/[pcre]/code/trunk/doc/pcretest.1
ViewVC logotype

Diff of /code/trunk/doc/pcretest.1

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 73 by nigel, Sat Feb 24 21:40:30 2007 UTC revision 691 by ph10, Sun Sep 11 14:31:21 2011 UTC
# Line 2  Line 2 
2  .SH NAME  .SH NAME
3  pcretest - a program for testing Perl-compatible regular expressions.  pcretest - a program for testing Perl-compatible regular expressions.
4  .SH SYNOPSIS  .SH SYNOPSIS
5  .B pcretest "[-d] [-i] [-m] [-o osize] [-p] [-t] [source] [destination]"  .rs
6    .sp
7  \fBpcretest\fR was written as a test program for the PCRE regular expression  .B pcretest "[options] [input file [output file]]"
8    .sp
9    \fBpcretest\fP was written as a test program for the PCRE regular expression
10  library itself, but it can also be used for experimenting with regular  library itself, but it can also be used for experimenting with regular
11  expressions. This document describes the features of the test program; for  expressions. This document describes the features of the test program; for
12  details of the regular expressions themselves, see the  details of the regular expressions themselves, see the
13  .\" HREF  .\" HREF
14  \fBpcrepattern\fR  \fBpcrepattern\fP
15  .\"  .\"
16  documentation. For details of PCRE and its options, see the  documentation. For details of the PCRE library function calls and their
17    options, see the
18  .\" HREF  .\" HREF
19  \fBpcreapi\fR  \fBpcreapi\fP
20  .\"  .\"
21  documentation.  documentation. The input for \fBpcretest\fP is a sequence of regular expression
22    patterns and strings to be matched, as described below. The output shows the
23  .SH OPTIONS  result of each match. Options on the command line and the patterns control PCRE
24    options and exactly what is output.
25    .
26    .
27    .SH COMMAND LINE OPTIONS
28  .rs  .rs
 .sp  
29  .TP 10  .TP 10
30  \fB-C\fR  \fB-b\fP
31    Behave as if each pattern has the \fB/B\fP (show byte code) modifier; the
32    internal form is output after compilation.
33    .TP 10
34    \fB-C\fP
35  Output the version number of the PCRE library, and all available information  Output the version number of the PCRE library, and all available information
36  about the optional features that are included, and then exit.  about the optional features that are included, and then exit.
37  .TP 10  .TP 10
38  \fB-d\fR  \fB-d\fP
39  Behave as if each regex had the \fB/D\fR modifier (see below); the internal  Behave as if each pattern has the \fB/D\fP (debug) modifier; the internal
40  form is output after compilation.  form and information about the compiled pattern is output after compilation;
41    \fB-d\fP is equivalent to \fB-b -i\fP.
42    .TP 10
43    \fB-dfa\fP
44    Behave as if each data line contains the \eD escape sequence; this causes the
45    alternative matching function, \fBpcre_dfa_exec()\fP, to be used instead of the
46    standard \fBpcre_exec()\fP function (more detail is given below).
47    .TP 10
48    \fB-help\fP
49    Output a brief summary these options and then exit.
50  .TP 10  .TP 10
51  \fB-i\fR  \fB-i\fP
52  Behave as if each regex had the \fB/I\fR modifier; information about the  Behave as if each pattern has the \fB/I\fP modifier; information about the
53  compiled pattern is given after compilation.  compiled pattern is given after compilation.
54  .TP 10  .TP 10
55  \fB-m\fR  \fB-M\fP
56    Behave as if each data line contains the \eM escape sequence; this causes
57    PCRE to discover the minimum MATCH_LIMIT and MATCH_LIMIT_RECURSION settings by
58    calling \fBpcre_exec()\fP repeatedly with different limits.
59    .TP 10
60    \fB-m\fP
61  Output the size of each compiled pattern after it has been compiled. This is  Output the size of each compiled pattern after it has been compiled. This is
62  equivalent to adding /M to each regular expression. For compatibility with  equivalent to adding \fB/M\fP to each regular expression.
63  earlier versions of pcretest, \fB-s\fR is a synonym for \fB-m\fR.  .TP 10
64    \fB-o\fP \fIosize\fP
65    Set the number of elements in the output vector that is used when calling
66    \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP to be \fIosize\fP. The default value
67    is 45, which is enough for 14 capturing subexpressions for \fBpcre_exec()\fP or
68    22 different matches for \fBpcre_dfa_exec()\fP. The vector size can be
69    changed for individual matching calls by including \eO in the data line (see
70    below).
71    .TP 10
72    \fB-p\fP
73    Behave as if each pattern has the \fB/P\fP modifier; the POSIX wrapper API is
74    used to call PCRE. None of the other options has any effect when \fB-p\fP is
75    set.
76    .TP 10
77    \fB-q\fP
78    Do not output the version number of \fBpcretest\fP at the start of execution.
79  .TP 10  .TP 10
80  \fB-o\fR \fIosize\fR  \fB-S\fP \fIsize\fP
81  Set the number of elements in the output vector that is used when calling PCRE  On Unix-like systems, set the size of the run-time stack to \fIsize\fP
82  to be \fIosize\fR. The default value is 45, which is enough for 14 capturing  megabytes.
 subexpressions. The vector size can be changed for individual matching calls by  
 including \\O in the data line (see below).  
 .TP 10  
 \fB-p\fR  
 Behave as if each regex has \fB/P\fR modifier; the POSIX wrapper API is used  
 to call PCRE. None of the other options has any effect when \fB-p\fR is set.  
83  .TP 10  .TP 10
84  \fB-t\fR  \fB-s\fP or \fB-s+\fP
85    Behave as if each pattern has the \fB/S\fP modifier; in other words, force each
86    pattern to be studied. If \fB-s+\fP is used, the PCRE_STUDY_JIT_COMPILE flag is
87    passed to \fBpcre_study()\fP, causing just-in-time optimization to be set up if
88    it is available. If the \fB/I\fP or \fB/D\fP option is present on a pattern
89    (requesting output about the compiled pattern), information about the result of
90    studying is not included when studying is caused only by \fB-s\fP and neither
91    \fB-i\fP nor \fB-d\fP is present on the command line. This behaviour means that
92    the output from tests that are run with and without \fB-s\fP should be
93    identical, except when options that output information about the actual running
94    of a match are set. The \fB-M\fP, \fB-t\fP, and \fB-tm\fP options, which give
95    information about resources used, are likely to produce different output with
96    and without \fB-s\fP. Output may also differ if the \fB/C\fP option is present
97    on an individual pattern. This uses callouts to trace the the matching process,
98    and this may be different between studied and non-studied patterns. If the
99    pattern contains (*MARK) items there may also be differences, for the same
100    reason. The \fB-s\fP command line option can be overridden for specific
101    patterns that should never be studied (see the \fB/S\fP pattern modifier
102    below).
103    .TP 10
104    \fB-t\fP
105  Run each compile, study, and match many times with a timer, and output  Run each compile, study, and match many times with a timer, and output
106  resulting time per compile or match (in milliseconds). Do not set \fB-t\fR with  resulting time per compile or match (in milliseconds). Do not set \fB-m\fP with
107  \fB-m\fR, because you will then get the size output 20000 times and the timing  \fB-t\fP, because you will then get the size output a zillion times, and the
108  will be distorted.  timing will be distorted. You can control the number of iterations that are
109    used for timing by following \fB-t\fP with a number (as a separate item on the
110    command line). For example, "-t 1000" would iterate 1000 times. The default is
111    to iterate 500000 times.
112    .TP 10
113    \fB-tm\fP
114    This is like \fB-t\fP except that it times only the matching phase, not the
115    compile or study phases.
116    .
117    .
118  .SH DESCRIPTION  .SH DESCRIPTION
119  .rs  .rs
120  .sp  .sp
121  If \fBpcretest\fR is given two filename arguments, it reads from the first and  If \fBpcretest\fP is given two filename arguments, it reads from the first and
122  writes to the second. If it is given only one filename argument, it reads from  writes to the second. If it is given only one filename argument, it reads from
123  that file and writes to stdout. Otherwise, it reads from stdin and writes to  that file and writes to stdout. Otherwise, it reads from stdin and writes to
124  stdout, and prompts for each line of input, using "re>" to prompt for regular  stdout, and prompts for each line of input, using "re>" to prompt for regular
125  expressions, and "data>" to prompt for data lines.  expressions, and "data>" to prompt for data lines.
126    .P
127    When \fBpcretest\fP is built, a configuration option can specify that it should
128    be linked with the \fBlibreadline\fP library. When this is done, if the input
129    is from a terminal, it is read using the \fBreadline()\fP function. This
130    provides line-editing and history facilities. The output from the \fB-help\fP
131    option states whether or not \fBreadline()\fP will be used.
132    .P
133  The program handles any number of sets of input on a single input file. Each  The program handles any number of sets of input on a single input file. Each
134  set starts with a regular expression, and continues with any number of data  set starts with a regular expression, and continues with any number of data
135  lines to be matched against the pattern.  lines to be matched against the pattern.
136    .P
137  Each line is matched separately and independently. If you want to do  Each data line is matched separately and independently. If you want to do
138  multiple-line matches, you have to use the \\n escape sequence in a single line  multi-line matches, you have to use the \en escape sequence (or \er or \er\en,
139  of input to encode the newline characters. The maximum length of data line is  etc., depending on the newline setting) in a single line of input to encode the
140  30,000 characters.  newline sequences. There is no limit on the length of data lines; the input
141    buffer is automatically extended if it is too small.
142    .P
143  An empty line signals the end of the data lines, at which point a new regular  An empty line signals the end of the data lines, at which point a new regular
144  expression is read. The regular expressions are given enclosed in any  expression is read. The regular expressions are given enclosed in any
145  non-alphameric delimiters other than backslash, for example  non-alphanumeric delimiters other than backslash, for example:
146    .sp
147    /(a|bc)x+yz/    /(a|bc)x+yz/
148    .sp
149  White space before the initial delimiter is ignored. A regular expression may  White space before the initial delimiter is ignored. A regular expression may
150  be continued over several input lines, in which case the newline characters are  be continued over several input lines, in which case the newline characters are
151  included within it. It is possible to include the delimiter within the pattern  included within it. It is possible to include the delimiter within the pattern
152  by escaping it, for example  by escaping it, for example
153    .sp
154    /abc\\/def/    /abc\e/def/
155    .sp
156  If you do so, the escape and the delimiter form part of the pattern, but since  If you do so, the escape and the delimiter form part of the pattern, but since
157  delimiters are always non-alphameric, this does not affect its interpretation.  delimiters are always non-alphanumeric, this does not affect its interpretation.
158  If the terminating delimiter is immediately followed by a backslash, for  If the terminating delimiter is immediately followed by a backslash, for
159  example,  example,
160    .sp
161    /abc/\\    /abc/\e
162    .sp
163  then a backslash is added to the end of the pattern. This is done to provide a  then a backslash is added to the end of the pattern. This is done to provide a
164  way of testing the error condition that arises if a pattern finishes with a  way of testing the error condition that arises if a pattern finishes with a
165  backslash, because  backslash, because
166    .sp
167    /abc\\/    /abc\e/
168    .sp
169  is interpreted as the first line of a pattern that starts with "abc/", causing  is interpreted as the first line of a pattern that starts with "abc/", causing
170  pcretest to read the next line as a continuation of the regular expression.  pcretest to read the next line as a continuation of the regular expression.
171    .
172  .SH PATTERN MODIFIERS  .
173    .SH "PATTERN MODIFIERS"
174  .rs  .rs
175  .sp  .sp
176  The pattern may be followed by \fBi\fR, \fBm\fR, \fBs\fR, or \fBx\fR to set the  A pattern may be followed by any number of modifiers, which are mostly single
177  PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, or PCRE_EXTENDED options,  characters. Following Perl usage, these are referred to below as, for example,
178  respectively. For example:  "the \fB/i\fP modifier", even though the delimiter of the pattern need not
179    always be a slash, and no slash is used when writing modifiers. White space may
180    appear between the final pattern delimiter and the first modifier, and between
181    the modifiers themselves.
182    .P
183    The \fB/i\fP, \fB/m\fP, \fB/s\fP, and \fB/x\fP modifiers set the PCRE_CASELESS,
184    PCRE_MULTILINE, PCRE_DOTALL, or PCRE_EXTENDED options, respectively, when
185    \fBpcre_compile()\fP is called. These four modifier letters have the same
186    effect as they do in Perl. For example:
187    .sp
188    /caseless/i    /caseless/i
189    .sp
190  These modifier letters have the same effect as they do in Perl. There are  The following table shows additional modifiers for setting PCRE compile-time
191  others that set PCRE options that do not correspond to anything in Perl:  options that do not correspond to anything in Perl:
192  \fB/A\fR, \fB/E\fR, \fB/N\fR, \fB/U\fR, and \fB/X\fR set PCRE_ANCHORED,  .sp
193  PCRE_DOLLAR_ENDONLY, PCRE_NO_AUTO_CAPTURE, PCRE_UNGREEDY, and PCRE_EXTRA    \fB/8\fP              PCRE_UTF8
194  respectively.    \fB/?\fP              PCRE_NO_UTF8_CHECK
195      \fB/A\fP              PCRE_ANCHORED
196      \fB/C\fP              PCRE_AUTO_CALLOUT
197      \fB/E\fP              PCRE_DOLLAR_ENDONLY
198      \fB/f\fP              PCRE_FIRSTLINE
199      \fB/J\fP              PCRE_DUPNAMES
200      \fB/N\fP              PCRE_NO_AUTO_CAPTURE
201      \fB/U\fP              PCRE_UNGREEDY
202      \fB/W\fP              PCRE_UCP
203      \fB/X\fP              PCRE_EXTRA
204      \fB/Y\fP              PCRE_NO_START_OPTIMIZE
205      \fB/<JS>\fP           PCRE_JAVASCRIPT_COMPAT
206      \fB/<cr>\fP           PCRE_NEWLINE_CR
207      \fB/<lf>\fP           PCRE_NEWLINE_LF
208      \fB/<crlf>\fP         PCRE_NEWLINE_CRLF
209      \fB/<anycrlf>\fP      PCRE_NEWLINE_ANYCRLF
210      \fB/<any>\fP          PCRE_NEWLINE_ANY
211      \fB/<bsr_anycrlf>\fP  PCRE_BSR_ANYCRLF
212      \fB/<bsr_unicode>\fP  PCRE_BSR_UNICODE
213    .sp
214    The modifiers that are enclosed in angle brackets are literal strings as shown,
215    including the angle brackets, but the letters within can be in either case.
216    This example sets multiline matching with CRLF as the line ending sequence:
217    .sp
218      /^abc/m<CRLF>
219    .sp
220    As well as turning on the PCRE_UTF8 option, the \fB/8\fP modifier also causes
221    any non-printing characters in output strings to be printed using the
222    \ex{hh...} notation if they are valid UTF-8 sequences. Full details of the PCRE
223    options are given in the
224    .\" HREF
225    \fBpcreapi\fP
226    .\"
227    documentation.
228    .
229    .
230    .SS "Finding all matches in a string"
231    .rs
232    .sp
233  Searching for all possible matches within each subject string can be requested  Searching for all possible matches within each subject string can be requested
234  by the \fB/g\fR or \fB/G\fR modifier. After finding a match, PCRE is called  by the \fB/g\fP or \fB/G\fP modifier. After finding a match, PCRE is called
235  again to search the remainder of the subject string. The difference between  again to search the remainder of the subject string. The difference between
236  \fB/g\fR and \fB/G\fR is that the former uses the \fIstartoffset\fR argument to  \fB/g\fP and \fB/G\fP is that the former uses the \fIstartoffset\fP argument to
237  \fBpcre_exec()\fR to start searching at a new point within the entire string  \fBpcre_exec()\fP to start searching at a new point within the entire string
238  (which is in effect what Perl does), whereas the latter passes over a shortened  (which is in effect what Perl does), whereas the latter passes over a shortened
239  substring. This makes a difference to the matching process if the pattern  substring. This makes a difference to the matching process if the pattern
240  begins with a lookbehind assertion (including \\b or \\B).  begins with a lookbehind assertion (including \eb or \eB).
241    .P
242  If any call to \fBpcre_exec()\fR in a \fB/g\fR or \fB/G\fR sequence matches an  If any call to \fBpcre_exec()\fP in a \fB/g\fP or \fB/G\fP sequence matches an
243  empty string, the next call is done with the PCRE_NOTEMPTY and PCRE_ANCHORED  empty string, the next call is done with the PCRE_NOTEMPTY_ATSTART and
244  flags set in order to search for another, non-empty, match at the same point.  PCRE_ANCHORED flags set in order to search for another, non-empty, match at the
245  If this second match fails, the start offset is advanced by one, and the normal  same point. If this second match fails, the start offset is advanced, and the
246  match is retried. This imitates the way Perl handles such cases when using the  normal match is retried. This imitates the way Perl handles such cases when
247  \fB/g\fR modifier or the \fBsplit()\fR function.  using the \fB/g\fP modifier or the \fBsplit()\fP function. Normally, the start
248    offset is advanced by one character, but if the newline convention recognizes
249  There are a number of other modifiers for controlling the way \fBpcretest\fR  CRLF as a newline, and the current character is CR followed by LF, an advance
250    of two is used.
251    .
252    .
253    .SS "Other modifiers"
254    .rs
255    .sp
256    There are yet more modifiers for controlling the way \fBpcretest\fP
257  operates.  operates.
258    .P
259  The \fB/+\fR modifier requests that as well as outputting the substring that  The \fB/+\fP modifier requests that as well as outputting the substring that
260  matched the entire pattern, pcretest should in addition output the remainder of  matched the entire pattern, \fBpcretest\fP should in addition output the
261  the subject string. This is useful for tests where the subject contains  remainder of the subject string. This is useful for tests where the subject
262  multiple copies of the same substring.  contains multiple copies of the same substring. If the \fB+\fP modifier appears
263    twice, the same action is taken for captured substrings. In each case the
264  The \fB/L\fR modifier must be followed directly by the name of a locale, for  remainder is output on the following line with a plus character following the
265    capture number. Note that this modifier must not immediately follow the /S
266    modifier because /S+ has another meaning.
267    .P
268    The \fB/=\fP modifier requests that the values of all potential captured
269    parentheses be output after a match by \fBpcre_exec()\fP. By default, only
270    those up to the highest one actually used in the match are output
271    (corresponding to the return code from \fBpcre_exec()\fP). Values in the
272    offsets vector corresponding to higher numbers should be set to -1, and these
273    are output as "<unset>". This modifier gives a way of checking that this is
274    happening.
275    .P
276    The \fB/B\fP modifier is a debugging feature. It requests that \fBpcretest\fP
277    output a representation of the compiled byte code after compilation. Normally
278    this information contains length and offset values; however, if \fB/Z\fP is
279    also present, this data is replaced by spaces. This is a special feature for
280    use in the automatic test scripts; it ensures that the same output is generated
281    for different internal link sizes.
282    .P
283    The \fB/D\fP modifier is a PCRE debugging feature, and is equivalent to
284    \fB/BI\fP, that is, both the \fB/B\fP and the \fB/I\fP modifiers.
285    .P
286    The \fB/F\fP modifier causes \fBpcretest\fP to flip the byte order of the
287    fields in the compiled pattern that contain 2-byte and 4-byte numbers. This
288    facility is for testing the feature in PCRE that allows it to execute patterns
289    that were compiled on a host with a different endianness. This feature is not
290    available when the POSIX interface to PCRE is being used, that is, when the
291    \fB/P\fP pattern modifier is specified. See also the section about saving and
292    reloading compiled patterns below.
293    .P
294    The \fB/I\fP modifier requests that \fBpcretest\fP output information about the
295    compiled pattern (whether it is anchored, has a fixed first character, and
296    so on). It does this by calling \fBpcre_fullinfo()\fP after compiling a
297    pattern. If the pattern is studied, the results of that are also output.
298    .P
299    The \fB/K\fP modifier requests \fBpcretest\fP to show names from backtracking
300    control verbs that are returned from calls to \fBpcre_exec()\fP. It causes
301    \fBpcretest\fP to create a \fBpcre_extra\fP block if one has not already been
302    created by a call to \fBpcre_study()\fP, and to set the PCRE_EXTRA_MARK flag
303    and the \fBmark\fP field within it, every time that \fBpcre_exec()\fP is
304    called. If the variable that the \fBmark\fP field points to is non-NULL for a
305    match, non-match, or partial match, \fBpcretest\fP prints the string to which
306    it points. For a match, this is shown on a line by itself, tagged with "MK:".
307    For a non-match it is added to the message.
308    .P
309    The \fB/L\fP modifier must be followed directly by the name of a locale, for
310  example,  example,
   
   /pattern/Lfr  
   
 For this reason, it must be the last modifier letter. The given locale is set,  
 \fBpcre_maketables()\fR is called to build a set of character tables for the  
 locale, and this is then passed to \fBpcre_compile()\fR when compiling the  
 regular expression. Without an \fB/L\fR modifier, NULL is passed as the tables  
 pointer; that is, \fB/L\fR applies only to the expression on which it appears.  
   
 The \fB/I\fR modifier requests that \fBpcretest\fR output information about the  
 compiled expression (whether it is anchored, has a fixed first character, and  
 so on). It does this by calling \fBpcre_fullinfo()\fR after compiling an  
 expression, and outputting the information it gets back. If the pattern is  
 studied, the results of that are also output.  
   
 The \fB/D\fR modifier is a PCRE debugging feature, which also assumes \fB/I\fR.  
 It causes the internal form of compiled regular expressions to be output after  
 compilation. If the pattern was studied, the information returned is also  
 output.  
   
 The \fB/S\fR modifier causes \fBpcre_study()\fR to be called after the  
 expression has been compiled, and the results used when the expression is  
 matched.  
   
 The \fB/M\fR modifier causes the size of memory block used to hold the compiled  
 pattern to be output.  
   
 The \fB/P\fR modifier causes \fBpcretest\fR to call PCRE via the POSIX wrapper  
 API rather than its native API. When this is done, all other modifiers except  
 \fB/i\fR, \fB/m\fR, and \fB/+\fR are ignored. REG_ICASE is set if \fB/i\fR is  
 present, and REG_NEWLINE is set if \fB/m\fR is present. The wrapper functions  
 force PCRE_DOLLAR_ENDONLY always, and PCRE_DOTALL unless REG_NEWLINE is set.  
   
 The \fB/8\fR modifier causes \fBpcretest\fR to call PCRE with the PCRE_UTF8  
 option set. This turns on support for UTF-8 character handling in PCRE,  
 provided that it was compiled with this support enabled. This modifier also  
 causes any non-printing characters in output strings to be printed using the  
 \\x{hh...} notation if they are valid UTF-8 sequences.  
   
 If the \fB/?\fR modifier is used with \fB/8\fR, it causes \fBpcretest\fR to  
 call \fBpcre_compile()\fR with the PCRE_NO_UTF8_CHECK option, to suppress the  
 checking of the string for UTF-8 validity.  
   
 .SH CALLOUTS  
 .rs  
311  .sp  .sp
312  If the pattern contains any callout requests, \fBpcretest\fR's callout function    /pattern/Lfr_FR
313  will be called. By default, it displays the callout number, and the start and  .sp
314  current positions in the text at the callout time. For example, the output  For this reason, it must be the last modifier. The given locale is set,
315    \fBpcre_maketables()\fP is called to build a set of character tables for the
316    --->pqrabcdef  locale, and this is then passed to \fBpcre_compile()\fP when compiling the
317      0    ^  ^  regular expression. Without an \fB/L\fP (or \fB/T\fP) modifier, NULL is passed
318    as the tables pointer; that is, \fB/L\fP applies only to the expression on
319  indicates that callout number 0 occurred for a match attempt starting at the  which it appears.
320  fourth character of the subject string, when the pointer was at the seventh  .P
321  character. The callout function returns zero (carry on matching) by default.  The \fB/M\fP modifier causes the size of memory block used to hold the compiled
322    pattern to be output.
323  Inserting callouts may be helpful when using \fBpcretest\fR to check  .P
324  complicated regular expressions. For further information about callouts, see  If the \fB/S\fP modifier appears once, it causes \fBpcre_study()\fP to be
325  the  called after the expression has been compiled, and the results used when the
326    expression is matched. If \fB/S\fP appears twice, it suppresses studying, even
327    if it was requested externally by the \fB-s\fP command line option. This makes
328    it possible to specify that certain patterns are always studied, and others are
329    never studied, independently of \fB-s\fP. This feature is used in the test
330    files in a few cases where the output is different when the pattern is studied.
331    .P
332    If the \fB/S\fP modifier is immediately followed by a + character, the call to
333    \fBpcre_study()\fP is made with the PCRE_STUDY_JIT_COMPILE option, requesting
334    just-in-time optimization support if it is available. Note that there is also a
335    \fB/+\fP modifier; it must not be given immediately after \fB/S\fP because this
336    will be misinterpreted. If JIT studying is successful, it will automatically be
337    used when \fBpcre_exec()\fP is run, except when incompatible run-time options
338    are specified. These include the partial matching options; a complete list is
339    given in the
340  .\" HREF  .\" HREF
341  \fBpcrecallout\fR  \fBpcrejit\fP
342  .\"  .\"
343  documentation.  documentation. See also the \fB\eJ\fP escape sequence below for a way of
344    setting the size of the JIT stack.
345  For testing the PCRE library, additional control of callout behaviour is  .P
346  available via escape sequences in the data, as described in the following  The \fB/T\fP modifier must be followed by a single digit. It causes a specific
347  section. In particular, it is possible to pass in a number as callout data (the  set of built-in character tables to be passed to \fBpcre_compile()\fP. It is
348  default is zero). If the callout function receives a non-zero number, it  used in the standard PCRE tests to check behaviour with different character
349  returns that value instead of zero.  tables. The digit specifies the tables as follows:
350    .sp
351  .SH DATA LINES    0   the default ASCII tables, as distributed in
352            pcre_chartables.c.dist
353      1   a set of tables defining ISO 8859 characters
354    .sp
355    In table 1, some characters whose codes are greater than 128 are identified as
356    letters, digits, spaces, etc.
357    .
358    .
359    .SS "Using the POSIX wrapper API"
360    .rs
361    .sp
362    The \fB/P\fP modifier causes \fBpcretest\fP to call PCRE via the POSIX wrapper
363    API rather than its native API. When \fB/P\fP is set, the following modifiers
364    set options for the \fBregcomp()\fP function:
365    .sp
366      /i    REG_ICASE
367      /m    REG_NEWLINE
368      /N    REG_NOSUB
369      /s    REG_DOTALL     )
370      /U    REG_UNGREEDY   ) These options are not part of
371      /W    REG_UCP        )   the POSIX standard
372      /8    REG_UTF8       )
373    .sp
374    The \fB/+\fP modifier works as described above. All other modifiers are
375    ignored.
376    .
377    .
378    .SH "DATA LINES"
379  .rs  .rs
380  .sp  .sp
381  Before each data line is passed to \fBpcre_exec()\fR, leading and trailing  Before each data line is passed to \fBpcre_exec()\fP, leading and trailing
382  whitespace is removed, and it is then scanned for \\ escapes. Some of these are  white space is removed, and it is then scanned for \e escapes. Some of these
383  pretty esoteric features, intended for checking out some of the more  are pretty esoteric features, intended for checking out some of the more
384  complicated features of PCRE. If you are just testing "ordinary" regular  complicated features of PCRE. If you are just testing "ordinary" regular
385  expressions, you probably don't need any of these. The following escapes are  expressions, you probably don't need any of these. The following escapes are
386  recognized:  recognized:
387    .sp
388    \\a         alarm (= BEL)    \ea         alarm (BEL, \ex07)
389    \\b         backspace    \eb         backspace (\ex08)
390    \\e         escape    \ee         escape (\ex27)
391    \\f         formfeed    \ef         form feed (\ex0c)
392    \\n         newline    \en         newline (\ex0a)
393    \\r         carriage return  .\" JOIN
394    \\t         tab    \eqdd       set the PCRE_MATCH_LIMIT limit to dd
395    \\v         vertical tab                 (any number of digits)
396    \\nnn       octal character (up to 3 octal digits)    \er         carriage return (\ex0d)
397    \\xhh       hexadecimal character (up to 2 hex digits)    \et         tab (\ex09)
398    \\x{hh...}  hexadecimal character, any number of digits    \ev         vertical tab (\ex0b)
399      \ennn       octal character (up to 3 octal digits)
400                   always a byte unless > 255 in UTF-8 mode
401      \exhh       hexadecimal byte (up to 2 hex digits)
402    .\" JOIN
403      \ex{hh...}  hexadecimal character, any number of digits
404                 in UTF-8 mode                 in UTF-8 mode
405    \\A         pass the PCRE_ANCHORED option to \fBpcre_exec()\fR  .\" JOIN
406    \\B         pass the PCRE_NOTBOL option to \fBpcre_exec()\fR    \eA         pass the PCRE_ANCHORED option to \fBpcre_exec()\fP
407    \\Cdd       call pcre_copy_substring() for substring dd                 or \fBpcre_dfa_exec()\fP
408                 after a successful match (any decimal number  .\" JOIN
409                 less than 32)    \eB         pass the PCRE_NOTBOL option to \fBpcre_exec()\fP
410    \\Cname     call pcre_copy_named_substring() for substring                 or \fBpcre_dfa_exec()\fP
411    .\" JOIN
412      \eCdd       call pcre_copy_substring() for substring dd
413                   after a successful match (number less than 32)
414    .\" JOIN
415      \eCname     call pcre_copy_named_substring() for substring
416                 "name" after a successful match (name termin-                 "name" after a successful match (name termin-
417                 ated by next non alphanumeric character)                 ated by next non alphanumeric character)
418    \\C+        show the current captured substrings at callout  .\" JOIN
419      \eC+        show the current captured substrings at callout
420                 time                 time
421    \\C-        do not supply a callout function    \eC-        do not supply a callout function
422    \\C!n       return 1 instead of 0 when callout number n is  .\" JOIN
423      \eC!n       return 1 instead of 0 when callout number n is
424                 reached                 reached
425    \\C!n!m     return 1 instead of 0 when callout number n is  .\" JOIN
426      \eC!n!m     return 1 instead of 0 when callout number n is
427                 reached for the nth time                 reached for the nth time
428    \\C*n       pass the number n (may be negative) as callout  .\" JOIN
429                 data    \eC*n       pass the number n (may be negative) as callout
430    \\Gdd       call pcre_get_substring() for substring dd                 data; this is used as the callout return value
431                 after a successful match (any decimal number    \eD         use the \fBpcre_dfa_exec()\fP match function
432                 less than 32)    \eF         only shortest match for \fBpcre_dfa_exec()\fP
433    \\Gname     call pcre_get_named_substring() for substring  .\" JOIN
434      \eGdd       call pcre_get_substring() for substring dd
435                   after a successful match (number less than 32)
436    .\" JOIN
437      \eGname     call pcre_get_named_substring() for substring
438                 "name" after a successful match (name termin-                 "name" after a successful match (name termin-
439                 ated by next non-alphanumeric character)                 ated by next non-alphanumeric character)
440    \\L         call pcre_get_substringlist() after a  .\" JOIN
441      \eJdd       set up a JIT stack of dd kilobytes maximum (any
442                   number of digits)
443    .\" JOIN
444      \eL         call pcre_get_substringlist() after a
445                 successful match                 successful match
446    \\M         discover the minimum MATCH_LIMIT setting  .\" JOIN
447    \\N         pass the PCRE_NOTEMPTY option to \fBpcre_exec()\fR    \eM         discover the minimum MATCH_LIMIT and
448    \\Odd       set the size of the output vector passed to                 MATCH_LIMIT_RECURSION settings
449                 \fBpcre_exec()\fR to dd (any number of decimal  .\" JOIN
450                 digits)    \eN         pass the PCRE_NOTEMPTY option to \fBpcre_exec()\fP
451    \\S         output details of memory get/free calls during matching                 or \fBpcre_dfa_exec()\fP; if used twice, pass the
452    \\Z         pass the PCRE_NOTEOL option to \fBpcre_exec()\fR                 PCRE_NOTEMPTY_ATSTART option
453    \\?         pass the PCRE_NO_UTF8_CHECK option to  .\" JOIN
454                 \fBpcre_exec()\fR    \eOdd       set the size of the output vector passed to
455                   \fBpcre_exec()\fP to dd (any number of digits)
456  If \\M is present, \fBpcretest\fR calls \fBpcre_exec()\fR several times, with  .\" JOIN
457  different values in the \fImatch_limit\fR field of the \fBpcre_extra\fR data    \eP         pass the PCRE_PARTIAL_SOFT option to \fBpcre_exec()\fP
458  structure, until it finds the minimum number that is needed for                 or \fBpcre_dfa_exec()\fP; if used twice, pass the
459  \fBpcre_exec()\fR to complete. This number is a measure of the amount of                 PCRE_PARTIAL_HARD option
460  recursion and backtracking that takes place, and checking it out can be  .\" JOIN
461  instructive. For most simple matches, the number is quite small, but for    \eQdd       set the PCRE_MATCH_LIMIT_RECURSION limit to dd
462  patterns with very large numbers of matching possibilities, it can become large                 (any number of digits)
463  very quickly with increasing length of subject string.    \eR         pass the PCRE_DFA_RESTART option to \fBpcre_dfa_exec()\fP
464      \eS         output details of memory get/free calls during matching
465  When \\O is used, it may be higher or lower than the size set by the \fB-O\fR  .\" JOIN
466  option (or defaulted to 45); \\O applies only to the call of \fBpcre_exec()\fR    \eY         pass the PCRE_NO_START_OPTIMIZE option to \fBpcre_exec()\fP
467  for the line in which it appears.                 or \fBpcre_dfa_exec()\fP
468    .\" JOIN
469  A backslash followed by anything else just escapes the anything else. If the    \eZ         pass the PCRE_NOTEOL option to \fBpcre_exec()\fP
470  very last character is a backslash, it is ignored. This gives a way of passing                 or \fBpcre_dfa_exec()\fP
471  an empty line as data, since a real empty line terminates the data input.  .\" JOIN
472      \e?         pass the PCRE_NO_UTF8_CHECK option to
473  If \fB/P\fR was present on the regex, causing the POSIX wrapper API to be used,                 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP
474  only \fB\B\fR, and \fB\Z\fR have any effect, causing REG_NOTBOL and REG_NOTEOL  .\" JOIN
475  to be passed to \fBregexec()\fR respectively.    \e>dd       start the match at offset dd (optional "-"; then
476                   any number of digits); this sets the \fIstartoffset\fP
477  The use of \\x{hh...} to represent UTF-8 characters is not dependent on the use                 argument for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP
478  of the \fB/8\fR modifier on the pattern. It is recognized always. There may be  .\" JOIN
479      \e<cr>      pass the PCRE_NEWLINE_CR option to \fBpcre_exec()\fP
480                   or \fBpcre_dfa_exec()\fP
481    .\" JOIN
482      \e<lf>      pass the PCRE_NEWLINE_LF option to \fBpcre_exec()\fP
483                   or \fBpcre_dfa_exec()\fP
484    .\" JOIN
485      \e<crlf>    pass the PCRE_NEWLINE_CRLF option to \fBpcre_exec()\fP
486                   or \fBpcre_dfa_exec()\fP
487    .\" JOIN
488      \e<anycrlf> pass the PCRE_NEWLINE_ANYCRLF option to \fBpcre_exec()\fP
489                   or \fBpcre_dfa_exec()\fP
490    .\" JOIN
491      \e<any>     pass the PCRE_NEWLINE_ANY option to \fBpcre_exec()\fP
492                   or \fBpcre_dfa_exec()\fP
493    .sp
494    Note that \exhh always specifies one byte, even in UTF-8 mode; this makes it
495    possible to construct invalid UTF-8 sequences for testing purposes. On the
496    other hand, \ex{hh} is interpreted as a UTF-8 character in UTF-8 mode,
497    generating more than one byte if the value is greater than 127. When not in
498    UTF-8 mode, it generates one byte for values less than 256, and causes an error
499    for greater values.
500    .P
501    The escapes that specify line ending sequences are literal strings, exactly as
502    shown. No more than one newline setting should be present in any data line.
503    .P
504    A backslash followed by anything else just escapes the anything else. If
505    the very last character is a backslash, it is ignored. This gives a way of
506    passing an empty line as data, since a real empty line terminates the data
507    input.
508    .P
509    The \fB\eJ\fP escape provides a way of setting the maximum stack size that is
510    used by the just-in-time optimization code. It is ignored if JIT optimization
511    is not being used. Providing a stack that is larger than the default 32K is
512    necessary only for very complicated patterns.
513    .P
514    If \eM is present, \fBpcretest\fP calls \fBpcre_exec()\fP several times, with
515    different values in the \fImatch_limit\fP and \fImatch_limit_recursion\fP
516    fields of the \fBpcre_extra\fP data structure, until it finds the minimum
517    numbers for each parameter that allow \fBpcre_exec()\fP to complete without
518    error. Because this is testing a specific feature of the normal interpretive
519    \fBpcre_exec()\fP execution, the use of any JIT optimization that might have
520    been set up by the \fB/S+\fP qualifier of \fB-s+\fP option is disabled.
521    .P
522    The \fImatch_limit\fP number is a measure of the amount of backtracking
523    that takes place, and checking it out can be instructive. For most simple
524    matches, the number is quite small, but for patterns with very large numbers of
525    matching possibilities, it can become large very quickly with increasing length
526    of subject string. The \fImatch_limit_recursion\fP number is a measure of how
527    much stack (or, if PCRE is compiled with NO_RECURSE, how much heap) memory is
528    needed to complete the match attempt.
529    .P
530    When \eO is used, the value specified may be higher or lower than the size set
531    by the \fB-O\fP command line option (or defaulted to 45); \eO applies only to
532    the call of \fBpcre_exec()\fP for the line in which it appears.
533    .P
534    If the \fB/P\fP modifier was present on the pattern, causing the POSIX wrapper
535    API to be used, the only option-setting sequences that have any effect are \eB,
536    \eN, and \eZ, causing REG_NOTBOL, REG_NOTEMPTY, and REG_NOTEOL, respectively,
537    to be passed to \fBregexec()\fP.
538    .P
539    The use of \ex{hh...} to represent UTF-8 characters is not dependent on the use
540    of the \fB/8\fP modifier on the pattern. It is recognized always. There may be
541  any number of hexadecimal digits inside the braces. The result is from one to  any number of hexadecimal digits inside the braces. The result is from one to
542  six bytes, encoded according to the UTF-8 rules.  six bytes, encoded according to the original UTF-8 rules of RFC 2279. This
543    allows for values in the range 0 to 0x7FFFFFFF. Note that not all of those are
544  .SH OUTPUT FROM PCRETEST  valid Unicode code points, or indeed valid UTF-8 characters according to the
545    later rules in RFC 3629.
546    .
547    .
548    .SH "THE ALTERNATIVE MATCHING FUNCTION"
549  .rs  .rs
550  .sp  .sp
551  When a match succeeds, pcretest outputs the list of captured substrings that  By default, \fBpcretest\fP uses the standard PCRE matching function,
552  \fBpcre_exec()\fR returns, starting with number 0 for the string that matched  \fBpcre_exec()\fP to match each data line. From release 6.0, PCRE supports an
553  the whole pattern. Here is an example of an interactive pcretest run.  alternative matching function, \fBpcre_dfa_test()\fP, which operates in a
554    different way, and has some restrictions. The differences between the two
555    functions are described in the
556    .\" HREF
557    \fBpcrematching\fP
558    .\"
559    documentation.
560    .P
561    If a data line contains the \eD escape sequence, or if the command line
562    contains the \fB-dfa\fP option, the alternative matching function is called.
563    This function finds all possible matches at a given point. If, however, the \eF
564    escape sequence is present in the data line, it stops after the first match is
565    found. This is always the shortest possible match.
566    .
567    .
568    .SH "DEFAULT OUTPUT FROM PCRETEST"
569    .rs
570    .sp
571    This section describes the output when the normal matching function,
572    \fBpcre_exec()\fP, is being used.
573    .P
574    When a match succeeds, \fBpcretest\fP outputs the list of captured substrings
575    that \fBpcre_exec()\fP returns, starting with number 0 for the string that
576    matched the whole pattern. Otherwise, it outputs "No match" when the return is
577    PCRE_ERROR_NOMATCH, and "Partial match:" followed by the partially matching
578    substring when \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL. (Note that this is
579    the entire substring that was inspected during the partial match; it may
580    include characters before the actual match start if a lookbehind assertion,
581    \eK, \eb, or \eB was involved.) For any other return, \fBpcretest\fP outputs
582    the PCRE negative error number and a short descriptive phrase. If the error is
583    a failed UTF-8 string check, the byte offset of the start of the failing
584    character and the reason code are also output, provided that the size of the
585    output vector is at least two. Here is an example of an interactive
586    \fBpcretest\fP run.
587    .sp
588    $ pcretest    $ pcretest
589    PCRE version 4.00 08-Jan-2003    PCRE version 8.13 2011-04-30
590    .sp
591      re> /^abc(\\d+)/      re> /^abc(\ed+)/
592    data> abc123    data> abc123
593     0: abc123     0: abc123
594     1: 123     1: 123
595    data> xyz    data> xyz
596    No match    No match
597    .sp
598  If the strings contain any non-printing characters, they are output as \\0x  Unset capturing substrings that are not followed by one that is set are not
599  escapes, or as \\x{...} escapes if the \fB/8\fR modifier was present on the  returned by \fBpcre_exec()\fP, and are not shown by \fBpcretest\fP. In the
600  pattern. If the pattern has the \fB/+\fR modifier, then the output for  following example, there are two capturing substrings, but when the first data
601  substring 0 is followed by the the rest of the subject string, identified by  line is matched, the second, unset substring is not shown. An "internal" unset
602  "0+" like this:  substring is shown as "<unset>", as for the second data line.
603    .sp
604        re> /(a)|(b)/
605      data> a
606       0: a
607       1: a
608      data> b
609       0: b
610       1: <unset>
611       2: b
612    .sp
613    If the strings contain any non-printing characters, they are output as \e0x
614    escapes, or as \ex{...} escapes if the \fB/8\fP modifier was present on the
615    pattern. See below for the definition of non-printing characters. If the
616    pattern has the \fB/+\fP modifier, the output for substring 0 is followed by
617    the the rest of the subject string, identified by "0+" like this:
618    .sp
619      re> /cat/+      re> /cat/+
620    data> cataract    data> cataract
621     0: cat     0: cat
622     0+ aract     0+ aract
623    .sp
624  If the pattern has the \fB/g\fR or \fB/G\fR modifier, the results of successive  If the pattern has the \fB/g\fP or \fB/G\fP modifier, the results of successive
625  matching attempts are output in sequence, like this:  matching attempts are output in sequence, like this:
626    .sp
627      re> /\\Bi(\\w\\w)/g      re> /\eBi(\ew\ew)/g
628    data> Mississippi    data> Mississippi
629     0: iss     0: iss
630     1: ss     1: ss
# Line 335  matching attempts are output in sequence Line 632  matching attempts are output in sequence
632     1: ss     1: ss
633     0: ipp     0: ipp
634     1: pp     1: pp
635    .sp
636  "No match" is output only if the first match attempt fails.  "No match" is output only if the first match attempt fails. Here is an example
637    of a failure message (the offset 4 that is specified by \e>4 is past the end of
638  If any of the sequences \fB\\C\fR, \fB\\G\fR, or \fB\\L\fR are present in a  the subject string):
639    .sp
640        re> /xyz/
641      data> xyz\e>4
642      Error -24 (bad offset value)
643    .P
644    If any of the sequences \fB\eC\fP, \fB\eG\fP, or \fB\eL\fP are present in a
645  data line that is successfully matched, the substrings extracted by the  data line that is successfully matched, the substrings extracted by the
646  convenience functions are output with C, G, or L after the string number  convenience functions are output with C, G, or L after the string number
647  instead of a colon. This is in addition to the normal full list. The string  instead of a colon. This is in addition to the normal full list. The string
648  length (that is, the return from the extraction function) is given in  length (that is, the return from the extraction function) is given in
649  parentheses after each string for \fB\\C\fR and \fB\\G\fR.  parentheses after each string for \fB\eC\fP and \fB\eG\fP.
650    .P
651  Note that while patterns can be continued over several lines (a plain ">"  Note that whereas patterns can be continued over several lines (a plain ">"
652  prompt is used for continuations), data lines may not. However newlines can be  prompt is used for continuations), data lines may not. However newlines can be
653  included in data by means of the \\n escape.  included in data by means of the \en escape (or \er, \er\en, etc., depending on
654    the newline sequence setting).
655    .
656    .
657    .
658    .SH "OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION"
659    .rs
660    .sp
661    When the alternative matching function, \fBpcre_dfa_exec()\fP, is used (by
662    means of the \eD escape sequence or the \fB-dfa\fP command line option), the
663    output consists of a list of all the matches that start at the first point in
664    the subject where there is at least one match. For example:
665    .sp
666        re> /(tang|tangerine|tan)/
667      data> yellow tangerine\eD
668       0: tangerine
669       1: tang
670       2: tan
671    .sp
672    (Using the normal matching function on this data finds only "tang".) The
673    longest matching string is always given first (and numbered zero). After a
674    PCRE_ERROR_PARTIAL return, the output is "Partial match:", followed by the
675    partially matching substring. (Note that this is the entire substring that was
676    inspected during the partial match; it may include characters before the actual
677    match start if a lookbehind assertion, \eK, \eb, or \eB was involved.)
678    .P
679    If \fB/g\fP is present on the pattern, the search for further matches resumes
680    at the end of the longest match. For example:
681    .sp
682        re> /(tang|tangerine|tan)/g
683      data> yellow tangerine and tangy sultana\eD
684       0: tangerine
685       1: tang
686       2: tan
687       0: tang
688       1: tan
689       0: tan
690    .sp
691    Since the matching function does not support substring capture, the escape
692    sequences that are concerned with captured substrings are not relevant.
693    .
694    .
695    .SH "RESTARTING AFTER A PARTIAL MATCH"
696    .rs
697    .sp
698    When the alternative matching function has given the PCRE_ERROR_PARTIAL return,
699    indicating that the subject partially matched the pattern, you can restart the
700    match with additional subject data by means of the \eR escape sequence. For
701    example:
702    .sp
703        re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
704      data> 23ja\eP\eD
705      Partial match: 23ja
706      data> n05\eR\eD
707       0: n05
708    .sp
709    For further information about partial matching, see the
710    .\" HREF
711    \fBpcrepartial\fP
712    .\"
713    documentation.
714    .
715    .
716    .SH CALLOUTS
717    .rs
718    .sp
719    If the pattern contains any callout requests, \fBpcretest\fP's callout function
720    is called during matching. This works with both matching functions. By default,
721    the called function displays the callout number, the start and current
722    positions in the text at the callout time, and the next pattern item to be
723    tested. For example, the output
724    .sp
725      --->pqrabcdef
726        0    ^  ^     \ed
727    .sp
728    indicates that callout number 0 occurred for a match attempt starting at the
729    fourth character of the subject string, when the pointer was at the seventh
730    character of the data, and when the next pattern item was \ed. Just one
731    circumflex is output if the start and current positions are the same.
732    .P
733    Callouts numbered 255 are assumed to be automatic callouts, inserted as a
734    result of the \fB/C\fP pattern modifier. In this case, instead of showing the
735    callout number, the offset in the pattern, preceded by a plus, is output. For
736    example:
737    .sp
738        re> /\ed?[A-E]\e*/C
739      data> E*
740      --->E*
741       +0 ^      \ed?
742       +3 ^      [A-E]
743       +8 ^^     \e*
744      +10 ^ ^
745       0: E*
746    .sp
747    If a pattern contains (*MARK) items, an additional line is output whenever
748    a change of latest mark is passed to the callout function. For example:
749    .sp
750        re> /a(*MARK:X)bc/C
751      data> abc
752      --->abc
753       +0 ^       a
754       +1 ^^      (*MARK:X)
755      +10 ^^      b
756      Latest Mark: X
757      +11 ^ ^     c
758      +12 ^  ^
759       0: abc
760    .sp
761    The mark changes between matching "a" and "b", but stays the same for the rest
762    of the match, so nothing more is output. If, as a result of backtracking, the
763    mark reverts to being unset, the text "<unset>" is output.
764    .P
765    The callout function in \fBpcretest\fP returns zero (carry on matching) by
766    default, but you can use a \eC item in a data line (as described above) to
767    change this and other parameters of the callout.
768    .P
769    Inserting callouts can be helpful when using \fBpcretest\fP to check
770    complicated regular expressions. For further information about callouts, see
771    the
772    .\" HREF
773    \fBpcrecallout\fP
774    .\"
775    documentation.
776    .
777    .
778    .
779    .SH "NON-PRINTING CHARACTERS"
780    .rs
781    .sp
782    When \fBpcretest\fP is outputting text in the compiled version of a pattern,
783    bytes other than 32-126 are always treated as non-printing characters are are
784    therefore shown as hex escapes.
785    .P
786    When \fBpcretest\fP is outputting text that is a matched part of a subject
787    string, it behaves in the same way, unless a different locale has been set for
788    the pattern (using the \fB/L\fP modifier). In this case, the \fBisprint()\fP
789    function to distinguish printing and non-printing characters.
790    .
791    .
792    .
793    .SH "SAVING AND RELOADING COMPILED PATTERNS"
794    .rs
795    .sp
796    The facilities described in this section are not available when the POSIX
797    interface to PCRE is being used, that is, when the \fB/P\fP pattern modifier is
798    specified.
799    .P
800    When the POSIX interface is not in use, you can cause \fBpcretest\fP to write a
801    compiled pattern to a file, by following the modifiers with > and a file name.
802    For example:
803    .sp
804      /pattern/im >/some/file
805    .sp
806    See the
807    .\" HREF
808    \fBpcreprecompile\fP
809    .\"
810    documentation for a discussion about saving and re-using compiled patterns.
811    Note that if the pattern was successfully studied with JIT optimization, the
812    JIT data cannot be saved.
813    .P
814    The data that is written is binary. The first eight bytes are the length of the
815    compiled pattern data followed by the length of the optional study data, each
816    written as four bytes in big-endian order (most significant byte first). If
817    there is no study data (either the pattern was not studied, or studying did not
818    return any data), the second length is zero. The lengths are followed by an
819    exact copy of the compiled pattern. If there is additional study data, this
820    (excluding any JIT data) follows immediately after the compiled pattern. After
821    writing the file, \fBpcretest\fP expects to read a new pattern.
822    .P
823    A saved pattern can be reloaded into \fBpcretest\fP by specifying < and a file
824    name instead of a pattern. The name of the file must not contain a < character,
825    as otherwise \fBpcretest\fP will interpret the line as a pattern delimited by <
826    characters.
827    For example:
828    .sp
829       re> </some/file
830      Compiled pattern loaded from /some/file
831      No study data
832    .sp
833    If the pattern was previously studied with the JIT optimization, the JIT
834    information cannot be saved and restored, and so is lost. When the pattern has
835    been loaded, \fBpcretest\fP proceeds to read data lines in the usual way.
836    .P
837    You can copy a file written by \fBpcretest\fP to a different host and reload it
838    there, even if the new host has opposite endianness to the one on which the
839    pattern was compiled. For example, you can compile on an i86 machine and run on
840    a SPARC machine.
841    .P
842    File names for saving and reloading can be absolute or relative, but note that
843    the shell facility of expanding a file name that starts with a tilde (~) is not
844    available.
845    .P
846    The ability to save and reload files in \fBpcretest\fP is intended for testing
847    and experimentation. It is not intended for production use because only a
848    single pattern can be written to a file. Furthermore, there is no facility for
849    supplying custom character tables for use with a reloaded pattern. If the
850    original pattern was compiled with custom tables, an attempt to match a subject
851    string using a reloaded pattern is likely to cause \fBpcretest\fP to crash.
852    Finally, if you attempt to load a file that is not in the correct format, the
853    result is undefined.
854    .
855    .
856    .SH "SEE ALSO"
857    .rs
858    .sp
859    \fBpcre\fP(3), \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrejit\fP,
860    \fBpcrematching\fP(3), \fBpcrepartial\fP(d), \fBpcrepattern\fP(3),
861    \fBpcreprecompile\fP(3).
862    .
863    .
864  .SH AUTHOR  .SH AUTHOR
865  .rs  .rs
866  .sp  .sp
867  Philip Hazel <ph10@cam.ac.uk>  .nf
868  .br  Philip Hazel
869  University Computing Service,  University Computing Service
870  .br  Cambridge CB2 3QH, England.
871  Cambridge CB2 3QG, England.  .fi
872    .
873  .in 0  .
874  Last updated: 09 December 2003  .SH REVISION
875  .br  .rs
876  Copyright (c) 1997-2003 University of Cambridge.  .sp
877    .nf
878    Last updated: 26 August 2011
879    Copyright (c) 1997-2011 University of Cambridge.
880    .fi

Legend:
Removed from v.73  
changed lines
  Added in v.691

  ViewVC Help
Powered by ViewVC 1.1.5