each string for \C and \G.
653 nigel 53
654 nigel 93 Note that whereas patterns can be continued over several lines (a plain
655 nigel 73 ">" prompt is used for continuations), data lines may not. However new-
656 ph10 836 lines can be included in data by means of the \n escape (or \r, \r\n,
657 nigel 93 etc., depending on the newline sequence setting).
658 nigel 53
659    
660 nigel 77 OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION
661    
662 ph10 869 When the alternative matching function, pcre[16]_dfa_exec(), is used
663     (by means of the \D escape sequence or the -dfa command line option),
664     the output consists of a list of all the matches that start at the
665     first point in the subject where there is at least one match. For exam-
666     ple:
667 nigel 77
668     re> /(tang|tangerine|tan)/
669     data> yellow tangerine\D
670     0: tangerine
671     1: tang
672     2: tan
673    
674 ph10 869 (Using the normal matching function on this data finds only "tang".)
675     The longest matching string is always given first (and numbered zero).
676 ph10 429 After a PCRE_ERROR_PARTIAL return, the output is "Partial match:", fol-
677 ph10 869 lowed by the partially matching substring. (Note that this is the
678     entire substring that was inspected during the partial match; it may
679 ph10 567 include characters before the actual match start if a lookbehind asser-
680     tion, \K, \b, or \B was involved.)
681 nigel 77
682 nigel 93 If /g is present on the pattern, the search for further matches resumes
683     at the end of the longest match. For example:
684 nigel 77
685     re> /(tang|tangerine|tan)/g
686     data> yellow tangerine and tangy sultana\D
687     0: tangerine
688     1: tang
689     2: tan
690     0: tang
691     1: tan
692     0: tan
693    
694 ph10 869 Since the matching function does not support substring capture, the
695     escape sequences that are concerned with captured substrings are not
696 nigel 77 relevant.
697    
698    
699     RESTARTING AFTER A PARTIAL MATCH
700    
701     When the alternative matching function has given the PCRE_ERROR_PARTIAL
702 ph10 869 return, indicating that the subject partially matched the pattern, you
703     can restart the match with additional subject data by means of the \R
704 nigel 77 escape sequence. For example:
705    
706 ph10 155 re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
707 nigel 77 data> 23ja\P\D
708     Partial match: 23ja
709     data> n05\R\D
710     0: n05
711    
712 ph10 869 For further information about partial matching, see the pcrepartial
713 nigel 77 documentation.
714    
715    
716 nigel 75 CALLOUTS
717    
718 ph10 869 If the pattern contains any callout requests, pcretest's callout func-
719     tion is called during matching. This works with both matching func-
720 nigel 77 tions. By default, the called function displays the callout number, the
721 ph10 869 start and current positions in the text at the callout time, and the
722     next pattern item to be tested. For example:
723 nigel 75
724     --->pqrabcdef
725     0 ^ ^ \d
726    
727 ph10 869 This output indicates that callout number 0 occurred for a match
728     attempt starting at the fourth character of the subject string, when
729     the pointer was at the seventh character of the data, and when the next
730     pattern item was \d. Just one circumflex is output if the start and
731     current positions are the same.
732 nigel 75
733     Callouts numbered 255 are assumed to be automatic callouts, inserted as
734 ph10 869 a result of the /C pattern modifier. In this case, instead of showing
735     the callout number, the offset in the pattern, preceded by a plus, is
736 nigel 75 output. For example:
737    
738     re> /\d?[A-E]\*/C
739     data> E*
740     --->E*
741     +0 ^ \d?
742     +3 ^ [A-E]
743     +8 ^^ \*
744     +10 ^ ^
745     0: E*
746    
747 ph10 654 If a pattern contains (*MARK) items, an additional line is output when-
748 ph10 869 ever a change of latest mark is passed to the callout function. For
749 ph10 654 example:
750    
751     re> /a(*MARK:X)bc/C
752     data> abc
753     --->abc
754     +0 ^ a
755     +1 ^^ (*MARK:X)
756     +10 ^^ b
757     Latest Mark: X
758     +11 ^ ^ c
759     +12 ^ ^
760     0: abc
761    
762 ph10 869 The mark changes between matching "a" and "b", but stays the same for
763     the rest of the match, so nothing more is output. If, as a result of
764     backtracking, the mark reverts to being unset, the text "<unset>" is
765 ph10 654 output.
766    
767 ph10 869 The callout function in pcretest returns zero (carry on matching) by
768     default, but you can use a \C item in a data line (as described above)
769 ph10 654 to change this and other parameters of the callout.
770 nigel 75
771 ph10 869 Inserting callouts can be helpful when using pcretest to check compli-
772     cated regular expressions. For further information about callouts, see
773 nigel 75 the pcrecallout documentation.
774    
775    
776 nigel 93 NON-PRINTING CHARACTERS
777    
778 ph10 869 When pcretest is outputting text in the compiled version of a pattern,
779     bytes other than 32-126 are always treated as non-printing characters
780 nigel 93 are are therefore shown as hex escapes.
781    
782 ph10 869 When pcretest is outputting text that is a matched part of a subject
783     string, it behaves in the same way, unless a different locale has been
784     set for the pattern (using the /L modifier). In this case, the
785 nigel 93 isprint() function to distinguish printing and non-printing characters.
786    
787    
788 nigel 75 SAVING AND RELOADING COMPILED PATTERNS
789    
790 ph10 869 The facilities described in this section are not available when the
791     POSIX interface to PCRE is being used, that is, when the /P pattern
792 ph10 654 modifier is specified.
793 nigel 75
794     When the POSIX interface is not in use, you can cause pcretest to write
795 ph10 869 a compiled pattern to a file, by following the modifiers with > and a
796 nigel 75 file name. For example:
797    
798     /pattern/im >/some/file
799    
800 ph10 869 See the pcreprecompile documentation for a discussion about saving and
801     re-using compiled patterns. Note that if the pattern was successfully
802 ph10 691 studied with JIT optimization, the JIT data cannot be saved.
803 nigel 75
804 ph10 869 The data that is written is binary. The first eight bytes are the
805     length of the compiled pattern data followed by the length of the
806     optional study data, each written as four bytes in big-endian order
807     (most significant byte first). If there is no study data (either the
808 nigel 75 pattern was not studied, or studying did not return any data), the sec-
809 ph10 869 ond length is zero. The lengths are followed by an exact copy of the
810     compiled pattern. If there is additional study data, this (excluding
811     any JIT data) follows immediately after the compiled pattern. After
812 ph10 691 writing the file, pcretest expects to read a new pattern.
813 nigel 75
814 ph10 869 A saved pattern can be reloaded into pcretest by specifying < and a
815 ph10 654 file name instead of a pattern. The name of the file must not contain a
816     < character, as otherwise pcretest will interpret the line as a pattern
817 nigel 75 delimited by < characters. For example:
818    
819     re> </some/file
820 ph10 654 Compiled pattern loaded from /some/file
821 nigel 75 No study data
822    
823 ph10 869 If the pattern was previously studied with the JIT optimization, the
824     JIT information cannot be saved and restored, and so is lost. When the
825     pattern has been loaded, pcretest proceeds to read data lines in the
826 ph10 691 usual way.
827 nigel 75
828 ph10 869 You can copy a file written by pcretest to a different host and reload
829     it there, even if the new host has opposite endianness to the one on
830     which the pattern was compiled. For example, you can compile on an i86
831     machine and run on a SPARC machine. When a pattern is reloaded on a
832     host with different endianness, the confirmation message is changed to:
833 nigel 75
834 ph10 869 Compiled pattern (byte-inverted) loaded from /some/file
835    
836     The test suite contains some saved pre-compiled patterns with different
837     endianness. These are reloaded using "<!" instead of just "<". This
838     suppresses the "(byte-inverted)" text so that the output is the same on
839     all hosts. It also forces debugging output once the pattern has been
840     reloaded.
841    
842 ph10 836 File names for saving and reloading can be absolute or relative, but
843     note that the shell facility of expanding a file name that starts with
844 nigel 75 a tilde (~) is not available.
845    
846 ph10 836 The ability to save and reload files in pcretest is intended for test-
847     ing and experimentation. It is not intended for production use because
848     only a single pattern can be written to a file. Furthermore, there is
849     no facility for supplying custom character tables for use with a
850     reloaded pattern. If the original pattern was compiled with custom
851     tables, an attempt to match a subject string using a reloaded pattern
852     is likely to cause pcretest to crash. Finally, if you attempt to load
853 nigel 75 a file that is not in the correct format, the result is undefined.
854    
855    
856 nigel 93 SEE ALSO
857    
858 ph10 869 pcre(3), pcre16(3), pcreapi(3), pcrecallout(3), pcrejit, pcrematch-
859     ing(3), pcrepartial(d), pcrepattern(3), pcreprecompile(3).
860 nigel 93
861    
862 nigel 63 AUTHOR
863 nigel 53
864 nigel 77 Philip Hazel
865 ph10 99 University Computing Service
866 nigel 93 Cambridge CB2 3QH, England.
867 nigel 53
868 ph10 99
869     REVISION
870    
871 ph10 869 Last updated: 13 January 2012
872     Copyright (c) 1997-2012 University of Cambridge.

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12