• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1<html>
2<head>
3<title>pcre2test specification</title>
4</head>
5<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6<h1>pcre2test man page</h1>
7<p>
8Return to the <a href="index.html">PCRE2 index page</a>.
9</p>
10<p>
11This page is part of the PCRE2 HTML documentation. It was generated
12automatically from the original man page. If there is any nonsense in it,
13please consult the man page, in case the conversion went wrong.
14<br>
15<ul>
16<li><a name="TOC1" href="#SEC1">SYNOPSIS</a>
17<li><a name="TOC2" href="#SEC2">PCRE2's 8-BIT, 16-BIT AND 32-BIT LIBRARIES</a>
18<li><a name="TOC3" href="#SEC3">INPUT ENCODING</a>
19<li><a name="TOC4" href="#SEC4">COMMAND LINE OPTIONS</a>
20<li><a name="TOC5" href="#SEC5">DESCRIPTION</a>
21<li><a name="TOC6" href="#SEC6">COMMAND LINES</a>
22<li><a name="TOC7" href="#SEC7">MODIFIER SYNTAX</a>
23<li><a name="TOC8" href="#SEC8">PATTERN SYNTAX</a>
24<li><a name="TOC9" href="#SEC9">SUBJECT LINE SYNTAX</a>
25<li><a name="TOC10" href="#SEC10">PATTERN MODIFIERS</a>
26<li><a name="TOC11" href="#SEC11">SUBJECT MODIFIERS</a>
27<li><a name="TOC12" href="#SEC12">THE ALTERNATIVE MATCHING FUNCTION</a>
28<li><a name="TOC13" href="#SEC13">DEFAULT OUTPUT FROM pcre2test</a>
29<li><a name="TOC14" href="#SEC14">OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION</a>
30<li><a name="TOC15" href="#SEC15">RESTARTING AFTER A PARTIAL MATCH</a>
31<li><a name="TOC16" href="#SEC16">CALLOUTS</a>
32<li><a name="TOC17" href="#SEC17">NON-PRINTING CHARACTERS</a>
33<li><a name="TOC18" href="#SEC18">SAVING AND RESTORING COMPILED PATTERNS</a>
34<li><a name="TOC19" href="#SEC19">SEE ALSO</a>
35<li><a name="TOC20" href="#SEC20">AUTHOR</a>
36<li><a name="TOC21" href="#SEC21">REVISION</a>
37</ul>
38<br><a name="SEC1" href="#TOC1">SYNOPSIS</a><br>
39<P>
40<b>pcre2test [options] [input file [output file]]</b>
41<br>
42<br>
43<b>pcre2test</b> is a test program for the PCRE2 regular expression libraries,
44but it can also be used for experimenting with regular expressions. This
45document describes the features of the test program; for details of the regular
46expressions themselves, see the
47<a href="pcre2pattern.html"><b>pcre2pattern</b></a>
48documentation. For details of the PCRE2 library function calls and their
49options, see the
50<a href="pcre2api.html"><b>pcre2api</b></a>
51documentation.
52</P>
53<P>
54The input for <b>pcre2test</b> is a sequence of regular expression patterns and
55subject strings to be matched. There are also command lines for setting
56defaults and controlling some special actions. The output shows the result of
57each match attempt. Modifiers on external or internal command lines, the
58patterns, and the subject lines specify PCRE2 function options, control how the
59subject is processed, and what output is produced.
60</P>
61<P>
62There are many obscure modifiers, some of which are specifically designed for
63use in conjunction with the test script and data files that are distributed as
64part of PCRE2. All the modifiers are documented here, some without much
65justification, but many of them are unlikely to be of use except when testing
66the libraries.
67</P>
68<br><a name="SEC2" href="#TOC1">PCRE2's 8-BIT, 16-BIT AND 32-BIT LIBRARIES</a><br>
69<P>
70Different versions of the PCRE2 library can be built to support character
71strings that are encoded in 8-bit, 16-bit, or 32-bit code units. One, two, or
72all three of these libraries may be simultaneously installed. The
73<b>pcre2test</b> program can be used to test all the libraries. However, its own
74input and output are always in 8-bit format. When testing the 16-bit or 32-bit
75libraries, patterns and subject strings are converted to 16-bit or 32-bit
76format before being passed to the library functions. Results are converted back
77to 8-bit code units for output.
78</P>
79<P>
80In the rest of this document, the names of library functions and structures
81are given in generic form, for example, <b>pcre2_compile()</b>. The actual
82names used in the libraries have a suffix _8, _16, or _32, as appropriate.
83<a name="inputencoding"></a></P>
84<br><a name="SEC3" href="#TOC1">INPUT ENCODING</a><br>
85<P>
86Input to <b>pcre2test</b> is processed line by line, either by calling the C
87library's <b>fgets()</b> function, or via the <b>libreadline</b> or <b>libedit</b>
88library. In some Windows environments character 26 (hex 1A) causes an immediate
89end of file, and no further data is read, so this character should be avoided
90unless you really want that action.
91</P>
92<P>
93The input is processed using using C's string functions, so must not
94contain binary zeros, even though in Unix-like environments, <b>fgets()</b>
95treats any bytes other than newline as data characters. An error is generated
96if a binary zero is encountered. By default subject lines are processed for
97backslash escapes, which makes it possible to include any data value in strings
98that are passed to the library for matching. For patterns, there is a facility
99for specifying some or all of the 8-bit input characters as hexadecimal pairs,
100which makes it possible to include binary zeros.
101</P>
102<br><b>
103Input for the 16-bit and 32-bit libraries
104</b><br>
105<P>
106When testing the 16-bit or 32-bit libraries, there is a need to be able to
107generate character code points greater than 255 in the strings that are passed
108to the library. For subject lines, backslash escapes can be used. In addition,
109when the <b>utf</b> modifier (see
110<a href="#optionmodifiers">"Setting compilation options"</a>
111below) is set, the pattern and any following subject lines are interpreted as
112UTF-8 strings and translated to UTF-16 or UTF-32 as appropriate.
113</P>
114<P>
115For non-UTF testing of wide characters, the <b>utf8_input</b> modifier can be
116used. This is mutually exclusive with <b>utf</b>, and is allowed only in 16-bit
117or 32-bit mode. It causes the pattern and following subject lines to be treated
118as UTF-8 according to the original definition (RFC 2279), which allows for
119character values up to 0x7fffffff. Each character is placed in one 16-bit or
12032-bit code unit (in the 16-bit case, values greater than 0xffff cause an error
121to occur).
122</P>
123<P>
124UTF-8 (in its original definition) is not capable of encoding values greater
125than 0x7fffffff, but such values can be handled by the 32-bit library. When
126testing this library in non-UTF mode with <b>utf8_input</b> set, if any
127character is preceded by the byte 0xff (which is an invalid byte in UTF-8)
1280x80000000 is added to the character's value. This is the only way of passing
129such code points in a pattern string. For subject strings, using an escape
130sequence is preferable.
131</P>
132<br><a name="SEC4" href="#TOC1">COMMAND LINE OPTIONS</a><br>
133<P>
134<b>-8</b>
135If the 8-bit library has been built, this option causes it to be used (this is
136the default). If the 8-bit library has not been built, this option causes an
137error.
138</P>
139<P>
140<b>-16</b>
141If the 16-bit library has been built, this option causes it to be used. If only
142the 16-bit library has been built, this is the default. If the 16-bit library
143has not been built, this option causes an error.
144</P>
145<P>
146<b>-32</b>
147If the 32-bit library has been built, this option causes it to be used. If only
148the 32-bit library has been built, this is the default. If the 32-bit library
149has not been built, this option causes an error.
150</P>
151<P>
152<b>-ac</b>
153Behave as if each pattern has the <b>auto_callout</b> modifier, that is, insert
154automatic callouts into every pattern that is compiled.
155</P>
156<P>
157<b>-AC</b>
158As for <b>-ac</b>, but in addition behave as if each subject line has the
159<b>callout_extra</b> modifier, that is, show additional information from
160callouts.
161</P>
162<P>
163<b>-b</b>
164Behave as if each pattern has the <b>fullbincode</b> modifier; the full
165internal binary form of the pattern is output after compilation.
166</P>
167<P>
168<b>-C</b>
169Output the version number of the PCRE2 library, and all available information
170about the optional features that are included, and then exit with zero exit
171code. All other options are ignored. If both -C and -LM are present, whichever
172is first is recognized.
173</P>
174<P>
175<b>-C</b> <i>option</i>
176Output information about a specific build-time option, then exit. This
177functionality is intended for use in scripts such as <b>RunTest</b>. The
178following options output the value and set the exit code as indicated:
179<pre>
180  ebcdic-nl  the code for LF (= NL) in an EBCDIC environment:
181               0x15 or 0x25
182               0 if used in an ASCII environment
183               exit code is always 0
184  linksize   the configured internal link size (2, 3, or 4)
185               exit code is set to the link size
186  newline    the default newline setting:
187               CR, LF, CRLF, ANYCRLF, ANY, or NUL
188               exit code is always 0
189  bsr        the default setting for what \R matches:
190               ANYCRLF or ANY
191               exit code is always 0
192</pre>
193The following options output 1 for true or 0 for false, and set the exit code
194to the same value:
195<pre>
196  backslash-C  \C is supported (not locked out)
197  ebcdic       compiled for an EBCDIC environment
198  jit          just-in-time support is available
199  pcre2-16     the 16-bit library was built
200  pcre2-32     the 32-bit library was built
201  pcre2-8      the 8-bit library was built
202  unicode      Unicode support is available
203</pre>
204If an unknown option is given, an error message is output; the exit code is 0.
205</P>
206<P>
207<b>-d</b>
208Behave as if each pattern has the <b>debug</b> modifier; the internal
209form and information about the compiled pattern is output after compilation;
210<b>-d</b> is equivalent to <b>-b -i</b>.
211</P>
212<P>
213<b>-dfa</b>
214Behave as if each subject line has the <b>dfa</b> modifier; matching is done
215using the <b>pcre2_dfa_match()</b> function instead of the default
216<b>pcre2_match()</b>.
217</P>
218<P>
219<b>-error</b> <i>number[,number,...]</i>
220Call <b>pcre2_get_error_message()</b> for each of the error numbers in the
221comma-separated list, display the resulting messages on the standard output,
222then exit with zero exit code. The numbers may be positive or negative. This is
223a convenience facility for PCRE2 maintainers.
224</P>
225<P>
226<b>-help</b>
227Output a brief summary these options and then exit.
228</P>
229<P>
230<b>-i</b>
231Behave as if each pattern has the <b>info</b> modifier; information about the
232compiled pattern is given after compilation.
233</P>
234<P>
235<b>-jit</b>
236Behave as if each pattern line has the <b>jit</b> modifier; after successful
237compilation, each pattern is passed to the just-in-time compiler, if available.
238</P>
239<P>
240<b>-jitfast</b>
241Behave as if each pattern line has the <b>jitfast</b> modifier; after
242successful compilation, each pattern is passed to the just-in-time compiler, if
243available, and each subject line is passed directly to the JIT matcher via its
244"fast path".
245</P>
246<P>
247<b>-jitverify</b>
248Behave as if each pattern line has the <b>jitverify</b> modifier; after
249successful compilation, each pattern is passed to the just-in-time compiler, if
250available, and the use of JIT for matching is verified.
251</P>
252<P>
253<b>-LM</b>
254List modifiers: write a list of available pattern and subject modifiers to the
255standard output, then exit with zero exit code. All other options are ignored.
256If both -C and any -Lx options are present, whichever is first is recognized.
257</P>
258<P>
259<b>-LP</b>
260List properties: write a list of recognized Unicode properties to the standard
261output, then exit with zero exit code. All other options are ignored. If both
262-C and any -Lx options are present, whichever is first is recognized.
263</P>
264<P>
265<b>-LS</b>
266List scripts: write a list of recogized Unicode script names to the standard
267output, then exit with zero exit code. All other options are ignored. If both
268-C and any -Lx options are present, whichever is first is recognized.
269</P>
270<P>
271<b>-pattern</b> <i>modifier-list</i>
272Behave as if each pattern line contains the given modifiers.
273</P>
274<P>
275<b>-q</b>
276Do not output the version number of <b>pcre2test</b> at the start of execution.
277</P>
278<P>
279<b>-S</b> <i>size</i>
280On Unix-like systems, set the size of the run-time stack to <i>size</i>
281mebibytes (units of 1024*1024 bytes).
282</P>
283<P>
284<b>-subject</b> <i>modifier-list</i>
285Behave as if each subject line contains the given modifiers.
286</P>
287<P>
288<b>-t</b>
289Run each compile and match many times with a timer, and output the resulting
290times per compile or match. When JIT is used, separate times are given for the
291initial compile and the JIT compile. You can control the number of iterations
292that are used for timing by following <b>-t</b> with a number (as a separate
293item on the command line). For example, "-t 1000" iterates 1000 times. The
294default is to iterate 500,000 times.
295</P>
296<P>
297<b>-tm</b>
298This is like <b>-t</b> except that it times only the matching phase, not the
299compile phase.
300</P>
301<P>
302<b>-T</b> <b>-TM</b>
303These behave like <b>-t</b> and <b>-tm</b>, but in addition, at the end of a run,
304the total times for all compiles and matches are output.
305</P>
306<P>
307<b>-version</b>
308Output the PCRE2 version number and then exit.
309</P>
310<br><a name="SEC5" href="#TOC1">DESCRIPTION</a><br>
311<P>
312If <b>pcre2test</b> is given two filename arguments, it reads from the first and
313writes to the second. If the first name is "-", input is taken from the
314standard input. If <b>pcre2test</b> is given only one argument, it reads from
315that file and writes to stdout. Otherwise, it reads from stdin and writes to
316stdout.
317</P>
318<P>
319When <b>pcre2test</b> is built, a configuration option can specify that it
320should be linked with the <b>libreadline</b> or <b>libedit</b> library. When this
321is done, if the input is from a terminal, it is read using the <b>readline()</b>
322function. This provides line-editing and history facilities. The output from
323the <b>-help</b> option states whether or not <b>readline()</b> will be used.
324</P>
325<P>
326The program handles any number of tests, each of which consists of a set of
327input lines. Each set starts with a regular expression pattern, followed by any
328number of subject lines to be matched against that pattern. In between sets of
329test data, command lines that begin with # may appear. This file format, with
330some restrictions, can also be processed by the <b>perltest.sh</b> script that
331is distributed with PCRE2 as a means of checking that the behaviour of PCRE2
332and Perl is the same. For a specification of <b>perltest.sh</b>, see the
333comments near its beginning. See also the #perltest command below.
334</P>
335<P>
336When the input is a terminal, <b>pcre2test</b> prompts for each line of input,
337using "re&#62;" to prompt for regular expression patterns, and "data&#62;" to prompt
338for subject lines. Command lines starting with # can be entered only in
339response to the "re&#62;" prompt.
340</P>
341<P>
342Each subject line is matched separately and independently. If you want to do
343multi-line matches, you have to use the \n escape sequence (or \r or \r\n,
344etc., depending on the newline setting) in a single line of input to encode the
345newline sequences. There is no limit on the length of subject lines; the input
346buffer is automatically extended if it is too small. There are replication
347features that makes it possible to generate long repetitive pattern or subject
348lines without having to supply them explicitly.
349</P>
350<P>
351An empty line or the end of the file signals the end of the subject lines for a
352test, at which point a new pattern or command line is expected if there is
353still input to be read.
354</P>
355<br><a name="SEC6" href="#TOC1">COMMAND LINES</a><br>
356<P>
357In between sets of test data, a line that begins with # is interpreted as a
358command line. If the first character is followed by white space or an
359exclamation mark, the line is treated as a comment, and ignored. Otherwise, the
360following commands are recognized:
361<pre>
362  #forbid_utf
363</pre>
364Subsequent patterns automatically have the PCRE2_NEVER_UTF and PCRE2_NEVER_UCP
365options set, which locks out the use of the PCRE2_UTF and PCRE2_UCP options and
366the use of (*UTF) and (*UCP) at the start of patterns. This command also forces
367an error if a subsequent pattern contains any occurrences of \P, \p, or \X,
368which are still supported when PCRE2_UTF is not set, but which require Unicode
369property support to be included in the library.
370</P>
371<P>
372This is a trigger guard that is used in test files to ensure that UTF or
373Unicode property tests are not accidentally added to files that are used when
374Unicode support is not included in the library. Setting PCRE2_NEVER_UTF and
375PCRE2_NEVER_UCP as a default can also be obtained by the use of <b>#pattern</b>;
376the difference is that <b>#forbid_utf</b> cannot be unset, and the automatic
377options are not displayed in pattern information, to avoid cluttering up test
378output.
379<pre>
380  #load &#60;filename&#62;
381</pre>
382This command is used to load a set of precompiled patterns from a file, as
383described in the section entitled "Saving and restoring compiled patterns"
384<a href="#saverestore">below.</a>
385<pre>
386  #loadtables &#60;filename&#62;
387</pre>
388This command is used to load a set of binary character tables that can be
389accessed by the tables=3 qualifier. Such tables can be created by the
390<b>pcre2_dftables</b> program with the -b option.
391<pre>
392  #newline_default [&#60;newline-list&#62;]
393</pre>
394When PCRE2 is built, a default newline convention can be specified. This
395determines which characters and/or character pairs are recognized as indicating
396a newline in a pattern or subject string. The default can be overridden when a
397pattern is compiled. The standard test files contain tests of various newline
398conventions, but the majority of the tests expect a single linefeed to be
399recognized as a newline by default. Without special action the tests would fail
400when PCRE2 is compiled with either CR or CRLF as the default newline.
401</P>
402<P>
403The #newline_default command specifies a list of newline types that are
404acceptable as the default. The types must be one of CR, LF, CRLF, ANYCRLF,
405ANY, or NUL (in upper or lower case), for example:
406<pre>
407  #newline_default LF Any anyCRLF
408</pre>
409If the default newline is in the list, this command has no effect. Otherwise,
410except when testing the POSIX API, a <b>newline</b> modifier that specifies the
411first newline convention in the list (LF in the above example) is added to any
412pattern that does not already have a <b>newline</b> modifier. If the newline
413list is empty, the feature is turned off. This command is present in a number
414of the standard test input files.
415</P>
416<P>
417When the POSIX API is being tested there is no way to override the default
418newline convention, though it is possible to set the newline convention from
419within the pattern. A warning is given if the <b>posix</b> or <b>posix_nosub</b>
420modifier is used when <b>#newline_default</b> would set a default for the
421non-POSIX API.
422<pre>
423  #pattern &#60;modifier-list&#62;
424</pre>
425This command sets a default modifier list that applies to all subsequent
426patterns. Modifiers on a pattern can change these settings.
427<pre>
428  #perltest
429</pre>
430This line is used in test files that can also be processed by <b>perltest.sh</b>
431to confirm that Perl gives the same results as PCRE2. Subsequent tests are
432checked for the use of <b>pcre2test</b> features that are incompatible with the
433<b>perltest.sh</b> script.
434</P>
435<P>
436Patterns must use '/' as their delimiter, and only certain modifiers are
437supported. Comment lines, #pattern commands, and #subject commands that set or
438unset "mark" are recognized and acted on. The #perltest, #forbid_utf, and
439#newline_default commands, which are needed in the relevant pcre2test files,
440are silently ignored. All other command lines are ignored, but give a warning
441message. The <b>#perltest</b> command helps detect tests that are accidentally
442put in the wrong file or use the wrong delimiter. For more details of the
443<b>perltest.sh</b> script see the comments it contains.
444<pre>
445  #pop [&#60;modifiers&#62;]
446  #popcopy [&#60;modifiers&#62;]
447</pre>
448These commands are used to manipulate the stack of compiled patterns, as
449described in the section entitled "Saving and restoring compiled patterns"
450<a href="#saverestore">below.</a>
451<pre>
452  #save &#60;filename&#62;
453</pre>
454This command is used to save a set of compiled patterns to a file, as described
455in the section entitled "Saving and restoring compiled patterns"
456<a href="#saverestore">below.</a>
457<pre>
458  #subject &#60;modifier-list&#62;
459</pre>
460This command sets a default modifier list that applies to all subsequent
461subject lines. Modifiers on a subject line can change these settings.
462</P>
463<br><a name="SEC7" href="#TOC1">MODIFIER SYNTAX</a><br>
464<P>
465Modifier lists are used with both pattern and subject lines. Items in a list
466are separated by commas followed by optional white space. Trailing whitespace
467in a modifier list is ignored. Some modifiers may be given for both patterns
468and subject lines, whereas others are valid only for one or the other. Each
469modifier has a long name, for example "anchored", and some of them must be
470followed by an equals sign and a value, for example, "offset=12". Values cannot
471contain comma characters, but may contain spaces. Modifiers that do not take
472values may be preceded by a minus sign to turn off a previous setting.
473</P>
474<P>
475A few of the more common modifiers can also be specified as single letters, for
476example "i" for "caseless". In documentation, following the Perl convention,
477these are written with a slash ("the /i modifier") for clarity. Abbreviated
478modifiers must all be concatenated in the first item of a modifier list. If the
479first item is not recognized as a long modifier name, it is interpreted as a
480sequence of these abbreviations. For example:
481<pre>
482  /abc/ig,newline=cr,jit=3
483</pre>
484This is a pattern line whose modifier list starts with two one-letter modifiers
485(/i and /g). The lower-case abbreviated modifiers are the same as used in Perl.
486</P>
487<br><a name="SEC8" href="#TOC1">PATTERN SYNTAX</a><br>
488<P>
489A pattern line must start with one of the following characters (common symbols,
490excluding pattern meta-characters):
491<pre>
492  / ! " ' ` - = _ : ; , % & @ ~
493</pre>
494This is interpreted as the pattern's delimiter. A regular expression may be
495continued over several input lines, in which case the newline characters are
496included within it. It is possible to include the delimiter as a literal within
497the pattern by escaping it with a backslash, for example
498<pre>
499  /abc\/def/
500</pre>
501If you do this, the escape and the delimiter form part of the pattern, but
502since the delimiters are all non-alphanumeric, the inclusion of the backslash
503does not affect the pattern's interpretation. Note, however, that this trick
504does not work within \Q...\E literal bracketing because the backslash will
505itself be interpreted as a literal. If the terminating delimiter is immediately
506followed by a backslash, for example,
507<pre>
508  /abc/\
509</pre>
510then a backslash is added to the end of the pattern. This is done to provide a
511way of testing the error condition that arises if a pattern finishes with a
512backslash, because
513<pre>
514  /abc\/
515</pre>
516is interpreted as the first line of a pattern that starts with "abc/", causing
517pcre2test to read the next line as a continuation of the regular expression.
518</P>
519<P>
520A pattern can be followed by a modifier list (details below).
521</P>
522<br><a name="SEC9" href="#TOC1">SUBJECT LINE SYNTAX</a><br>
523<P>
524Before each subject line is passed to <b>pcre2_match()</b>,
525<b>pcre2_dfa_match()</b>, or <b>pcre2_jit_match()</b>, leading and trailing white
526space is removed, and the line is scanned for backslash escapes, unless the
527<b>subject_literal</b> modifier was set for the pattern. The following provide a
528means of encoding non-printing characters in a visible way:
529<pre>
530  \a         alarm (BEL, \x07)
531  \b         backspace (\x08)
532  \e         escape (\x27)
533  \f         form feed (\x0c)
534  \n         newline (\x0a)
535  \r         carriage return (\x0d)
536  \t         tab (\x09)
537  \v         vertical tab (\x0b)
538  \nnn       octal character (up to 3 octal digits); always
539               a byte unless &#62; 255 in UTF-8 or 16-bit or 32-bit mode
540  \o{dd...}  octal character (any number of octal digits}
541  \xhh       hexadecimal byte (up to 2 hex digits)
542  \x{hh...}  hexadecimal character (any number of hex digits)
543</pre>
544The use of \x{hh...} is not dependent on the use of the <b>utf</b> modifier on
545the pattern. It is recognized always. There may be any number of hexadecimal
546digits inside the braces; invalid values provoke error messages.
547</P>
548<P>
549Note that \xhh specifies one byte rather than one character in UTF-8 mode;
550this makes it possible to construct invalid UTF-8 sequences for testing
551purposes. On the other hand, \x{hh} is interpreted as a UTF-8 character in
552UTF-8 mode, generating more than one byte if the value is greater than 127.
553When testing the 8-bit library not in UTF-8 mode, \x{hh} generates one byte
554for values less than 256, and causes an error for greater values.
555</P>
556<P>
557In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it
558possible to construct invalid UTF-16 sequences for testing purposes.
559</P>
560<P>
561In UTF-32 mode, all 4- to 8-digit \x{...} values are accepted. This makes it
562possible to construct invalid UTF-32 sequences for testing purposes.
563</P>
564<P>
565There is a special backslash sequence that specifies replication of one or more
566characters:
567<pre>
568  \[&#60;characters&#62;]{&#60;count&#62;}
569</pre>
570This makes it possible to test long strings without having to provide them as
571part of the file. For example:
572<pre>
573  \[abc]{4}
574</pre>
575is converted to "abcabcabcabc". This feature does not support nesting. To
576include a closing square bracket in the characters, code it as \x5D.
577</P>
578<P>
579A backslash followed by an equals sign marks the end of the subject string and
580the start of a modifier list. For example:
581<pre>
582  abc\=notbol,notempty
583</pre>
584If the subject string is empty and \= is followed by whitespace, the line is
585treated as a comment line, and is not used for matching. For example:
586<pre>
587  \= This is a comment.
588  abc\= This is an invalid modifier list.
589</pre>
590A backslash followed by any other non-alphanumeric character just escapes that
591character. A backslash followed by anything else causes an error. However, if
592the very last character in the line is a backslash (and there is no modifier
593list), it is ignored. This gives a way of passing an empty line as data, since
594a real empty line terminates the data input.
595</P>
596<P>
597If the <b>subject_literal</b> modifier is set for a pattern, all subject lines
598that follow are treated as literals, with no special treatment of backslashes.
599No replication is possible, and any subject modifiers must be set as defaults
600by a <b>#subject</b> command.
601</P>
602<br><a name="SEC10" href="#TOC1">PATTERN MODIFIERS</a><br>
603<P>
604There are several types of modifier that can appear in pattern lines. Except
605where noted below, they may also be used in <b>#pattern</b> commands. A
606pattern's modifier list can add to or override default modifiers that were set
607by a previous <b>#pattern</b> command.
608<a name="optionmodifiers"></a></P>
609<br><b>
610Setting compilation options
611</b><br>
612<P>
613The following modifiers set options for <b>pcre2_compile()</b>. Most of them set
614bits in the options argument of that function, but those whose names start with
615PCRE2_EXTRA are additional options that are set in the compile context. For the
616main options, there are some single-letter abbreviations that are the same as
617Perl options. There is special handling for /x: if a second x is present,
618PCRE2_EXTENDED is converted into PCRE2_EXTENDED_MORE as in Perl. A third
619appearance adds PCRE2_EXTENDED as well, though this makes no difference to the
620way <b>pcre2_compile()</b> behaves. See
621<a href="pcre2api.html"><b>pcre2api</b></a>
622for a description of the effects of these options.
623<pre>
624      allow_empty_class         set PCRE2_ALLOW_EMPTY_CLASS
625      allow_lookaround_bsk      set PCRE2_EXTRA_ALLOW_LOOKAROUND_BSK
626      allow_surrogate_escapes   set PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES
627      alt_bsux                  set PCRE2_ALT_BSUX
628      alt_circumflex            set PCRE2_ALT_CIRCUMFLEX
629      alt_verbnames             set PCRE2_ALT_VERBNAMES
630      anchored                  set PCRE2_ANCHORED
631      auto_callout              set PCRE2_AUTO_CALLOUT
632      bad_escape_is_literal     set PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL
633  /i  caseless                  set PCRE2_CASELESS
634      dollar_endonly            set PCRE2_DOLLAR_ENDONLY
635  /s  dotall                    set PCRE2_DOTALL
636      dupnames                  set PCRE2_DUPNAMES
637      endanchored               set PCRE2_ENDANCHORED
638      escaped_cr_is_lf          set PCRE2_EXTRA_ESCAPED_CR_IS_LF
639  /x  extended                  set PCRE2_EXTENDED
640  /xx extended_more             set PCRE2_EXTENDED_MORE
641      extra_alt_bsux            set PCRE2_EXTRA_ALT_BSUX
642      firstline                 set PCRE2_FIRSTLINE
643      literal                   set PCRE2_LITERAL
644      match_line                set PCRE2_EXTRA_MATCH_LINE
645      match_invalid_utf         set PCRE2_MATCH_INVALID_UTF
646      match_unset_backref       set PCRE2_MATCH_UNSET_BACKREF
647      match_word                set PCRE2_EXTRA_MATCH_WORD
648  /m  multiline                 set PCRE2_MULTILINE
649      never_backslash_c         set PCRE2_NEVER_BACKSLASH_C
650      never_ucp                 set PCRE2_NEVER_UCP
651      never_utf                 set PCRE2_NEVER_UTF
652  /n  no_auto_capture           set PCRE2_NO_AUTO_CAPTURE
653      no_auto_possess           set PCRE2_NO_AUTO_POSSESS
654      no_dotstar_anchor         set PCRE2_NO_DOTSTAR_ANCHOR
655      no_start_optimize         set PCRE2_NO_START_OPTIMIZE
656      no_utf_check              set PCRE2_NO_UTF_CHECK
657      ucp                       set PCRE2_UCP
658      ungreedy                  set PCRE2_UNGREEDY
659      use_offset_limit          set PCRE2_USE_OFFSET_LIMIT
660      utf                       set PCRE2_UTF
661</pre>
662As well as turning on the PCRE2_UTF option, the <b>utf</b> modifier causes all
663non-printing characters in output strings to be printed using the \x{hh...}
664notation. Otherwise, those less than 0x100 are output in hex without the curly
665brackets. Setting <b>utf</b> in 16-bit or 32-bit mode also causes pattern and
666subject strings to be translated to UTF-16 or UTF-32, respectively, before
667being passed to library functions.
668<a name="controlmodifiers"></a></P>
669<br><b>
670Setting compilation controls
671</b><br>
672<P>
673The following modifiers affect the compilation process or request information
674about the pattern. There are single-letter abbreviations for some that are
675heavily used in the test files.
676<pre>
677      bsr=[anycrlf|unicode]     specify \R handling
678  /B  bincode                   show binary code without lengths
679      callout_info              show callout information
680      convert=&#60;options&#62;         request foreign pattern conversion
681      convert_glob_escape=c     set glob escape character
682      convert_glob_separator=c  set glob separator character
683      convert_length            set convert buffer length
684      debug                     same as info,fullbincode
685      framesize                 show matching frame size
686      fullbincode               show binary code with lengths
687  /I  info                      show info about compiled pattern
688      hex                       unquoted characters are hexadecimal
689      jit[=&#60;number&#62;]            use JIT
690      jitfast                   use JIT fast path
691      jitverify                 verify JIT use
692      locale=&#60;name&#62;             use this locale
693      max_pattern_length=&#60;n&#62;    set the maximum pattern length
694      memory                    show memory used
695      newline=&#60;type&#62;            set newline type
696      null_context              compile with a NULL context
697      parens_nest_limit=&#60;n&#62;     set maximum parentheses depth
698      posix                     use the POSIX API
699      posix_nosub               use the POSIX API with REG_NOSUB
700      push                      push compiled pattern onto the stack
701      pushcopy                  push a copy onto the stack
702      stackguard=&#60;number&#62;       test the stackguard feature
703      subject_literal           treat all subject lines as literal
704      tables=[0|1|2|3]          select internal tables
705      use_length                do not zero-terminate the pattern
706      utf8_input                treat input as UTF-8
707</pre>
708The effects of these modifiers are described in the following sections.
709</P>
710<br><b>
711Newline and \R handling
712</b><br>
713<P>
714The <b>bsr</b> modifier specifies what \R in a pattern should match. If it is
715set to "anycrlf", \R matches CR, LF, or CRLF only. If it is set to "unicode",
716\R matches any Unicode newline sequence. The default can be specified when
717PCRE2 is built; if it is not, the default is set to Unicode.
718</P>
719<P>
720The <b>newline</b> modifier specifies which characters are to be interpreted as
721newlines, both in the pattern and in subject lines. The type must be one of CR,
722LF, CRLF, ANYCRLF, ANY, or NUL (in upper or lower case).
723</P>
724<br><b>
725Information about a pattern
726</b><br>
727<P>
728The <b>debug</b> modifier is a shorthand for <b>info,fullbincode</b>, requesting
729all available information.
730</P>
731<P>
732The <b>bincode</b> modifier causes a representation of the compiled code to be
733output after compilation. This information does not contain length and offset
734values, which ensures that the same output is generated for different internal
735link sizes and different code unit widths. By using <b>bincode</b>, the same
736regression tests can be used in different environments.
737</P>
738<P>
739The <b>fullbincode</b> modifier, by contrast, <i>does</i> include length and
740offset values. This is used in a few special tests that run only for specific
741code unit widths and link sizes, and is also useful for one-off tests.
742</P>
743<P>
744The <b>info</b> modifier requests information about the compiled pattern
745(whether it is anchored, has a fixed first character, and so on). The
746information is obtained from the <b>pcre2_pattern_info()</b> function. Here are
747some typical examples:
748<pre>
749    re&#62; /(?i)(^a|^b)/m,info
750  Capture group count = 1
751  Compile options: multiline
752  Overall options: caseless multiline
753  First code unit at start or follows newline
754  Subject length lower bound = 1
755
756    re&#62; /(?i)abc/info
757  Capture group count = 0
758  Compile options: &#60;none&#62;
759  Overall options: caseless
760  First code unit = 'a' (caseless)
761  Last code unit = 'c' (caseless)
762  Subject length lower bound = 3
763</pre>
764"Compile options" are those specified by modifiers; "overall options" have
765added options that are taken or deduced from the pattern. If both sets of
766options are the same, just a single "options" line is output; if there are no
767options, the line is omitted. "First code unit" is where any match must start;
768if there is more than one they are listed as "starting code units". "Last code
769unit" is the last literal code unit that must be present in any match. This is
770not necessarily the last character. These lines are omitted if no starting or
771ending code units are recorded. The subject length line is omitted when
772<b>no_start_optimize</b> is set because the minimum length is not calculated
773when it can never be used.
774</P>
775<P>
776The <b>framesize</b> modifier shows the size, in bytes, of the storage frames
777used by <b>pcre2_match()</b> for handling backtracking. The size depends on the
778number of capturing parentheses in the pattern.
779</P>
780<P>
781The <b>callout_info</b> modifier requests information about all the callouts in
782the pattern. A list of them is output at the end of any other information that
783is requested. For each callout, either its number or string is given, followed
784by the item that follows it in the pattern.
785</P>
786<br><b>
787Passing a NULL context
788</b><br>
789<P>
790Normally, <b>pcre2test</b> passes a context block to <b>pcre2_compile()</b>. If
791the <b>null_context</b> modifier is set, however, NULL is passed. This is for
792testing that <b>pcre2_compile()</b> behaves correctly in this case (it uses
793default values).
794</P>
795<br><b>
796Specifying pattern characters in hexadecimal
797</b><br>
798<P>
799The <b>hex</b> modifier specifies that the characters of the pattern, except for
800substrings enclosed in single or double quotes, are to be interpreted as pairs
801of hexadecimal digits. This feature is provided as a way of creating patterns
802that contain binary zeros and other non-printing characters. White space is
803permitted between pairs of digits. For example, this pattern contains three
804characters:
805<pre>
806  /ab 32 59/hex
807</pre>
808Parts of such a pattern are taken literally if quoted. This pattern contains
809nine characters, only two of which are specified in hexadecimal:
810<pre>
811  /ab "literal" 32/hex
812</pre>
813Either single or double quotes may be used. There is no way of including
814the delimiter within a substring. The <b>hex</b> and <b>expand</b> modifiers are
815mutually exclusive.
816</P>
817<br><b>
818Specifying the pattern's length
819</b><br>
820<P>
821By default, patterns are passed to the compiling functions as zero-terminated
822strings but can be passed by length instead of being zero-terminated. The
823<b>use_length</b> modifier causes this to happen. Using a length happens
824automatically (whether or not <b>use_length</b> is set) when <b>hex</b> is set,
825because patterns specified in hexadecimal may contain binary zeros.
826</P>
827<P>
828If <b>hex</b> or <b>use_length</b> is used with the POSIX wrapper API (see
829<a href="#posixwrapper">"Using the POSIX wrapper API"</a>
830below), the REG_PEND extension is used to pass the pattern's length.
831</P>
832<br><b>
833Specifying wide characters in 16-bit and 32-bit modes
834</b><br>
835<P>
836In 16-bit and 32-bit modes, all input is automatically treated as UTF-8 and
837translated to UTF-16 or UTF-32 when the <b>utf</b> modifier is set. For testing
838the 16-bit and 32-bit libraries in non-UTF mode, the <b>utf8_input</b> modifier
839can be used. It is mutually exclusive with <b>utf</b>. Input lines are
840interpreted as UTF-8 as a means of specifying wide characters. More details are
841given in
842<a href="#inputencoding">"Input encoding"</a>
843above.
844</P>
845<br><b>
846Generating long repetitive patterns
847</b><br>
848<P>
849Some tests use long patterns that are very repetitive. Instead of creating a
850very long input line for such a pattern, you can use a special repetition
851feature, similar to the one described for subject lines above. If the
852<b>expand</b> modifier is present on a pattern, parts of the pattern that have
853the form
854<pre>
855  \[&#60;characters&#62;]{&#60;count&#62;}
856</pre>
857are expanded before the pattern is passed to <b>pcre2_compile()</b>. For
858example, \[AB]{6000} is expanded to "ABAB..." 6000 times. This construction
859cannot be nested. An initial "\[" sequence is recognized only if "]{" followed
860by decimal digits and "}" is found later in the pattern. If not, the characters
861remain in the pattern unaltered. The <b>expand</b> and <b>hex</b> modifiers are
862mutually exclusive.
863</P>
864<P>
865If part of an expanded pattern looks like an expansion, but is really part of
866the actual pattern, unwanted expansion can be avoided by giving two values in
867the quantifier. For example, \[AB]{6000,6000} is not recognized as an
868expansion item.
869</P>
870<P>
871If the <b>info</b> modifier is set on an expanded pattern, the result of the
872expansion is included in the information that is output.
873</P>
874<br><b>
875JIT compilation
876</b><br>
877<P>
878Just-in-time (JIT) compiling is a heavyweight optimization that can greatly
879speed up pattern matching. See the
880<a href="pcre2jit.html"><b>pcre2jit</b></a>
881documentation for details. JIT compiling happens, optionally, after a pattern
882has been successfully compiled into an internal form. The JIT compiler converts
883this to optimized machine code. It needs to know whether the match-time options
884PCRE2_PARTIAL_HARD and PCRE2_PARTIAL_SOFT are going to be used, because
885different code is generated for the different cases. See the <b>partial</b>
886modifier in "Subject Modifiers"
887<a href="#subjectmodifiers">below</a>
888for details of how these options are specified for each match attempt.
889</P>
890<P>
891JIT compilation is requested by the <b>jit</b> pattern modifier, which may
892optionally be followed by an equals sign and a number in the range 0 to 7.
893The three bits that make up the number specify which of the three JIT operating
894modes are to be compiled:
895<pre>
896  1  compile JIT code for non-partial matching
897  2  compile JIT code for soft partial matching
898  4  compile JIT code for hard partial matching
899</pre>
900The possible values for the <b>jit</b> modifier are therefore:
901<pre>
902  0  disable JIT
903  1  normal matching only
904  2  soft partial matching only
905  3  normal and soft partial matching
906  4  hard partial matching only
907  6  soft and hard partial matching only
908  7  all three modes
909</pre>
910If no number is given, 7 is assumed. The phrase "partial matching" means a call
911to <b>pcre2_match()</b> with either the PCRE2_PARTIAL_SOFT or the
912PCRE2_PARTIAL_HARD option set. Note that such a call may return a complete
913match; the options enable the possibility of a partial match, but do not
914require it. Note also that if you request JIT compilation only for partial
915matching (for example, jit=2) but do not set the <b>partial</b> modifier on a
916subject line, that match will not use JIT code because none was compiled for
917non-partial matching.
918</P>
919<P>
920If JIT compilation is successful, the compiled JIT code will automatically be
921used when an appropriate type of match is run, except when incompatible
922run-time options are specified. For more details, see the
923<a href="pcre2jit.html"><b>pcre2jit</b></a>
924documentation. See also the <b>jitstack</b> modifier below for a way of
925setting the size of the JIT stack.
926</P>
927<P>
928If the <b>jitfast</b> modifier is specified, matching is done using the JIT
929"fast path" interface, <b>pcre2_jit_match()</b>, which skips some of the sanity
930checks that are done by <b>pcre2_match()</b>, and of course does not work when
931JIT is not supported. If <b>jitfast</b> is specified without <b>jit</b>, jit=7 is
932assumed.
933</P>
934<P>
935If the <b>jitverify</b> modifier is specified, information about the compiled
936pattern shows whether JIT compilation was or was not successful. If
937<b>jitverify</b> is specified without <b>jit</b>, jit=7 is assumed. If JIT
938compilation is successful when <b>jitverify</b> is set, the text "(JIT)" is
939added to the first output line after a match or non match when JIT-compiled
940code was actually used in the match.
941</P>
942<br><b>
943Setting a locale
944</b><br>
945<P>
946The <b>locale</b> modifier must specify the name of a locale, for example:
947<pre>
948  /pattern/locale=fr_FR
949</pre>
950The given locale is set, <b>pcre2_maketables()</b> is called to build a set of
951character tables for the locale, and this is then passed to
952<b>pcre2_compile()</b> when compiling the regular expression. The same tables
953are used when matching the following subject lines. The <b>locale</b> modifier
954applies only to the pattern on which it appears, but can be given in a
955<b>#pattern</b> command if a default is needed. Setting a locale and alternate
956character tables are mutually exclusive.
957</P>
958<br><b>
959Showing pattern memory
960</b><br>
961<P>
962The <b>memory</b> modifier causes the size in bytes of the memory used to hold
963the compiled pattern to be output. This does not include the size of the
964<b>pcre2_code</b> block; it is just the actual compiled data. If the pattern is
965subsequently passed to the JIT compiler, the size of the JIT compiled code is
966also output. Here is an example:
967<pre>
968    re&#62; /a(b)c/jit,memory
969  Memory allocation (code space): 21
970  Memory allocation (JIT code): 1910
971
972</PRE>
973</P>
974<br><b>
975Limiting nested parentheses
976</b><br>
977<P>
978The <b>parens_nest_limit</b> modifier sets a limit on the depth of nested
979parentheses in a pattern. Breaching the limit causes a compilation error.
980The default for the library is set when PCRE2 is built, but <b>pcre2test</b>
981sets its own default of 220, which is required for running the standard test
982suite.
983</P>
984<br><b>
985Limiting the pattern length
986</b><br>
987<P>
988The <b>max_pattern_length</b> modifier sets a limit, in code units, to the
989length of pattern that <b>pcre2_compile()</b> will accept. Breaching the limit
990causes a compilation error. The default is the largest number a PCRE2_SIZE
991variable can hold (essentially unlimited).
992<a name="posixwrapper"></a></P>
993<br><b>
994Using the POSIX wrapper API
995</b><br>
996<P>
997The <b>posix</b> and <b>posix_nosub</b> modifiers cause <b>pcre2test</b> to call
998PCRE2 via the POSIX wrapper API rather than its native API. When
999<b>posix_nosub</b> is used, the POSIX option REG_NOSUB is passed to
1000<b>regcomp()</b>. The POSIX wrapper supports only the 8-bit library. Note that
1001it does not imply POSIX matching semantics; for more detail see the
1002<a href="pcre2posix.html"><b>pcre2posix</b></a>
1003documentation. The following pattern modifiers set options for the
1004<b>regcomp()</b> function:
1005<pre>
1006  caseless           REG_ICASE
1007  multiline          REG_NEWLINE
1008  dotall             REG_DOTALL     )
1009  ungreedy           REG_UNGREEDY   ) These options are not part of
1010  ucp                REG_UCP        )   the POSIX standard
1011  utf                REG_UTF8       )
1012</pre>
1013The <b>regerror_buffsize</b> modifier specifies a size for the error buffer that
1014is passed to <b>regerror()</b> in the event of a compilation error. For example:
1015<pre>
1016  /abc/posix,regerror_buffsize=20
1017</pre>
1018This provides a means of testing the behaviour of <b>regerror()</b> when the
1019buffer is too small for the error message. If this modifier has not been set, a
1020large buffer is used.
1021</P>
1022<P>
1023The <b>aftertext</b> and <b>allaftertext</b> subject modifiers work as described
1024below. All other modifiers are either ignored, with a warning message, or cause
1025an error.
1026</P>
1027<P>
1028The pattern is passed to <b>regcomp()</b> as a zero-terminated string by
1029default, but if the <b>use_length</b> or <b>hex</b> modifiers are set, the
1030REG_PEND extension is used to pass it by length.
1031</P>
1032<br><b>
1033Testing the stack guard feature
1034</b><br>
1035<P>
1036The <b>stackguard</b> modifier is used to test the use of
1037<b>pcre2_set_compile_recursion_guard()</b>, a function that is provided to
1038enable stack availability to be checked during compilation (see the
1039<a href="pcre2api.html"><b>pcre2api</b></a>
1040documentation for details). If the number specified by the modifier is greater
1041than zero, <b>pcre2_set_compile_recursion_guard()</b> is called to set up
1042callback from <b>pcre2_compile()</b> to a local function. The argument it
1043receives is the current nesting parenthesis depth; if this is greater than the
1044value given by the modifier, non-zero is returned, causing the compilation to
1045be aborted.
1046</P>
1047<br><b>
1048Using alternative character tables
1049</b><br>
1050<P>
1051The value specified for the <b>tables</b> modifier must be one of the digits 0,
10521, 2, or 3. It causes a specific set of built-in character tables to be passed
1053to <b>pcre2_compile()</b>. This is used in the PCRE2 tests to check behaviour
1054with different character tables. The digit specifies the tables as follows:
1055<pre>
1056  0   do not pass any special character tables
1057  1   the default ASCII tables, as distributed in
1058        pcre2_chartables.c.dist
1059  2   a set of tables defining ISO 8859 characters
1060  3   a set of tables loaded by the #loadtables command
1061</pre>
1062In tables 2, some characters whose codes are greater than 128 are identified as
1063letters, digits, spaces, etc. Tables 3 can be used only after a
1064<b>#loadtables</b> command has loaded them from a binary file. Setting alternate
1065character tables and a locale are mutually exclusive.
1066</P>
1067<br><b>
1068Setting certain match controls
1069</b><br>
1070<P>
1071The following modifiers are really subject modifiers, and are described under
1072"Subject Modifiers" below. However, they may be included in a pattern's
1073modifier list, in which case they are applied to every subject line that is
1074processed with that pattern. These modifiers do not affect the compilation
1075process.
1076<pre>
1077      aftertext                   show text after match
1078      allaftertext                show text after captures
1079      allcaptures                 show all captures
1080      allvector                   show the entire ovector
1081      allusedtext                 show all consulted text
1082      altglobal                   alternative global matching
1083  /g  global                      global matching
1084      jitstack=&#60;n&#62;                set size of JIT stack
1085      mark                        show mark values
1086      replace=&#60;string&#62;            specify a replacement string
1087      startchar                   show starting character when relevant
1088      substitute_callout          use substitution callouts
1089      substitute_extended         use PCRE2_SUBSTITUTE_EXTENDED
1090      substitute_literal          use PCRE2_SUBSTITUTE_LITERAL
1091      substitute_matched          use PCRE2_SUBSTITUTE_MATCHED
1092      substitute_overflow_length  use PCRE2_SUBSTITUTE_OVERFLOW_LENGTH
1093      substitute_replacement_only use PCRE2_SUBSTITUTE_REPLACEMENT_ONLY
1094      substitute_skip=&#60;n&#62;         skip substitution &#60;n&#62;
1095      substitute_stop=&#60;n&#62;         skip substitution &#60;n&#62; and following
1096      substitute_unknown_unset    use PCRE2_SUBSTITUTE_UNKNOWN_UNSET
1097      substitute_unset_empty      use PCRE2_SUBSTITUTE_UNSET_EMPTY
1098</pre>
1099These modifiers may not appear in a <b>#pattern</b> command. If you want them as
1100defaults, set them in a <b>#subject</b> command.
1101</P>
1102<br><b>
1103Specifying literal subject lines
1104</b><br>
1105<P>
1106If the <b>subject_literal</b> modifier is present on a pattern, all the subject
1107lines that it matches are taken as literal strings, with no interpretation of
1108backslashes. It is not possible to set subject modifiers on such lines, but any
1109that are set as defaults by a <b>#subject</b> command are recognized.
1110</P>
1111<br><b>
1112Saving a compiled pattern
1113</b><br>
1114<P>
1115When a pattern with the <b>push</b> modifier is successfully compiled, it is
1116pushed onto a stack of compiled patterns, and <b>pcre2test</b> expects the next
1117line to contain a new pattern (or a command) instead of a subject line. This
1118facility is used when saving compiled patterns to a file, as described in the
1119section entitled "Saving and restoring compiled patterns"
1120<a href="#saverestore">below.</a>
1121If <b>pushcopy</b> is used instead of <b>push</b>, a copy of the compiled
1122pattern is stacked, leaving the original as current, ready to match the
1123following input lines. This provides a way of testing the
1124<b>pcre2_code_copy()</b> function.
1125The <b>push</b> and <b>pushcopy </b> modifiers are incompatible with compilation
1126modifiers such as <b>global</b> that act at match time. Any that are specified
1127are ignored (for the stacked copy), with a warning message, except for
1128<b>replace</b>, which causes an error. Note that <b>jitverify</b>, which is
1129allowed, does not carry through to any subsequent matching that uses a stacked
1130pattern.
1131</P>
1132<br><b>
1133Testing foreign pattern conversion
1134</b><br>
1135<P>
1136The experimental foreign pattern conversion functions in PCRE2 can be tested by
1137setting the <b>convert</b> modifier. Its argument is a colon-separated list of
1138options, which set the equivalent option for the <b>pcre2_pattern_convert()</b>
1139function:
1140<pre>
1141  glob                    PCRE2_CONVERT_GLOB
1142  glob_no_starstar        PCRE2_CONVERT_GLOB_NO_STARSTAR
1143  glob_no_wild_separator  PCRE2_CONVERT_GLOB_NO_WILD_SEPARATOR
1144  posix_basic             PCRE2_CONVERT_POSIX_BASIC
1145  posix_extended          PCRE2_CONVERT_POSIX_EXTENDED
1146  unset                   Unset all options
1147</pre>
1148The "unset" value is useful for turning off a default that has been set by a
1149<b>#pattern</b> command. When one of these options is set, the input pattern is
1150passed to <b>pcre2_pattern_convert()</b>. If the conversion is successful, the
1151result is reflected in the output and then passed to <b>pcre2_compile()</b>. The
1152normal <b>utf</b> and <b>no_utf_check</b> options, if set, cause the
1153PCRE2_CONVERT_UTF and PCRE2_CONVERT_NO_UTF_CHECK options to be passed to
1154<b>pcre2_pattern_convert()</b>.
1155</P>
1156<P>
1157By default, the conversion function is allowed to allocate a buffer for its
1158output. However, if the <b>convert_length</b> modifier is set to a value greater
1159than zero, <b>pcre2test</b> passes a buffer of the given length. This makes it
1160possible to test the length check.
1161</P>
1162<P>
1163The <b>convert_glob_escape</b> and <b>convert_glob_separator</b> modifiers can be
1164used to specify the escape and separator characters for glob processing,
1165overriding the defaults, which are operating-system dependent.
1166<a name="subjectmodifiers"></a></P>
1167<br><a name="SEC11" href="#TOC1">SUBJECT MODIFIERS</a><br>
1168<P>
1169The modifiers that can appear in subject lines and the <b>#subject</b>
1170command are of two types.
1171</P>
1172<br><b>
1173Setting match options
1174</b><br>
1175<P>
1176The following modifiers set options for <b>pcre2_match()</b> or
1177<b>pcre2_dfa_match()</b>. See
1178<a href="pcreapi.html"><b>pcreapi</b></a>
1179for a description of their effects.
1180<pre>
1181      anchored                  set PCRE2_ANCHORED
1182      endanchored               set PCRE2_ENDANCHORED
1183      dfa_restart               set PCRE2_DFA_RESTART
1184      dfa_shortest              set PCRE2_DFA_SHORTEST
1185      no_jit                    set PCRE2_NO_JIT
1186      no_utf_check              set PCRE2_NO_UTF_CHECK
1187      notbol                    set PCRE2_NOTBOL
1188      notempty                  set PCRE2_NOTEMPTY
1189      notempty_atstart          set PCRE2_NOTEMPTY_ATSTART
1190      noteol                    set PCRE2_NOTEOL
1191      partial_hard (or ph)      set PCRE2_PARTIAL_HARD
1192      partial_soft (or ps)      set PCRE2_PARTIAL_SOFT
1193</pre>
1194The partial matching modifiers are provided with abbreviations because they
1195appear frequently in tests.
1196</P>
1197<P>
1198If the <b>posix</b> or <b>posix_nosub</b> modifier was present on the pattern,
1199causing the POSIX wrapper API to be used, the only option-setting modifiers
1200that have any effect are <b>notbol</b>, <b>notempty</b>, and <b>noteol</b>,
1201causing REG_NOTBOL, REG_NOTEMPTY, and REG_NOTEOL, respectively, to be passed to
1202<b>regexec()</b>. The other modifiers are ignored, with a warning message.
1203</P>
1204<P>
1205There is one additional modifier that can be used with the POSIX wrapper. It is
1206ignored (with a warning) if used for non-POSIX matching.
1207<pre>
1208      posix_startend=&#60;n&#62;[:&#60;m&#62;]
1209</pre>
1210This causes the subject string to be passed to <b>regexec()</b> using the
1211REG_STARTEND option, which uses offsets to specify which part of the string is
1212searched. If only one number is given, the end offset is passed as the end of
1213the subject string. For more detail of REG_STARTEND, see the
1214<a href="pcre2posix.html"><b>pcre2posix</b></a>
1215documentation. If the subject string contains binary zeros (coded as escapes
1216such as \x{00} because <b>pcre2test</b> does not support actual binary zeros in
1217its input), you must use <b>posix_startend</b> to specify its length.
1218</P>
1219<br><b>
1220Setting match controls
1221</b><br>
1222<P>
1223The following modifiers affect the matching process or request additional
1224information. Some of them may also be specified on a pattern line (see above),
1225in which case they apply to every subject line that is matched against that
1226pattern, but can be overridden by modifiers on the subject.
1227<pre>
1228      aftertext                  show text after match
1229      allaftertext               show text after captures
1230      allcaptures                show all captures
1231      allvector                  show the entire ovector
1232      allusedtext                show all consulted text (non-JIT only)
1233      altglobal                  alternative global matching
1234      callout_capture            show captures at callout time
1235      callout_data=&#60;n&#62;           set a value to pass via callouts
1236      callout_error=&#60;n&#62;[:&#60;m&#62;]    control callout error
1237      callout_extra              show extra callout information
1238      callout_fail=&#60;n&#62;[:&#60;m&#62;]     control callout failure
1239      callout_no_where           do not show position of a callout
1240      callout_none               do not supply a callout function
1241      copy=&#60;number or name&#62;      copy captured substring
1242      depth_limit=&#60;n&#62;            set a depth limit
1243      dfa                        use <b>pcre2_dfa_match()</b>
1244      find_limits                find match and depth limits
1245      get=&#60;number or name&#62;       extract captured substring
1246      getall                     extract all captured substrings
1247  /g  global                     global matching
1248      heap_limit=&#60;n&#62;             set a limit on heap memory (Kbytes)
1249      jitstack=&#60;n&#62;               set size of JIT stack
1250      mark                       show mark values
1251      match_limit=&#60;n&#62;            set a match limit
1252      memory                     show heap memory usage
1253      null_context               match with a NULL context
1254      null_replacement           substitute with NULL replacement
1255      null_subject               match with NULL subject
1256      offset=&#60;n&#62;                 set starting offset
1257      offset_limit=&#60;n&#62;           set offset limit
1258      ovector=&#60;n&#62;                set size of output vector
1259      recursion_limit=&#60;n&#62;        obsolete synonym for depth_limit
1260      replace=&#60;string&#62;           specify a replacement string
1261      startchar                  show startchar when relevant
1262      startoffset=&#60;n&#62;            same as offset=&#60;n&#62;
1263      substitute_callout         use substitution callouts
1264      substitute_extedded        use PCRE2_SUBSTITUTE_EXTENDED
1265      substitute_literal         use PCRE2_SUBSTITUTE_LITERAL
1266      substitute_matched         use PCRE2_SUBSTITUTE_MATCHED
1267      substitute_overflow_length use PCRE2_SUBSTITUTE_OVERFLOW_LENGTH
1268      substitute_replacement_only use PCRE2_SUBSTITUTE_REPLACEMENT_ONLY
1269      substitute_skip=&#60;n&#62;        skip substitution number n
1270      substitute_stop=&#60;n&#62;        skip substitution number n and greater
1271      substitute_unknown_unset   use PCRE2_SUBSTITUTE_UNKNOWN_UNSET
1272      substitute_unset_empty     use PCRE2_SUBSTITUTE_UNSET_EMPTY
1273      zero_terminate             pass the subject as zero-terminated
1274</pre>
1275The effects of these modifiers are described in the following sections. When
1276matching via the POSIX wrapper API, the <b>aftertext</b>, <b>allaftertext</b>,
1277and <b>ovector</b> subject modifiers work as described below. All other
1278modifiers are either ignored, with a warning message, or cause an error.
1279</P>
1280<br><b>
1281Showing more text
1282</b><br>
1283<P>
1284The <b>aftertext</b> modifier requests that as well as outputting the part of
1285the subject string that matched the entire pattern, <b>pcre2test</b> should in
1286addition output the remainder of the subject string. This is useful for tests
1287where the subject contains multiple copies of the same substring. The
1288<b>allaftertext</b> modifier requests the same action for captured substrings as
1289well as the main matched substring. In each case the remainder is output on the
1290following line with a plus character following the capture number.
1291</P>
1292<P>
1293The <b>allusedtext</b> modifier requests that all the text that was consulted
1294during a successful pattern match by the interpreter should be shown, for both
1295full and partial matches. This feature is not supported for JIT matching, and
1296if requested with JIT it is ignored (with a warning message). Setting this
1297modifier affects the output if there is a lookbehind at the start of a match,
1298or, for a complete match, a lookahead at the end, or if \K is used in the
1299pattern. Characters that precede or follow the start and end of the actual
1300match are indicated in the output by '&#60;' or '&#62;' characters underneath them.
1301Here is an example:
1302<pre>
1303    re&#62; /(?&#60;=pqr)abc(?=xyz)/
1304  data&#62; 123pqrabcxyz456\=allusedtext
1305   0: pqrabcxyz
1306      &#60;&#60;&#60;   &#62;&#62;&#62;
1307  data&#62; 123pqrabcxy\=ph,allusedtext
1308  Partial match: pqrabcxy
1309                 &#60;&#60;&#60;
1310</pre>
1311The first, complete match shows that the matched string is "abc", with the
1312preceding and following strings "pqr" and "xyz" having been consulted during
1313the match (when processing the assertions). The partial match can indicate only
1314the preceding string.
1315</P>
1316<P>
1317The <b>startchar</b> modifier requests that the starting character for the match
1318be indicated, if it is different to the start of the matched string. The only
1319time when this occurs is when \K has been processed as part of the match. In
1320this situation, the output for the matched string is displayed from the
1321starting character instead of from the match point, with circumflex characters
1322under the earlier characters. For example:
1323<pre>
1324    re&#62; /abc\Kxyz/
1325  data&#62; abcxyz\=startchar
1326   0: abcxyz
1327      ^^^
1328</pre>
1329Unlike <b>allusedtext</b>, the <b>startchar</b> modifier can be used with JIT.
1330However, these two modifiers are mutually exclusive.
1331</P>
1332<br><b>
1333Showing the value of all capture groups
1334</b><br>
1335<P>
1336The <b>allcaptures</b> modifier requests that the values of all potential
1337captured parentheses be output after a match. By default, only those up to the
1338highest one actually used in the match are output (corresponding to the return
1339code from <b>pcre2_match()</b>). Groups that did not take part in the match
1340are output as "&#60;unset&#62;". This modifier is not relevant for DFA matching (which
1341does no capturing) and does not apply when <b>replace</b> is specified; it is
1342ignored, with a warning message, if present.
1343</P>
1344<br><b>
1345Showing the entire ovector, for all outcomes
1346</b><br>
1347<P>
1348The <b>allvector</b> modifier requests that the entire ovector be shown,
1349whatever the outcome of the match. Compare <b>allcaptures</b>, which shows only
1350up to the maximum number of capture groups for the pattern, and then only for a
1351successful complete non-DFA match. This modifier, which acts after any match
1352result, and also for DFA matching, provides a means of checking that there are
1353no unexpected modifications to ovector fields. Before each match attempt, the
1354ovector is filled with a special value, and if this is found in both elements
1355of a capturing pair, "&#60;unchanged&#62;" is output. After a successful match, this
1356applies to all groups after the maximum capture group for the pattern. In other
1357cases it applies to the entire ovector. After a partial match, the first two
1358elements are the only ones that should be set. After a DFA match, the amount of
1359ovector that is used depends on the number of matches that were found.
1360</P>
1361<br><b>
1362Testing pattern callouts
1363</b><br>
1364<P>
1365A callout function is supplied when <b>pcre2test</b> calls the library matching
1366functions, unless <b>callout_none</b> is specified. Its behaviour can be
1367controlled by various modifiers listed above whose names begin with
1368<b>callout_</b>. Details are given in the section entitled "Callouts"
1369<a href="#callouts">below.</a>
1370Testing callouts from <b>pcre2_substitute()</b> is decribed separately in
1371"Testing the substitution function"
1372<a href="#substitution">below.</a>
1373</P>
1374<br><b>
1375Finding all matches in a string
1376</b><br>
1377<P>
1378Searching for all possible matches within a subject can be requested by the
1379<b>global</b> or <b>altglobal</b> modifier. After finding a match, the matching
1380function is called again to search the remainder of the subject. The difference
1381between <b>global</b> and <b>altglobal</b> is that the former uses the
1382<i>start_offset</i> argument to <b>pcre2_match()</b> or <b>pcre2_dfa_match()</b>
1383to start searching at a new point within the entire string (which is what Perl
1384does), whereas the latter passes over a shortened subject. This makes a
1385difference to the matching process if the pattern begins with a lookbehind
1386assertion (including \b or \B).
1387</P>
1388<P>
1389If an empty string is matched, the next match is done with the
1390PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED flags set, in order to search for
1391another, non-empty, match at the same point in the subject. If this match
1392fails, the start offset is advanced, and the normal match is retried. This
1393imitates the way Perl handles such cases when using the <b>/g</b> modifier or
1394the <b>split()</b> function. Normally, the start offset is advanced by one
1395character, but if the newline convention recognizes CRLF as a newline, and the
1396current character is CR followed by LF, an advance of two characters occurs.
1397</P>
1398<br><b>
1399Testing substring extraction functions
1400</b><br>
1401<P>
1402The <b>copy</b> and <b>get</b> modifiers can be used to test the
1403<b>pcre2_substring_copy_xxx()</b> and <b>pcre2_substring_get_xxx()</b> functions.
1404They can be given more than once, and each can specify a capture group name or
1405number, for example:
1406<pre>
1407   abcd\=copy=1,copy=3,get=G1
1408</pre>
1409If the <b>#subject</b> command is used to set default copy and/or get lists,
1410these can be unset by specifying a negative number to cancel all numbered
1411groups and an empty name to cancel all named groups.
1412</P>
1413<P>
1414The <b>getall</b> modifier tests <b>pcre2_substring_list_get()</b>, which
1415extracts all captured substrings.
1416</P>
1417<P>
1418If the subject line is successfully matched, the substrings extracted by the
1419convenience functions are output with C, G, or L after the string number
1420instead of a colon. This is in addition to the normal full list. The string
1421length (that is, the return from the extraction function) is given in
1422parentheses after each substring, followed by the name when the extraction was
1423by name.
1424<a name="substitution"></a></P>
1425<br><b>
1426Testing the substitution function
1427</b><br>
1428<P>
1429If the <b>replace</b> modifier is set, the <b>pcre2_substitute()</b> function is
1430called instead of one of the matching functions (or after one call of
1431<b>pcre2_match()</b> in the case of PCRE2_SUBSTITUTE_MATCHED). Note that
1432replacement strings cannot contain commas, because a comma signifies the end of
1433a modifier. This is not thought to be an issue in a test program.
1434</P>
1435<P>
1436Specifying a completely empty replacement string disables this modifier.
1437However, it is possible to specify an empty replacement by providing a buffer
1438length, as described below, for an otherwise empty replacement.
1439</P>
1440<P>
1441Unlike subject strings, <b>pcre2test</b> does not process replacement strings
1442for escape sequences. In UTF mode, a replacement string is checked to see if it
1443is a valid UTF-8 string. If so, it is correctly converted to a UTF string of
1444the appropriate code unit width. If it is not a valid UTF-8 string, the
1445individual code units are copied directly. This provides a means of passing an
1446invalid UTF-8 string for testing purposes.
1447</P>
1448<P>
1449The following modifiers set options (in additional to the normal match options)
1450for <b>pcre2_substitute()</b>:
1451<pre>
1452  global                      PCRE2_SUBSTITUTE_GLOBAL
1453  substitute_extended         PCRE2_SUBSTITUTE_EXTENDED
1454  substitute_literal          PCRE2_SUBSTITUTE_LITERAL
1455  substitute_matched          PCRE2_SUBSTITUTE_MATCHED
1456  substitute_overflow_length  PCRE2_SUBSTITUTE_OVERFLOW_LENGTH
1457  substitute_replacement_only PCRE2_SUBSTITUTE_REPLACEMENT_ONLY
1458  substitute_unknown_unset    PCRE2_SUBSTITUTE_UNKNOWN_UNSET
1459  substitute_unset_empty      PCRE2_SUBSTITUTE_UNSET_EMPTY
1460</pre>
1461See the
1462<a href="pcre2api.html"><b>pcre2api</b></a>
1463documentation for details of these options.
1464</P>
1465<P>
1466After a successful substitution, the modified string is output, preceded by the
1467number of replacements. This may be zero if there were no matches. Here is a
1468simple example of a substitution test:
1469<pre>
1470  /abc/replace=xxx
1471      =abc=abc=
1472   1: =xxx=abc=
1473      =abc=abc=\=global
1474   2: =xxx=xxx=
1475</pre>
1476Subject and replacement strings should be kept relatively short (fewer than 256
1477characters) for substitution tests, as fixed-size buffers are used. To make it
1478easy to test for buffer overflow, if the replacement string starts with a
1479number in square brackets, that number is passed to <b>pcre2_substitute()</b> as
1480the size of the output buffer, with the replacement string starting at the next
1481character. Here is an example that tests the edge case:
1482<pre>
1483  /abc/
1484      123abc123\=replace=[10]XYZ
1485   1: 123XYZ123
1486      123abc123\=replace=[9]XYZ
1487  Failed: error -47: no more memory
1488</pre>
1489The default action of <b>pcre2_substitute()</b> is to return
1490PCRE2_ERROR_NOMEMORY when the output buffer is too small. However, if the
1491PCRE2_SUBSTITUTE_OVERFLOW_LENGTH option is set (by using the
1492<b>substitute_overflow_length</b> modifier), <b>pcre2_substitute()</b> continues
1493to go through the motions of matching and substituting (but not doing any
1494callouts), in order to compute the size of buffer that is required. When this
1495happens, <b>pcre2test</b> shows the required buffer length (which includes space
1496for the trailing zero) as part of the error message. For example:
1497<pre>
1498  /abc/substitute_overflow_length
1499      123abc123\=replace=[9]XYZ
1500  Failed: error -47: no more memory: 10 code units are needed
1501</pre>
1502A replacement string is ignored with POSIX and DFA matching. Specifying partial
1503matching provokes an error return ("bad option value") from
1504<b>pcre2_substitute()</b>.
1505</P>
1506<br><b>
1507Testing substitute callouts
1508</b><br>
1509<P>
1510If the <b>substitute_callout</b> modifier is set, a substitution callout
1511function is set up. The <b>null_context</b> modifier must not be set, because
1512the address of the callout function is passed in a match context. When the
1513callout function is called (after each substitution), details of the the input
1514and output strings are output. For example:
1515<pre>
1516  /abc/g,replace=&#60;$0&#62;,substitute_callout
1517      abcdefabcpqr
1518   1(1) Old 0 3 "abc" New 0 5 "&#60;abc&#62;"
1519   2(1) Old 6 9 "abc" New 8 13 "&#60;abc&#62;"
1520   2: &#60;abc&#62;def&#60;abc&#62;pqr
1521</pre>
1522The first number on each callout line is the count of matches. The
1523parenthesized number is the number of pairs that are set in the ovector (that
1524is, one more than the number of capturing groups that were set). Then are
1525listed the offsets of the old substring, its contents, and the same for the
1526replacement.
1527</P>
1528<P>
1529By default, the substitution callout function returns zero, which accepts the
1530replacement and causes matching to continue if /g was used. Two further
1531modifiers can be used to test other return values. If <b>substitute_skip</b> is
1532set to a value greater than zero the callout function returns +1 for the match
1533of that number, and similarly <b>substitute_stop</b> returns -1. These cause the
1534replacement to be rejected, and -1 causes no further matching to take place. If
1535either of them are set, <b>substitute_callout</b> is assumed. For example:
1536<pre>
1537  /abc/g,replace=&#60;$0&#62;,substitute_skip=1
1538      abcdefabcpqr
1539   1(1) Old 0 3 "abc" New 0 5 "&#60;abc&#62; SKIPPED"
1540   2(1) Old 6 9 "abc" New 6 11 "&#60;abc&#62;"
1541   2: abcdef&#60;abc&#62;pqr
1542      abcdefabcpqr\=substitute_stop=1
1543   1(1) Old 0 3 "abc" New 0 5 "&#60;abc&#62; STOPPED"
1544   1: abcdefabcpqr
1545</pre>
1546If both are set for the same number, stop takes precedence. Only a single skip
1547or stop is supported, which is sufficient for testing that the feature works.
1548</P>
1549<br><b>
1550Setting the JIT stack size
1551</b><br>
1552<P>
1553The <b>jitstack</b> modifier provides a way of setting the maximum stack size
1554that is used by the just-in-time optimization code. It is ignored if JIT
1555optimization is not being used. The value is a number of kibibytes (units of
15561024 bytes). Setting zero reverts to the default of 32KiB. Providing a stack
1557that is larger than the default is necessary only for very complicated
1558patterns. If <b>jitstack</b> is set non-zero on a subject line it overrides any
1559value that was set on the pattern.
1560</P>
1561<br><b>
1562Setting heap, match, and depth limits
1563</b><br>
1564<P>
1565The <b>heap_limit</b>, <b>match_limit</b>, and <b>depth_limit</b> modifiers set
1566the appropriate limits in the match context. These values are ignored when the
1567<b>find_limits</b> modifier is specified.
1568</P>
1569<br><b>
1570Finding minimum limits
1571</b><br>
1572<P>
1573If the <b>find_limits</b> modifier is present on a subject line, <b>pcre2test</b>
1574calls the relevant matching function several times, setting different values in
1575the match context via <b>pcre2_set_heap_limit()</b>,
1576<b>pcre2_set_match_limit()</b>, or <b>pcre2_set_depth_limit()</b> until it finds
1577the minimum values for each parameter that allows the match to complete without
1578error. If JIT is being used, only the match limit is relevant.
1579</P>
1580<P>
1581When using this modifier, the pattern should not contain any limit settings
1582such as (*LIMIT_MATCH=...) within it. If such a setting is present and is
1583lower than the minimum matching value, the minimum value cannot be found
1584because <b>pcre2_set_match_limit()</b> etc. are only able to reduce the value of
1585an in-pattern limit; they cannot increase it.
1586</P>
1587<P>
1588For non-DFA matching, the minimum <i>depth_limit</i> number is a measure of how
1589much nested backtracking happens (that is, how deeply the pattern's tree is
1590searched). In the case of DFA matching, <i>depth_limit</i> controls the depth of
1591recursive calls of the internal function that is used for handling pattern
1592recursion, lookaround assertions, and atomic groups.
1593</P>
1594<P>
1595For non-DFA matching, the <i>match_limit</i> number is a measure of the amount
1596of backtracking that takes place, and learning the minimum value can be
1597instructive. For most simple matches, the number is quite small, but for
1598patterns with very large numbers of matching possibilities, it can become large
1599very quickly with increasing length of subject string. In the case of DFA
1600matching, <i>match_limit</i> controls the total number of calls, both recursive
1601and non-recursive, to the internal matching function, thus controlling the
1602overall amount of computing resource that is used.
1603</P>
1604<P>
1605For both kinds of matching, the <i>heap_limit</i> number, which is in kibibytes
1606(units of 1024 bytes), limits the amount of heap memory used for matching. A
1607value of zero disables the use of any heap memory; many simple pattern matches
1608can be done without using the heap, so zero is not an unreasonable setting.
1609</P>
1610<br><b>
1611Showing MARK names
1612</b><br>
1613<P>
1614The <b>mark</b> modifier causes the names from backtracking control verbs that
1615are returned from calls to <b>pcre2_match()</b> to be displayed. If a mark is
1616returned for a match, non-match, or partial match, <b>pcre2test</b> shows it.
1617For a match, it is on a line by itself, tagged with "MK:". Otherwise, it
1618is added to the non-match message.
1619</P>
1620<br><b>
1621Showing memory usage
1622</b><br>
1623<P>
1624The <b>memory</b> modifier causes <b>pcre2test</b> to log the sizes of all heap
1625memory allocation and freeing calls that occur during a call to
1626<b>pcre2_match()</b> or <b>pcre2_dfa_match()</b>. These occur only when a match
1627requires a bigger vector than the default for remembering backtracking points
1628(<b>pcre2_match()</b>) or for internal workspace (<b>pcre2_dfa_match()</b>). In
1629many cases there will be no heap memory used and therefore no additional
1630output. No heap memory is allocated during matching with JIT, so in that case
1631the <b>memory</b> modifier never has any effect. For this modifier to work, the
1632<b>null_context</b> modifier must not be set on both the pattern and the
1633subject, though it can be set on one or the other.
1634</P>
1635<br><b>
1636Setting a starting offset
1637</b><br>
1638<P>
1639The <b>offset</b> modifier sets an offset in the subject string at which
1640matching starts. Its value is a number of code units, not characters.
1641</P>
1642<br><b>
1643Setting an offset limit
1644</b><br>
1645<P>
1646The <b>offset_limit</b> modifier sets a limit for unanchored matches. If a match
1647cannot be found starting at or before this offset in the subject, a "no match"
1648return is given. The data value is a number of code units, not characters. When
1649this modifier is used, the <b>use_offset_limit</b> modifier must have been set
1650for the pattern; if not, an error is generated.
1651</P>
1652<br><b>
1653Setting the size of the output vector
1654</b><br>
1655<P>
1656The <b>ovector</b> modifier applies only to the subject line in which it
1657appears, though of course it can also be used to set a default in a
1658<b>#subject</b> command. It specifies the number of pairs of offsets that are
1659available for storing matching information. The default is 15.
1660</P>
1661<P>
1662A value of zero is useful when testing the POSIX API because it causes
1663<b>regexec()</b> to be called with a NULL capture vector. When not testing the
1664POSIX API, a value of zero is used to cause
1665<b>pcre2_match_data_create_from_pattern()</b> to be called, in order to create a
1666match block of exactly the right size for the pattern. (It is not possible to
1667create a match block with a zero-length ovector; there is always at least one
1668pair of offsets.)
1669</P>
1670<br><b>
1671Passing the subject as zero-terminated
1672</b><br>
1673<P>
1674By default, the subject string is passed to a native API matching function with
1675its correct length. In order to test the facility for passing a zero-terminated
1676string, the <b>zero_terminate</b> modifier is provided. It causes the length to
1677be passed as PCRE2_ZERO_TERMINATED. When matching via the POSIX interface,
1678this modifier is ignored, with a warning.
1679</P>
1680<P>
1681When testing <b>pcre2_substitute()</b>, this modifier also has the effect of
1682passing the replacement string as zero-terminated.
1683</P>
1684<br><b>
1685Passing a NULL context, subject, or replacement
1686</b><br>
1687<P>
1688Normally, <b>pcre2test</b> passes a context block to <b>pcre2_match()</b>,
1689<b>pcre2_dfa_match()</b>, <b>pcre2_jit_match()</b> or <b>pcre2_substitute()</b>.
1690If the <b>null_context</b> modifier is set, however, NULL is passed. This is for
1691testing that the matching and substitution functions behave correctly in this
1692case (they use default values). This modifier cannot be used with the
1693<b>find_limits</b> or <b>substitute_callout</b> modifiers.
1694</P>
1695<P>
1696Similarly, for testing purposes, if the <b>null_subject</b> or
1697<b>null_replacement</b> modifier is set, the subject or replacement string
1698pointers are passed as NULL, respectively, to the relevant functions.
1699</P>
1700<br><a name="SEC12" href="#TOC1">THE ALTERNATIVE MATCHING FUNCTION</a><br>
1701<P>
1702By default, <b>pcre2test</b> uses the standard PCRE2 matching function,
1703<b>pcre2_match()</b> to match each subject line. PCRE2 also supports an
1704alternative matching function, <b>pcre2_dfa_match()</b>, which operates in a
1705different way, and has some restrictions. The differences between the two
1706functions are described in the
1707<a href="pcre2matching.html"><b>pcre2matching</b></a>
1708documentation.
1709</P>
1710<P>
1711If the <b>dfa</b> modifier is set, the alternative matching function is used.
1712This function finds all possible matches at a given point in the subject. If,
1713however, the <b>dfa_shortest</b> modifier is set, processing stops after the
1714first match is found. This is always the shortest possible match.
1715</P>
1716<br><a name="SEC13" href="#TOC1">DEFAULT OUTPUT FROM pcre2test</a><br>
1717<P>
1718This section describes the output when the normal matching function,
1719<b>pcre2_match()</b>, is being used.
1720</P>
1721<P>
1722When a match succeeds, <b>pcre2test</b> outputs the list of captured substrings,
1723starting with number 0 for the string that matched the whole pattern.
1724Otherwise, it outputs "No match" when the return is PCRE2_ERROR_NOMATCH, or
1725"Partial match:" followed by the partially matching substring when the
1726return is PCRE2_ERROR_PARTIAL. (Note that this is the
1727entire substring that was inspected during the partial match; it may include
1728characters before the actual match start if a lookbehind assertion, \K, \b,
1729or \B was involved.)
1730</P>
1731<P>
1732For any other return, <b>pcre2test</b> outputs the PCRE2 negative error number
1733and a short descriptive phrase. If the error is a failed UTF string check, the
1734code unit offset of the start of the failing character is also output. Here is
1735an example of an interactive <b>pcre2test</b> run.
1736<pre>
1737  $ pcre2test
1738  PCRE2 version 10.22 2016-07-29
1739
1740    re&#62; /^abc(\d+)/
1741  data&#62; abc123
1742   0: abc123
1743   1: 123
1744  data&#62; xyz
1745  No match
1746</pre>
1747Unset capturing substrings that are not followed by one that is set are not
1748shown by <b>pcre2test</b> unless the <b>allcaptures</b> modifier is specified. In
1749the following example, there are two capturing substrings, but when the first
1750data line is matched, the second, unset substring is not shown. An "internal"
1751unset substring is shown as "&#60;unset&#62;", as for the second data line.
1752<pre>
1753    re&#62; /(a)|(b)/
1754  data&#62; a
1755   0: a
1756   1: a
1757  data&#62; b
1758   0: b
1759   1: &#60;unset&#62;
1760   2: b
1761</pre>
1762If the strings contain any non-printing characters, they are output as \xhh
1763escapes if the value is less than 256 and UTF mode is not set. Otherwise they
1764are output as \x{hh...} escapes. See below for the definition of non-printing
1765characters. If the <b>aftertext</b> modifier is set, the output for substring
17660 is followed by the the rest of the subject string, identified by "0+" like
1767this:
1768<pre>
1769    re&#62; /cat/aftertext
1770  data&#62; cataract
1771   0: cat
1772   0+ aract
1773</pre>
1774If global matching is requested, the results of successive matching attempts
1775are output in sequence, like this:
1776<pre>
1777    re&#62; /\Bi(\w\w)/g
1778  data&#62; Mississippi
1779   0: iss
1780   1: ss
1781   0: iss
1782   1: ss
1783   0: ipp
1784   1: pp
1785</pre>
1786"No match" is output only if the first match attempt fails. Here is an example
1787of a failure message (the offset 4 that is specified by the <b>offset</b>
1788modifier is past the end of the subject string):
1789<pre>
1790    re&#62; /xyz/
1791  data&#62; xyz\=offset=4
1792  Error -24 (bad offset value)
1793</PRE>
1794</P>
1795<P>
1796Note that whereas patterns can be continued over several lines (a plain "&#62;"
1797prompt is used for continuations), subject lines may not. However newlines can
1798be included in a subject by means of the \n escape (or \r, \r\n, etc.,
1799depending on the newline sequence setting).
1800</P>
1801<br><a name="SEC14" href="#TOC1">OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION</a><br>
1802<P>
1803When the alternative matching function, <b>pcre2_dfa_match()</b>, is used, the
1804output consists of a list of all the matches that start at the first point in
1805the subject where there is at least one match. For example:
1806<pre>
1807    re&#62; /(tang|tangerine|tan)/
1808  data&#62; yellow tangerine\=dfa
1809   0: tangerine
1810   1: tang
1811   2: tan
1812</pre>
1813Using the normal matching function on this data finds only "tang". The
1814longest matching string is always given first (and numbered zero). After a
1815PCRE2_ERROR_PARTIAL return, the output is "Partial match:", followed by the
1816partially matching substring. Note that this is the entire substring that was
1817inspected during the partial match; it may include characters before the actual
1818match start if a lookbehind assertion, \b, or \B was involved. (\K is not
1819supported for DFA matching.)
1820</P>
1821<P>
1822If global matching is requested, the search for further matches resumes
1823at the end of the longest match. For example:
1824<pre>
1825    re&#62; /(tang|tangerine|tan)/g
1826  data&#62; yellow tangerine and tangy sultana\=dfa
1827   0: tangerine
1828   1: tang
1829   2: tan
1830   0: tang
1831   1: tan
1832   0: tan
1833</pre>
1834The alternative matching function does not support substring capture, so the
1835modifiers that are concerned with captured substrings are not relevant.
1836</P>
1837<br><a name="SEC15" href="#TOC1">RESTARTING AFTER A PARTIAL MATCH</a><br>
1838<P>
1839When the alternative matching function has given the PCRE2_ERROR_PARTIAL
1840return, indicating that the subject partially matched the pattern, you can
1841restart the match with additional subject data by means of the
1842<b>dfa_restart</b> modifier. For example:
1843<pre>
1844    re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
1845  data&#62; 23ja\=ps,dfa
1846  Partial match: 23ja
1847  data&#62; n05\=dfa,dfa_restart
1848   0: n05
1849</pre>
1850For further information about partial matching, see the
1851<a href="pcre2partial.html"><b>pcre2partial</b></a>
1852documentation.
1853<a name="callouts"></a></P>
1854<br><a name="SEC16" href="#TOC1">CALLOUTS</a><br>
1855<P>
1856If the pattern contains any callout requests, <b>pcre2test</b>'s callout
1857function is called during matching unless <b>callout_none</b> is specified. This
1858works with both matching functions, and with JIT, though there are some
1859differences in behaviour. The output for callouts with numerical arguments and
1860those with string arguments is slightly different.
1861</P>
1862<br><b>
1863Callouts with numerical arguments
1864</b><br>
1865<P>
1866By default, the callout function displays the callout number, the start and
1867current positions in the subject text at the callout time, and the next pattern
1868item to be tested. For example:
1869<pre>
1870  ---&#62;pqrabcdef
1871    0    ^  ^     \d
1872</pre>
1873This output indicates that callout number 0 occurred for a match attempt
1874starting at the fourth character of the subject string, when the pointer was at
1875the seventh character, and when the next pattern item was \d. Just
1876one circumflex is output if the start and current positions are the same, or if
1877the current position precedes the start position, which can happen if the
1878callout is in a lookbehind assertion.
1879</P>
1880<P>
1881Callouts numbered 255 are assumed to be automatic callouts, inserted as a
1882result of the <b>auto_callout</b> pattern modifier. In this case, instead of
1883showing the callout number, the offset in the pattern, preceded by a plus, is
1884output. For example:
1885<pre>
1886    re&#62; /\d?[A-E]\*/auto_callout
1887  data&#62; E*
1888  ---&#62;E*
1889   +0 ^      \d?
1890   +3 ^      [A-E]
1891   +8 ^^     \*
1892  +10 ^ ^
1893   0: E*
1894</pre>
1895If a pattern contains (*MARK) items, an additional line is output whenever
1896a change of latest mark is passed to the callout function. For example:
1897<pre>
1898    re&#62; /a(*MARK:X)bc/auto_callout
1899  data&#62; abc
1900  ---&#62;abc
1901   +0 ^       a
1902   +1 ^^      (*MARK:X)
1903  +10 ^^      b
1904  Latest Mark: X
1905  +11 ^ ^     c
1906  +12 ^  ^
1907   0: abc
1908</pre>
1909The mark changes between matching "a" and "b", but stays the same for the rest
1910of the match, so nothing more is output. If, as a result of backtracking, the
1911mark reverts to being unset, the text "&#60;unset&#62;" is output.
1912</P>
1913<br><b>
1914Callouts with string arguments
1915</b><br>
1916<P>
1917The output for a callout with a string argument is similar, except that instead
1918of outputting a callout number before the position indicators, the callout
1919string and its offset in the pattern string are output before the reflection of
1920the subject string, and the subject string is reflected for each callout. For
1921example:
1922<pre>
1923    re&#62; /^ab(?C'first')cd(?C"second")ef/
1924  data&#62; abcdefg
1925  Callout (7): 'first'
1926  ---&#62;abcdefg
1927      ^ ^         c
1928  Callout (20): "second"
1929  ---&#62;abcdefg
1930      ^   ^       e
1931   0: abcdef
1932
1933</PRE>
1934</P>
1935<br><b>
1936Callout modifiers
1937</b><br>
1938<P>
1939The callout function in <b>pcre2test</b> returns zero (carry on matching) by
1940default, but you can use a <b>callout_fail</b> modifier in a subject line to
1941change this and other parameters of the callout (see below).
1942</P>
1943<P>
1944If the <b>callout_capture</b> modifier is set, the current captured groups are
1945output when a callout occurs. This is useful only for non-DFA matching, as
1946<b>pcre2_dfa_match()</b> does not support capturing, so no captures are ever
1947shown.
1948</P>
1949<P>
1950The normal callout output, showing the callout number or pattern offset (as
1951described above) is suppressed if the <b>callout_no_where</b> modifier is set.
1952</P>
1953<P>
1954When using the interpretive matching function <b>pcre2_match()</b> without JIT,
1955setting the <b>callout_extra</b> modifier causes additional output from
1956<b>pcre2test</b>'s callout function to be generated. For the first callout in a
1957match attempt at a new starting position in the subject, "New match attempt" is
1958output. If there has been a backtrack since the last callout (or start of
1959matching if this is the first callout), "Backtrack" is output, followed by "No
1960other matching paths" if the backtrack ended the previous match attempt. For
1961example:
1962<pre>
1963   re&#62; /(a+)b/auto_callout,no_start_optimize,no_auto_possess
1964  data&#62; aac\=callout_extra
1965  New match attempt
1966  ---&#62;aac
1967   +0 ^       (
1968   +1 ^       a+
1969   +3 ^ ^     )
1970   +4 ^ ^     b
1971  Backtrack
1972  ---&#62;aac
1973   +3 ^^      )
1974   +4 ^^      b
1975  Backtrack
1976  No other matching paths
1977  New match attempt
1978  ---&#62;aac
1979   +0  ^      (
1980   +1  ^      a+
1981   +3  ^^     )
1982   +4  ^^     b
1983  Backtrack
1984  No other matching paths
1985  New match attempt
1986  ---&#62;aac
1987   +0   ^     (
1988   +1   ^     a+
1989  Backtrack
1990  No other matching paths
1991  New match attempt
1992  ---&#62;aac
1993   +0    ^    (
1994   +1    ^    a+
1995  No match
1996</pre>
1997Notice that various optimizations must be turned off if you want all possible
1998matching paths to be scanned. If <b>no_start_optimize</b> is not used, there is
1999an immediate "no match", without any callouts, because the starting
2000optimization fails to find "b" in the subject, which it knows must be present
2001for any match. If <b>no_auto_possess</b> is not used, the "a+" item is turned
2002into "a++", which reduces the number of backtracks.
2003</P>
2004<P>
2005The <b>callout_extra</b> modifier has no effect if used with the DFA matching
2006function, or with JIT.
2007</P>
2008<br><b>
2009Return values from callouts
2010</b><br>
2011<P>
2012The default return from the callout function is zero, which allows matching to
2013continue. The <b>callout_fail</b> modifier can be given one or two numbers. If
2014there is only one number, 1 is returned instead of 0 (causing matching to
2015backtrack) when a callout of that number is reached. If two numbers (&#60;n&#62;:&#60;m&#62;)
2016are given, 1 is returned when callout &#60;n&#62; is reached and there have been at
2017least &#60;m&#62; callouts. The <b>callout_error</b> modifier is similar, except that
2018PCRE2_ERROR_CALLOUT is returned, causing the entire matching process to be
2019aborted. If both these modifiers are set for the same callout number,
2020<b>callout_error</b> takes precedence. Note that callouts with string arguments
2021are always given the number zero.
2022</P>
2023<P>
2024The <b>callout_data</b> modifier can be given an unsigned or a negative number.
2025This is set as the "user data" that is passed to the matching function, and
2026passed back when the callout function is invoked. Any value other than zero is
2027used as a return from <b>pcre2test</b>'s callout function.
2028</P>
2029<P>
2030Inserting callouts can be helpful when using <b>pcre2test</b> to check
2031complicated regular expressions. For further information about callouts, see
2032the
2033<a href="pcre2callout.html"><b>pcre2callout</b></a>
2034documentation.
2035</P>
2036<br><a name="SEC17" href="#TOC1">NON-PRINTING CHARACTERS</a><br>
2037<P>
2038When <b>pcre2test</b> is outputting text in the compiled version of a pattern,
2039bytes other than 32-126 are always treated as non-printing characters and are
2040therefore shown as hex escapes.
2041</P>
2042<P>
2043When <b>pcre2test</b> is outputting text that is a matched part of a subject
2044string, it behaves in the same way, unless a different locale has been set for
2045the pattern (using the <b>locale</b> modifier). In this case, the
2046<b>isprint()</b> function is used to distinguish printing and non-printing
2047characters.
2048<a name="saverestore"></a></P>
2049<br><a name="SEC18" href="#TOC1">SAVING AND RESTORING COMPILED PATTERNS</a><br>
2050<P>
2051It is possible to save compiled patterns on disc or elsewhere, and reload them
2052later, subject to a number of restrictions. JIT data cannot be saved. The host
2053on which the patterns are reloaded must be running the same version of PCRE2,
2054with the same code unit width, and must also have the same endianness, pointer
2055width and PCRE2_SIZE type. Before compiled patterns can be saved they must be
2056serialized, that is, converted to a stream of bytes. A single byte stream may
2057contain any number of compiled patterns, but they must all use the same
2058character tables. A single copy of the tables is included in the byte stream
2059(its size is 1088 bytes).
2060</P>
2061<P>
2062The functions whose names begin with <b>pcre2_serialize_</b> are used
2063for serializing and de-serializing. They are described in the
2064<a href="pcre2serialize.html"><b>pcre2serialize</b></a>
2065documentation. In this section we describe the features of <b>pcre2test</b> that
2066can be used to test these functions.
2067</P>
2068<P>
2069Note that "serialization" in PCRE2 does not convert compiled patterns to an
2070abstract format like Java or .NET. It just makes a reloadable byte code stream.
2071Hence the restrictions on reloading mentioned above.
2072</P>
2073<P>
2074In <b>pcre2test</b>, when a pattern with <b>push</b> modifier is successfully
2075compiled, it is pushed onto a stack of compiled patterns, and <b>pcre2test</b>
2076expects the next line to contain a new pattern (or command) instead of a
2077subject line. By contrast, the <b>pushcopy</b> modifier causes a copy of the
2078compiled pattern to be stacked, leaving the original available for immediate
2079matching. By using <b>push</b> and/or <b>pushcopy</b>, a number of patterns can
2080be compiled and retained. These modifiers are incompatible with <b>posix</b>,
2081and control modifiers that act at match time are ignored (with a message) for
2082the stacked patterns. The <b>jitverify</b> modifier applies only at compile
2083time.
2084</P>
2085<P>
2086The command
2087<pre>
2088  #save &#60;filename&#62;
2089</pre>
2090causes all the stacked patterns to be serialized and the result written to the
2091named file. Afterwards, all the stacked patterns are freed. The command
2092<pre>
2093  #load &#60;filename&#62;
2094</pre>
2095reads the data in the file, and then arranges for it to be de-serialized, with
2096the resulting compiled patterns added to the pattern stack. The pattern on the
2097top of the stack can be retrieved by the #pop command, which must be followed
2098by lines of subjects that are to be matched with the pattern, terminated as
2099usual by an empty line or end of file. This command may be followed by a
2100modifier list containing only
2101<a href="#controlmodifiers">control modifiers</a>
2102that act after a pattern has been compiled. In particular, <b>hex</b>,
2103<b>posix</b>, <b>posix_nosub</b>, <b>push</b>, and <b>pushcopy</b> are not allowed,
2104nor are any
2105<a href="#optionmodifiers">option-setting modifiers.</a>
2106The JIT modifiers are, however permitted. Here is an example that saves and
2107reloads two patterns.
2108<pre>
2109  /abc/push
2110  /xyz/push
2111  #save tempfile
2112  #load tempfile
2113  #pop info
2114  xyz
2115
2116  #pop jit,bincode
2117  abc
2118</pre>
2119If <b>jitverify</b> is used with #pop, it does not automatically imply
2120<b>jit</b>, which is different behaviour from when it is used on a pattern.
2121</P>
2122<P>
2123The #popcopy command is analagous to the <b>pushcopy</b> modifier in that it
2124makes current a copy of the topmost stack pattern, leaving the original still
2125on the stack.
2126</P>
2127<br><a name="SEC19" href="#TOC1">SEE ALSO</a><br>
2128<P>
2129<b>pcre2</b>(3), <b>pcre2api</b>(3), <b>pcre2callout</b>(3),
2130<b>pcre2jit</b>, <b>pcre2matching</b>(3), <b>pcre2partial</b>(d),
2131<b>pcre2pattern</b>(3), <b>pcre2serialize</b>(3).
2132</P>
2133<br><a name="SEC20" href="#TOC1">AUTHOR</a><br>
2134<P>
2135Philip Hazel
2136<br>
2137Retired from University Computing Service
2138<br>
2139Cambridge, England.
2140<br>
2141</P>
2142<br><a name="SEC21" href="#TOC1">REVISION</a><br>
2143<P>
2144Last updated: 12 January 2022
2145<br>
2146Copyright &copy; 1997-2022 University of Cambridge.
2147<br>
2148<p>
2149Return to the <a href="index.html">PCRE2 index page</a>.
2150</p>
2151