/[pcre]/code/trunk/doc/pcreapi.3
ViewVC logotype

Contents of /code/trunk/doc/pcreapi.3

Parent Directory Parent Directory | Revision Log Revision Log


Revision 312 - (hide annotations) (download)
Wed Jan 23 18:02:23 2008 UTC (6 years, 2 months ago) by ph10
File size: 84145 byte(s)
Remove the automatic build of pcredemo.

1 nigel 79 .TH PCREAPI 3
2 nigel 63 .SH NAME
3     PCRE - Perl-compatible regular expressions
4 nigel 75 .SH "PCRE NATIVE API"
5 nigel 63 .rs
6     .sp
7     .B #include <pcre.h>
8     .PP
9     .SM
10 nigel 75 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
11 nigel 63 .ti +5n
12 nigel 75 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13 nigel 63 .ti +5n
14 nigel 75 .B const unsigned char *\fItableptr\fP);
15 nigel 63 .PP
16 nigel 77 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17     .ti +5n
18     .B int *\fIerrorcodeptr\fP,
19     .ti +5n
20     .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21     .ti +5n
22     .B const unsigned char *\fItableptr\fP);
23     .PP
24 nigel 75 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25 nigel 63 .ti +5n
26 nigel 75 .B const char **\fIerrptr\fP);
27 nigel 63 .PP
28 nigel 75 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
29 nigel 63 .ti +5n
30 nigel 75 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31 nigel 63 .ti +5n
32 nigel 75 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33 nigel 63 .PP
34 nigel 77 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35     .ti +5n
36     .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37     .ti +5n
38     .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39     .ti +5n
40     .B int *\fIworkspace\fP, int \fIwscount\fP);
41     .PP
42 nigel 75 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
43 nigel 63 .ti +5n
44 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
45 nigel 63 .ti +5n
46 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
47 nigel 63 .ti +5n
48 nigel 75 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49 nigel 63 .PP
50 nigel 75 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
51 nigel 63 .ti +5n
52 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53 nigel 63 .ti +5n
54 nigel 75 .B int \fIbuffersize\fP);
55 nigel 63 .PP
56 nigel 75 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
57 nigel 63 .ti +5n
58 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
59 nigel 63 .ti +5n
60 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
61 nigel 63 .ti +5n
62 nigel 75 .B const char **\fIstringptr\fP);
63 nigel 63 .PP
64 nigel 75 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65 nigel 63 .ti +5n
66 nigel 75 .B const char *\fIname\fP);
67 nigel 63 .PP
68 nigel 91 .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
69     .ti +5n
70     .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71     .PP
72 nigel 75 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
73 nigel 63 .ti +5n
74 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP,
75 nigel 63 .ti +5n
76 nigel 75 .B const char **\fIstringptr\fP);
77 nigel 63 .PP
78 nigel 75 .B int pcre_get_substring_list(const char *\fIsubject\fP,
79 nigel 63 .ti +5n
80 nigel 75 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81 nigel 63 .PP
82 nigel 75 .B void pcre_free_substring(const char *\fIstringptr\fP);
83 nigel 63 .PP
84 nigel 75 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
85 nigel 63 .PP
86     .B const unsigned char *pcre_maketables(void);
87     .PP
88 nigel 75 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
89 nigel 63 .ti +5n
90 nigel 75 .B int \fIwhat\fP, void *\fIwhere\fP);
91 nigel 63 .PP
92 nigel 75 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93     .B *\fIfirstcharptr\fP);
94 nigel 63 .PP
95 nigel 77 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
96     .PP
97 nigel 75 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
98 nigel 63 .PP
99     .B char *pcre_version(void);
100     .PP
101     .B void *(*pcre_malloc)(size_t);
102     .PP
103     .B void (*pcre_free)(void *);
104     .PP
105 nigel 73 .B void *(*pcre_stack_malloc)(size_t);
106     .PP
107     .B void (*pcre_stack_free)(void *);
108     .PP
109 nigel 63 .B int (*pcre_callout)(pcre_callout_block *);
110 nigel 75 .
111     .
112     .SH "PCRE API OVERVIEW"
113 nigel 63 .rs
114     .sp
115 nigel 93 PCRE has its own native API, which is described in this document. There are
116     also some wrapper functions that correspond to the POSIX regular expression
117 nigel 77 API. These are described in the
118 nigel 75 .\" HREF
119     \fBpcreposix\fP
120     .\"
121 nigel 77 documentation. Both of these APIs define a set of C function calls. A C++
122     wrapper is distributed with PCRE. It is documented in the
123     .\" HREF
124     \fBpcrecpp\fP
125     .\"
126     page.
127 nigel 75 .P
128 nigel 77 The native API C function prototypes are defined in the header file
129     \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130     It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131     an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132     and PCRE_MINOR to contain the major and minor release numbers for the library.
133 nigel 75 Applications can use these to include support for different releases of PCRE.
134     .P
135 nigel 77 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136     and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137     in a Perl-compatible manner. A sample program that demonstrates the simplest
138     way of using them is provided in the file called \fIpcredemo.c\fP in the source
139     distribution. The
140 nigel 75 .\" HREF
141     \fBpcresample\fP
142     .\"
143 ph10 312 documentation describes how to compile and run it.
144 nigel 75 .P
145 nigel 77 A second matching function, \fBpcre_dfa_exec()\fP, which is not
146     Perl-compatible, is also provided. This uses a different algorithm for the
147 nigel 91 matching. The alternative algorithm finds all possible matches (at a given
148 nigel 93 point in the subject), and scans the subject just once. However, this algorithm
149     does not return captured substrings. A description of the two matching
150     algorithms and their advantages and disadvantages is given in the
151 nigel 77 .\" HREF
152     \fBpcrematching\fP
153     .\"
154     documentation.
155     .P
156 nigel 75 In addition to the main compiling and matching functions, there are convenience
157 nigel 77 functions for extracting captured substrings from a subject string that is
158     matched by \fBpcre_exec()\fP. They are:
159 nigel 75 .sp
160     \fBpcre_copy_substring()\fP
161     \fBpcre_copy_named_substring()\fP
162     \fBpcre_get_substring()\fP
163     \fBpcre_get_named_substring()\fP
164     \fBpcre_get_substring_list()\fP
165     \fBpcre_get_stringnumber()\fP
166 nigel 91 \fBpcre_get_stringtable_entries()\fP
167 nigel 75 .sp
168     \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
169 nigel 63 provided, to free the memory used for extracted strings.
170 nigel 75 .P
171     The function \fBpcre_maketables()\fP is used to build a set of character tables
172 nigel 77 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
173     or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
174     specialist use. Most commonly, no special tables are passed, in which case
175     internal tables that are generated when PCRE is built are used.
176 nigel 75 .P
177     The function \fBpcre_fullinfo()\fP is used to find out information about a
178     compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
179 nigel 63 some of the available information, but is retained for backwards compatibility.
180 nigel 75 The function \fBpcre_version()\fP returns a pointer to a string containing the
181 nigel 63 version of PCRE and its date of release.
182 nigel 75 .P
183 nigel 77 The function \fBpcre_refcount()\fP maintains a reference count in a data block
184     containing a compiled pattern. This is provided for the benefit of
185     object-oriented applications.
186     .P
187 nigel 75 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
188     the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
189 nigel 63 respectively. PCRE calls the memory management functions via these variables,
190     so a calling program can replace them if it wishes to intercept the calls. This
191     should be done before calling any PCRE functions.
192 nigel 75 .P
193     The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
194 nigel 73 indirections to memory management functions. These special functions are used
195     only when PCRE is compiled to use the heap for remembering data, instead of
196 nigel 91 recursive function calls, when running the \fBpcre_exec()\fP function. See the
197     .\" HREF
198     \fBpcrebuild\fP
199     .\"
200     documentation for details of how to do this. It is a non-standard way of
201     building PCRE, for use in environments that have limited stacks. Because of the
202     greater use of memory management, it runs more slowly. Separate functions are
203     provided so that special-purpose external code can be used for this case. When
204     used, these functions are always called in a stack-like manner (last obtained,
205     first freed), and always for memory blocks of the same size. There is a
206     discussion about PCRE's stack usage in the
207     .\" HREF
208     \fBpcrestack\fP
209     .\"
210     documentation.
211 nigel 75 .P
212     The global variable \fBpcre_callout\fP initially contains NULL. It can be set
213 nigel 63 by the caller to a "callout" function, which PCRE will then call at specified
214 nigel 75 points during a matching operation. Details are given in the
215     .\" HREF
216     \fBpcrecallout\fP
217     .\"
218 nigel 63 documentation.
219 nigel 75 .
220     .
221 ph10 227 .\" HTML <a name="newlines"></a>
222 nigel 91 .SH NEWLINES
223 nigel 93 .rs
224 nigel 91 .sp
225 ph10 149 PCRE supports five different conventions for indicating line breaks in
226 nigel 93 strings: a single CR (carriage return) character, a single LF (linefeed)
227 ph10 149 character, the two-character sequence CRLF, any of the three preceding, or any
228     Unicode newline sequence. The Unicode newline sequences are the three just
229     mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
230     U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
231     (paragraph separator, U+2029).
232 nigel 93 .P
233     Each of the first three conventions is used by at least one operating system as
234     its standard newline sequence. When PCRE is built, a default can be specified.
235     The default default is LF, which is the Unix standard. When PCRE is run, the
236     default can be overridden, either when a pattern is compiled, or when it is
237     matched.
238     .P
239 ph10 227 At compile time, the newline convention can be specified by the \fIoptions\fP
240     argument of \fBpcre_compile()\fP, or it can be specified by special text at the
241     start of the pattern itself; this overrides any other settings. See the
242     .\" HREF
243     \fBpcrepattern\fP
244     .\"
245     page for details of the special character sequences.
246     .P
247 nigel 91 In the PCRE documentation the word "newline" is used to mean "the character or
248 nigel 93 pair of characters that indicate a line break". The choice of newline
249     convention affects the handling of the dot, circumflex, and dollar
250     metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
251     recognized line ending sequence, the match position advancement for a
252 ph10 226 non-anchored pattern. There is more detail about this in the
253     .\" HTML <a href="#execoptions">
254     .\" </a>
255     section on \fBpcre_exec()\fP options
256     .\"
257 ph10 231 below.
258     .P
259     The choice of newline convention does not affect the interpretation of
260     the \en or \er escape sequences, nor does it affect what \eR matches, which is
261     controlled in a similar way, but by separate options.
262 nigel 91 .
263     .
264 nigel 63 .SH MULTITHREADING
265     .rs
266     .sp
267     The PCRE functions can be used in multi-threading applications, with the
268 nigel 75 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
269     \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
270     callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
271     .P
272 nigel 63 The compiled form of a regular expression is not altered during matching, so
273     the same compiled pattern can safely be used by several threads at once.
274 nigel 75 .
275     .
276     .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
277 nigel 63 .rs
278     .sp
279 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
280     time, possibly by a different program, and even on a host other than the one on
281     which it was compiled. Details are given in the
282     .\" HREF
283     \fBpcreprecompile\fP
284     .\"
285 ph10 155 documentation. However, compiling a regular expression with one version of PCRE
286     for use with a different version is not guaranteed to work and may cause
287     crashes.
288 nigel 75 .
289     .
290     .SH "CHECKING BUILD-TIME OPTIONS"
291     .rs
292     .sp
293     .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
294 nigel 63 .PP
295 nigel 75 The function \fBpcre_config()\fP makes it possible for a PCRE client to
296 nigel 63 discover which optional features have been compiled into the PCRE library. The
297     .\" HREF
298 nigel 75 \fBpcrebuild\fP
299 nigel 63 .\"
300     documentation has more details about these optional features.
301 nigel 75 .P
302     The first argument for \fBpcre_config()\fP is an integer, specifying which
303 nigel 63 information is required; the second argument is a pointer to a variable into
304     which the information is placed. The following information is available:
305 nigel 75 .sp
306 nigel 63 PCRE_CONFIG_UTF8
307 nigel 75 .sp
308 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
309     otherwise it is set to zero.
310 nigel 75 .sp
311     PCRE_CONFIG_UNICODE_PROPERTIES
312     .sp
313     The output is an integer that is set to one if support for Unicode character
314     properties is available; otherwise it is set to zero.
315     .sp
316 nigel 63 PCRE_CONFIG_NEWLINE
317 nigel 75 .sp
318 nigel 91 The output is an integer whose value specifies the default character sequence
319 nigel 93 that is recognized as meaning "newline". The four values that are supported
320 ph10 149 are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
321     default should normally be the standard sequence for your operating system.
322 nigel 75 .sp
323 ph10 231 PCRE_CONFIG_BSR
324     .sp
325     The output is an integer whose value indicates what character sequences the \eR
326     escape sequence matches by default. A value of 0 means that \eR matches any
327     Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
328     or CRLF. The default can be overridden when a pattern is compiled or matched.
329     .sp
330 nigel 63 PCRE_CONFIG_LINK_SIZE
331 nigel 75 .sp
332 nigel 63 The output is an integer that contains the number of bytes used for internal
333     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
334     allow larger regular expressions to be compiled, at the expense of slower
335     matching. The default value of 2 is sufficient for all but the most massive
336     patterns, since it allows the compiled pattern to be up to 64K in size.
337 nigel 75 .sp
338 nigel 63 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
339 nigel 75 .sp
340 nigel 63 The output is an integer that contains the threshold above which the POSIX
341 nigel 75 interface uses \fBmalloc()\fP for output vectors. Further details are given in
342     the
343     .\" HREF
344     \fBpcreposix\fP
345     .\"
346     documentation.
347     .sp
348 nigel 63 PCRE_CONFIG_MATCH_LIMIT
349 nigel 75 .sp
350 nigel 63 The output is an integer that gives the default limit for the number of
351 nigel 75 internal matching function calls in a \fBpcre_exec()\fP execution. Further
352     details are given with \fBpcre_exec()\fP below.
353     .sp
354 nigel 87 PCRE_CONFIG_MATCH_LIMIT_RECURSION
355     .sp
356     The output is an integer that gives the default limit for the depth of
357     recursion when calling the internal matching function in a \fBpcre_exec()\fP
358     execution. Further details are given with \fBpcre_exec()\fP below.
359     .sp
360 nigel 73 PCRE_CONFIG_STACKRECURSE
361 nigel 75 .sp
362 nigel 77 The output is an integer that is set to one if internal recursion when running
363     \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
364     to remember their state. This is the usual way that PCRE is compiled. The
365     output is zero if PCRE was compiled to use blocks of data on the heap instead
366     of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
367     \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
368     avoiding the use of the stack.
369 nigel 75 .
370     .
371     .SH "COMPILING A PATTERN"
372 nigel 63 .rs
373     .sp
374 nigel 75 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
375 nigel 63 .ti +5n
376 nigel 75 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
377 nigel 63 .ti +5n
378 nigel 75 .B const unsigned char *\fItableptr\fP);
379 nigel 77 .sp
380     .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
381     .ti +5n
382     .B int *\fIerrorcodeptr\fP,
383     .ti +5n
384     .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
385     .ti +5n
386     .B const unsigned char *\fItableptr\fP);
387 nigel 75 .P
388 nigel 77 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
389     called to compile a pattern into an internal form. The only difference between
390     the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
391     \fIerrorcodeptr\fP, via which a numerical error code can be returned.
392 nigel 75 .P
393 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
394     \fIpattern\fP argument. A pointer to a single block of memory that is obtained
395     via \fBpcre_malloc\fP is returned. This contains the compiled code and related
396     data. The \fBpcre\fP type is defined for the returned block; this is a typedef
397     for a structure whose contents are not externally defined. It is up to the
398 nigel 91 caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
399 nigel 77 .P
400 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
401 nigel 75 depend on memory location, the complete \fBpcre\fP data block is not
402     fully relocatable, because it may contain a copy of the \fItableptr\fP
403     argument, which is an address (see below).
404     .P
405 nigel 93 The \fIoptions\fP argument contains various bit settings that affect the
406 nigel 75 compilation. It should be zero if no options are required. The available
407     options are described below. Some of them, in particular, those that are
408     compatible with Perl, can also be set and unset from within the pattern (see
409     the detailed description in the
410     .\" HREF
411     \fBpcrepattern\fP
412     .\"
413     documentation). For these options, the contents of the \fIoptions\fP argument
414     specifies their initial settings at the start of compilation and execution. The
415 nigel 91 PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
416     matching as well as at compile time.
417 nigel 75 .P
418     If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
419     Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
420     NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
421 nigel 87 error message. This is a static string that is part of the library. You must
422     not try to free it. The offset from the start of the pattern to the character
423     where the error was discovered is placed in the variable pointed to by
424 nigel 75 \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
425     .P
426 nigel 77 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
427     \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
428     returned via this argument in the event of an error. This is in addition to the
429     textual error message. Error codes and messages are listed below.
430     .P
431 nigel 75 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
432     character tables that are built when PCRE is compiled, using the default C
433     locale. Otherwise, \fItableptr\fP must be an address that is the result of a
434     call to \fBpcre_maketables()\fP. This value is stored with the compiled
435     pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
436     passed to it. For more discussion, see the section on locale support below.
437     .P
438     This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
439     .sp
440 nigel 63 pcre *re;
441     const char *error;
442     int erroffset;
443     re = pcre_compile(
444     "^A.*Z", /* the pattern */
445     0, /* default options */
446     &error, /* for error message */
447     &erroffset, /* for error offset */
448     NULL); /* use default character tables */
449 nigel 75 .sp
450     The following names for option bits are defined in the \fBpcre.h\fP header
451     file:
452     .sp
453 nigel 63 PCRE_ANCHORED
454 nigel 75 .sp
455 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
456 nigel 75 constrained to match only at the first matching point in the string that is
457 nigel 63 being searched (the "subject string"). This effect can also be achieved by
458     appropriate constructs in the pattern itself, which is the only way to do it in
459     Perl.
460 nigel 75 .sp
461     PCRE_AUTO_CALLOUT
462     .sp
463     If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
464     all with number 255, before each pattern item. For discussion of the callout
465     facility, see the
466     .\" HREF
467     \fBpcrecallout\fP
468     .\"
469     documentation.
470     .sp
471 ph10 231 PCRE_BSR_ANYCRLF
472     PCRE_BSR_UNICODE
473     .sp
474     These options (which are mutually exclusive) control what the \eR escape
475     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
476     match any Unicode newline sequence. The default is specified when PCRE is
477     built. It can be overridden from within the pattern, or by setting an option
478     when a compiled pattern is matched.
479     .sp
480 nigel 63 PCRE_CASELESS
481 nigel 75 .sp
482 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
483     letters. It is equivalent to Perl's /i option, and it can be changed within a
484 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
485     concept of case for characters whose values are less than 128, so caseless
486     matching is always possible. For characters with higher values, the concept of
487     case is supported if PCRE is compiled with Unicode property support, but not
488     otherwise. If you want to use caseless matching for characters 128 and above,
489     you must ensure that PCRE is compiled with Unicode property support as well as
490     with UTF-8 support.
491 nigel 75 .sp
492 nigel 63 PCRE_DOLLAR_ENDONLY
493 nigel 75 .sp
494 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
495     end of the subject string. Without this option, a dollar also matches
496 nigel 91 immediately before a newline at the end of the string (but not before any other
497     newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
498     There is no equivalent to this option in Perl, and no way to set it within a
499     pattern.
500 nigel 75 .sp
501 nigel 63 PCRE_DOTALL
502 nigel 75 .sp
503 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
504 nigel 91 including those that indicate newline. Without it, a dot does not match when
505     the current position is at a newline. This option is equivalent to Perl's /s
506     option, and it can be changed within a pattern by a (?s) option setting. A
507 nigel 93 negative class such as [^a] always matches newline characters, independent of
508     the setting of this option.
509 nigel 75 .sp
510 nigel 91 PCRE_DUPNAMES
511     .sp
512     If this bit is set, names used to identify capturing subpatterns need not be
513     unique. This can be helpful for certain types of pattern when it is known that
514     only one instance of the named subpattern can ever be matched. There are more
515     details of named subpatterns below; see also the
516     .\" HREF
517     \fBpcrepattern\fP
518     .\"
519     documentation.
520     .sp
521 nigel 63 PCRE_EXTENDED
522 nigel 75 .sp
523 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
524     ignored except when escaped or inside a character class. Whitespace does not
525     include the VT character (code 11). In addition, characters between an
526 nigel 91 unescaped # outside a character class and the next newline, inclusive, are also
527     ignored. This is equivalent to Perl's /x option, and it can be changed within a
528     pattern by a (?x) option setting.
529 nigel 75 .P
530 nigel 63 This option makes it possible to include comments inside complicated patterns.
531     Note, however, that this applies only to data characters. Whitespace characters
532     may never appear within special character sequences in a pattern, for example
533     within the sequence (?( which introduces a conditional subpattern.
534 nigel 75 .sp
535 nigel 63 PCRE_EXTRA
536 nigel 75 .sp
537 nigel 63 This option was invented in order to turn on additional functionality of PCRE
538     that is incompatible with Perl, but it is currently of very little use. When
539     set, any backslash in a pattern that is followed by a letter that has no
540     special meaning causes an error, thus reserving these combinations for future
541     expansion. By default, as in Perl, a backslash followed by a letter with no
542 nigel 91 special meaning is treated as a literal. (Perl can, however, be persuaded to
543     give a warning for this.) There are at present no other features controlled by
544     this option. It can also be set by a (?X) option setting within a pattern.
545 nigel 75 .sp
546 nigel 77 PCRE_FIRSTLINE
547     .sp
548     If this option is set, an unanchored pattern is required to match before or at
549 nigel 91 the first newline in the subject string, though the matched text may continue
550     over the newline.
551 nigel 77 .sp
552 nigel 63 PCRE_MULTILINE
553 nigel 75 .sp
554     By default, PCRE treats the subject string as consisting of a single line of
555     characters (even if it actually contains newlines). The "start of line"
556 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
557     line" metacharacter ($) matches only at the end of the string, or before a
558     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
559     Perl.
560 nigel 75 .P
561 nigel 63 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
562 nigel 91 match immediately following or immediately before internal newlines in the
563     subject string, respectively, as well as at the very start and end. This is
564     equivalent to Perl's /m option, and it can be changed within a pattern by a
565     (?m) option setting. If there are no newlines in a subject string, or no
566 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
567 nigel 75 .sp
568 nigel 91 PCRE_NEWLINE_CR
569     PCRE_NEWLINE_LF
570     PCRE_NEWLINE_CRLF
571 ph10 150 PCRE_NEWLINE_ANYCRLF
572 nigel 93 PCRE_NEWLINE_ANY
573 nigel 91 .sp
574     These options override the default newline definition that was chosen when PCRE
575     was built. Setting the first or the second specifies that a newline is
576 nigel 93 indicated by a single character (CR or LF, respectively). Setting
577     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
578 ph10 149 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
579     preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
580     that any Unicode newline sequence should be recognized. The Unicode newline
581     sequences are the three just mentioned, plus the single characters VT (vertical
582     tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
583     separator, U+2028), and PS (paragraph separator, U+2029). The last two are
584     recognized only in UTF-8 mode.
585 nigel 91 .P
586 nigel 93 The newline setting in the options word uses three bits that are treated
587 ph10 149 as a number, giving eight possibilities. Currently only six are used (default
588     plus the five values above). This means that if you set more than one newline
589 nigel 93 option, the combination may or may not be sensible. For example,
590     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
591 ph10 149 other combinations may yield unused numbers and cause an error.
592 nigel 93 .P
593     The only time that a line break is specially recognized when compiling a
594     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
595     class is encountered. This indicates a comment that lasts until after the next
596     line break sequence. In other circumstances, line break sequences are treated
597     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
598     as whitespace characters and are therefore ignored.
599     .P
600     The newline option that is set at compile time becomes the default that is used
601     for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
602 nigel 91 .sp
603 nigel 63 PCRE_NO_AUTO_CAPTURE
604 nigel 75 .sp
605 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
606     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
607     were followed by ?: but named parentheses can still be used for capturing (and
608     they acquire numbers in the usual way). There is no equivalent of this option
609     in Perl.
610 nigel 75 .sp
611 nigel 63 PCRE_UNGREEDY
612 nigel 75 .sp
613 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
614     greedy by default, but become greedy if followed by "?". It is not compatible
615     with Perl. It can also be set by a (?U) option setting within the pattern.
616 nigel 75 .sp
617 nigel 63 PCRE_UTF8
618 nigel 75 .sp
619 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
620     of UTF-8 characters instead of single-byte character strings. However, it is
621 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
622 nigel 63 of this option provokes an error. Details of how this option changes the
623     behaviour of PCRE are given in the
624     .\" HTML <a href="pcre.html#utf8support">
625     .\" </a>
626     section on UTF-8 support
627     .\"
628     in the main
629     .\" HREF
630 nigel 75 \fBpcre\fP
631 nigel 63 .\"
632     page.
633 nigel 75 .sp
634 nigel 71 PCRE_NO_UTF8_CHECK
635 nigel 75 .sp
636 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
637 ph10 211 automatically checked. There is a discussion about the
638     .\" HTML <a href="pcre.html#utf8strings">
639     .\" </a>
640     validity of UTF-8 strings
641     .\"
642     in the main
643     .\" HREF
644     \fBpcre\fP
645     .\"
646     page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
647     returns an error. If you already know that your pattern is valid, and you want
648     to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
649     option. When it is set, the effect of passing an invalid UTF-8 string as a
650     pattern is undefined. It may cause your program to crash. Note that this option
651     can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
652     the UTF-8 validity checking of subject strings.
653 nigel 75 .
654     .
655 nigel 77 .SH "COMPILATION ERROR CODES"
656     .rs
657     .sp
658     The following table lists the error codes than may be returned by
659     \fBpcre_compile2()\fP, along with the error messages that may be returned by
660 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
661     out of use. To avoid confusion, they have not been re-used.
662 nigel 77 .sp
663     0 no error
664     1 \e at end of pattern
665     2 \ec at end of pattern
666     3 unrecognized character follows \e
667     4 numbers out of order in {} quantifier
668     5 number too big in {} quantifier
669     6 missing terminating ] for character class
670     7 invalid escape sequence in character class
671     8 range out of order in character class
672     9 nothing to repeat
673 nigel 93 10 [this code is not in use]
674 nigel 77 11 internal error: unexpected repeat
675 ph10 290 12 unrecognized character after (? or (?-
676 nigel 77 13 POSIX named classes are supported only within a class
677     14 missing )
678     15 reference to non-existent subpattern
679     16 erroffset passed as NULL
680     17 unknown option bit(s) set
681     18 missing ) after comment
682 nigel 93 19 [this code is not in use]
683 ph10 290 20 regular expression is too large
684 nigel 77 21 failed to get memory
685     22 unmatched parentheses
686     23 internal error: code overflow
687     24 unrecognized character after (?<
688     25 lookbehind assertion is not fixed length
689 nigel 91 26 malformed number or name after (?(
690 nigel 77 27 conditional group contains more than two branches
691     28 assertion expected after (?(
692 ph10 181 29 (?R or (?[+-]digits must be followed by )
693 nigel 77 30 unknown POSIX class name
694     31 POSIX collating elements are not supported
695     32 this version of PCRE is not compiled with PCRE_UTF8 support
696 nigel 93 33 [this code is not in use]
697 nigel 77 34 character value in \ex{...} sequence is too large
698     35 invalid condition (?(0)
699     36 \eC not allowed in lookbehind assertion
700     37 PCRE does not support \eL, \el, \eN, \eU, or \eu
701     38 number after (?C is > 255
702     39 closing ) for (?C expected
703     40 recursive call could loop indefinitely
704     41 unrecognized character after (?P
705 nigel 93 42 syntax error in subpattern name (missing terminator)
706 nigel 91 43 two named subpatterns have the same name
707 nigel 77 44 invalid UTF-8 string
708     45 support for \eP, \ep, and \eX has not been compiled
709     46 malformed \eP or \ep sequence
710     47 unknown property name after \eP or \ep
711 nigel 91 48 subpattern name is too long (maximum 32 characters)
712 ph10 290 49 too many named subpatterns (maximum 10000)
713 ph10 202 50 [this code is not in use]
714 nigel 91 51 octal value is greater than \e377 (not in UTF-8 mode)
715 nigel 93 52 internal error: overran compiling workspace
716     53 internal error: previously-checked referenced subpattern not found
717     54 DEFINE group contains more than one branch
718     55 repeating a DEFINE group is not allowed
719 ph10 231 56 inconsistent NEWLINE options
720 ph10 182 57 \eg is not followed by a braced name or an optionally braced
721 ph10 181 non-zero number
722     58 (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
723 ph10 290 59 (*VERB) with an argument is not supported
724     60 (*VERB) not recognized
725     61 number is too big
726     62 subpattern name expected
727 ph10 292 63 digit expected after (?+
728 ph10 290 .sp
729 ph10 292 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
730 ph10 290 be used if the limits were changed when PCRE was built.
731 nigel 77 .
732     .
733 nigel 75 .SH "STUDYING A PATTERN"
734 nigel 63 .rs
735     .sp
736 nigel 77 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
737 nigel 63 .ti +5n
738 nigel 75 .B const char **\fIerrptr\fP);
739 nigel 63 .PP
740 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
741     more time analyzing it in order to speed up the time taken for matching. The
742     function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
743     argument. If studying the pattern produces additional information that will
744     help speed up matching, \fBpcre_study()\fP returns a pointer to a
745     \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
746     results of the study.
747     .P
748     The returned value from \fBpcre_study()\fP can be passed directly to
749     \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
750 nigel 63 fields that can be set by the caller before the block is passed; these are
751 nigel 75 described
752     .\" HTML <a href="#extradata">
753     .\" </a>
754     below
755     .\"
756     in the section on matching a pattern.
757     .P
758 nigel 77 If studying the pattern does not produce any additional information
759 nigel 75 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
760     wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
761     own \fBpcre_extra\fP block.
762     .P
763     The second argument of \fBpcre_study()\fP contains option bits. At present, no
764     options are defined, and this argument should always be zero.
765     .P
766     The third argument for \fBpcre_study()\fP is a pointer for an error message. If
767 nigel 63 studying succeeds (even if no data is returned), the variable it points to is
768 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
769     static string that is part of the library. You must not try to free it. You
770     should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
771     sure that it has run successfully.
772 nigel 75 .P
773     This is a typical call to \fBpcre_study\fP():
774     .sp
775 nigel 63 pcre_extra *pe;
776     pe = pcre_study(
777     re, /* result of pcre_compile() */
778     0, /* no options exist */
779     &error); /* set to NULL or points to a message */
780 nigel 75 .sp
781 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
782     not have a single fixed starting character. A bitmap of possible starting
783 nigel 75 bytes is created.
784     .
785     .
786 nigel 63 .\" HTML <a name="localesupport"></a>
787 nigel 75 .SH "LOCALE SUPPORT"
788 nigel 63 .rs
789     .sp
790 ph10 139 PCRE handles caseless matching, and determines whether characters are letters,
791 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
792 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
793 nigel 75 less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
794     can be tested with \ep if PCRE is built with Unicode character property
795 ph10 142 support. The use of locales with Unicode is discouraged. If you are handling
796     characters with codes greater than 128, you should either use UTF-8 and
797 ph10 139 Unicode, or use locales, but not try to mix the two.
798 nigel 75 .P
799 ph10 139 PCRE contains an internal set of tables that are used when the final argument
800     of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
801 ph10 142 Normally, the internal tables recognize only ASCII characters. However, when
802 ph10 139 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
803     default "C" locale of the local system, which may cause them to be different.
804 nigel 75 .P
805 ph10 139 The internal tables can always be overridden by tables supplied by the
806     application that calls PCRE. These may be created in a different locale from
807     the default. As more and more applications change to using Unicode, the need
808     for this locale support is expected to die away.
809     .P
810 nigel 75 External tables are built by calling the \fBpcre_maketables()\fP function,
811     which has no arguments, in the relevant locale. The result can then be passed
812     to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
813     example, to build and use tables that are appropriate for the French locale
814     (where accented characters with values greater than 128 are treated as letters),
815     the following code could be used:
816     .sp
817     setlocale(LC_CTYPE, "fr_FR");
818 nigel 63 tables = pcre_maketables();
819     re = pcre_compile(..., tables);
820 nigel 75 .sp
821 ph10 142 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
822 ph10 139 are using Windows, the name for the French locale is "french".
823     .P
824 nigel 75 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
825     obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
826     that the memory containing the tables remains available for as long as it is
827     needed.
828     .P
829     The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
830     pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
831     and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
832     pattern, compilation, studying and matching all happen in the same locale, but
833     different patterns can be compiled in different locales.
834     .P
835     It is possible to pass a table pointer or NULL (indicating the use of the
836     internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
837     this facility could be used to match a pattern in a different locale from the
838     one in which it was compiled. Passing table pointers at run time is discussed
839     below in the section on matching a pattern.
840     .
841     .
842     .SH "INFORMATION ABOUT A PATTERN"
843 nigel 63 .rs
844     .sp
845 nigel 75 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
846 nigel 63 .ti +5n
847 nigel 75 .B int \fIwhat\fP, void *\fIwhere\fP);
848 nigel 63 .PP
849 nigel 75 The \fBpcre_fullinfo()\fP function returns information about a compiled
850     pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
851 nigel 63 nevertheless retained for backwards compability (and is documented below).
852 nigel 75 .P
853     The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
854     pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
855 nigel 63 the pattern was not studied. The third argument specifies which piece of
856     information is required, and the fourth argument is a pointer to a variable
857     to receive the data. The yield of the function is zero for success, or one of
858     the following negative numbers:
859 nigel 75 .sp
860     PCRE_ERROR_NULL the argument \fIcode\fP was NULL
861     the argument \fIwhere\fP was NULL
862 nigel 63 PCRE_ERROR_BADMAGIC the "magic number" was not found
863 nigel 75 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
864     .sp
865     The "magic number" is placed at the start of each compiled pattern as an simple
866     check against passing an arbitrary memory pointer. Here is a typical call of
867     \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
868     .sp
869 nigel 63 int rc;
870 nigel 91 size_t length;
871 nigel 63 rc = pcre_fullinfo(
872     re, /* result of pcre_compile() */
873     pe, /* result of pcre_study(), or NULL */
874     PCRE_INFO_SIZE, /* what is required */
875     &length); /* where to put the data */
876 nigel 75 .sp
877     The possible values for the third argument are defined in \fBpcre.h\fP, and are
878 nigel 63 as follows:
879 nigel 75 .sp
880 nigel 63 PCRE_INFO_BACKREFMAX
881 nigel 75 .sp
882 nigel 63 Return the number of the highest back reference in the pattern. The fourth
883 nigel 75 argument should point to an \fBint\fP variable. Zero is returned if there are
884 nigel 63 no back references.
885 nigel 75 .sp
886 nigel 63 PCRE_INFO_CAPTURECOUNT
887 nigel 75 .sp
888 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
889 nigel 75 should point to an \fBint\fP variable.
890     .sp
891 nigel 77 PCRE_INFO_DEFAULT_TABLES
892 nigel 75 .sp
893     Return a pointer to the internal default character tables within PCRE. The
894     fourth argument should point to an \fBunsigned char *\fP variable. This
895     information call is provided for internal use by the \fBpcre_study()\fP
896     function. External callers can cause PCRE to use its internal tables by passing
897     a NULL table pointer.
898     .sp
899 nigel 63 PCRE_INFO_FIRSTBYTE
900 nigel 75 .sp
901 nigel 63 Return information about the first byte of any matched string, for a
902 nigel 91 non-anchored pattern. The fourth argument should point to an \fBint\fP
903     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
904     still recognized for backwards compatibility.)
905 nigel 75 .P
906     If there is a fixed first byte, for example, from a pattern such as
907 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
908 nigel 75 .sp
909 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
910     starts with "^", or
911 nigel 75 .sp
912 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
913     (if it were set, the pattern would be anchored),
914 nigel 75 .sp
915 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
916     subject string or after any newline within the string. Otherwise -2 is
917     returned. For anchored patterns, -2 is returned.
918 nigel 75 .sp
919 nigel 63 PCRE_INFO_FIRSTTABLE
920 nigel 75 .sp
921 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
922     table indicating a fixed set of bytes for the first byte in any matching
923     string, a pointer to the table is returned. Otherwise NULL is returned. The
924 nigel 75 fourth argument should point to an \fBunsigned char *\fP variable.
925     .sp
926 ph10 226 PCRE_INFO_HASCRORLF
927     .sp
928 ph10 227 Return 1 if the pattern contains any explicit matches for CR or LF characters,
929 ph10 243 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
930 ph10 231 explicit match is either a literal CR or LF character, or \er or \en.
931 ph10 226 .sp
932 ph10 169 PCRE_INFO_JCHANGED
933     .sp
934 ph10 278 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
935     0. The fourth argument should point to an \fBint\fP variable. (?J) and
936     (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
937 ph10 169 .sp
938 nigel 63 PCRE_INFO_LASTLITERAL
939 nigel 75 .sp
940 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
941     string, other than at its start, if such a byte has been recorded. The fourth
942 nigel 75 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
943 nigel 65 returned. For anchored patterns, a last literal byte is recorded only if it
944     follows something of variable length. For example, for the pattern
945 nigel 75 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
946 nigel 65 is -1.
947 nigel 75 .sp
948 nigel 63 PCRE_INFO_NAMECOUNT
949     PCRE_INFO_NAMEENTRYSIZE
950     PCRE_INFO_NAMETABLE
951 nigel 75 .sp
952 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
953     names are just an additional way of identifying the parentheses, which still
954 nigel 91 acquire numbers. Several convenience functions such as
955     \fBpcre_get_named_substring()\fP are provided for extracting captured
956     substrings by name. It is also possible to extract the data directly, by first
957     converting the name to a number in order to access the correct pointers in the
958     output vector (described with \fBpcre_exec()\fP below). To do the conversion,
959     you need to use the name-to-number map, which is described by these three
960     values.
961 nigel 75 .P
962 nigel 63 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
963     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
964 nigel 75 entry; both of these return an \fBint\fP value. The entry size depends on the
965 nigel 63 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
966 nigel 75 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
967 nigel 63 are the number of the capturing parenthesis, most significant byte first. The
968     rest of the entry is the corresponding name, zero terminated. The names are in
969 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
970     their parentheses numbers. For example, consider the following pattern (assume
971 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
972 nigel 75 .sp
973     .\" JOIN
974 nigel 93 (?<date> (?<year>(\ed\ed)?\ed\ed) -
975     (?<month>\ed\ed) - (?<day>\ed\ed) )
976 nigel 75 .sp
977 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
978     in the table is eight bytes long. The table is as follows, with non-printing
979 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
980     .sp
981 nigel 63 00 01 d a t e 00 ??
982     00 05 d a y 00 ?? ??
983     00 04 m o n t h 00
984     00 02 y e a r 00 ??
985 nigel 75 .sp
986     When writing code to extract data from named subpatterns using the
987 nigel 91 name-to-number map, remember that the length of the entries is likely to be
988 nigel 75 different for each compiled pattern.
989     .sp
990 ph10 169 PCRE_INFO_OKPARTIAL
991     .sp
992 ph10 172 Return 1 if the pattern can be used for partial matching, otherwise 0. The
993 ph10 169 fourth argument should point to an \fBint\fP variable. The
994     .\" HREF
995     \fBpcrepartial\fP
996     .\"
997     documentation lists the restrictions that apply to patterns when partial
998     matching is used.
999     .sp
1000 nigel 63 PCRE_INFO_OPTIONS
1001 nigel 75 .sp
1002 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
1003 nigel 75 argument should point to an \fBunsigned long int\fP variable. These option bits
1004     are those specified in the call to \fBpcre_compile()\fP, modified by any
1005 ph10 196 top-level option settings at the start of the pattern itself. In other words,
1006     they are the options that will be in force when matching starts. For example,
1007     if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1008     result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1009 nigel 75 .P
1010 nigel 63 A pattern is automatically anchored by PCRE if all of its top-level
1011     alternatives begin with one of the following:
1012 nigel 75 .sp
1013 nigel 63 ^ unless PCRE_MULTILINE is set
1014 nigel 75 \eA always
1015     \eG always
1016     .\" JOIN
1017 nigel 63 .* if PCRE_DOTALL is set and there are no back
1018     references to the subpattern in which .* appears
1019 nigel 75 .sp
1020 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1021 nigel 75 \fBpcre_fullinfo()\fP.
1022     .sp
1023 nigel 63 PCRE_INFO_SIZE
1024 nigel 75 .sp
1025 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
1026 nigel 75 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1027     place the compiled data. The fourth argument should point to a \fBsize_t\fP
1028 nigel 63 variable.
1029 nigel 75 .sp
1030 nigel 63 PCRE_INFO_STUDYSIZE
1031 nigel 75 .sp
1032     Return the size of the data block pointed to by the \fIstudy_data\fP field in
1033     a \fBpcre_extra\fP block. That is, it is the value that was passed to
1034     \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1035     created by \fBpcre_study()\fP. The fourth argument should point to a
1036     \fBsize_t\fP variable.
1037     .
1038     .
1039     .SH "OBSOLETE INFO FUNCTION"
1040 nigel 63 .rs
1041     .sp
1042 nigel 75 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1043     .B *\fIfirstcharptr\fP);
1044 nigel 63 .PP
1045 nigel 75 The \fBpcre_info()\fP function is now obsolete because its interface is too
1046 nigel 63 restrictive to return all the available data about a compiled pattern. New
1047 nigel 75 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1048     \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1049 nigel 63 following negative numbers:
1050 nigel 75 .sp
1051     PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1052 nigel 63 PCRE_ERROR_BADMAGIC the "magic number" was not found
1053 nigel 75 .sp
1054     If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1055 nigel 63 pattern was compiled is placed in the integer it points to (see
1056     PCRE_INFO_OPTIONS above).
1057 nigel 75 .P
1058     If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1059 nigel 63 it is used to pass back information about the first character of any matched
1060     string (see PCRE_INFO_FIRSTBYTE above).
1061 nigel 75 .
1062     .
1063 nigel 77 .SH "REFERENCE COUNTS"
1064 nigel 63 .rs
1065     .sp
1066 nigel 77 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1067     .PP
1068     The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1069     data block that contains a compiled pattern. It is provided for the benefit of
1070     applications that operate in an object-oriented manner, where different parts
1071     of the application may be using the same compiled pattern, but you want to free
1072     the block when they are all done.
1073     .P
1074     When a pattern is compiled, the reference count field is initialized to zero.
1075     It is changed only by calling this function, whose action is to add the
1076     \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1077     function is the new value. However, the value of the count is constrained to
1078     lie between 0 and 65535, inclusive. If the new value is outside these limits,
1079     it is forced to the appropriate limit value.
1080     .P
1081     Except when it is zero, the reference count is not correctly preserved if a
1082     pattern is compiled on one host and then transferred to a host whose byte-order
1083     is different. (This seems a highly unlikely scenario.)
1084     .
1085     .
1086     .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1087     .rs
1088     .sp
1089 nigel 75 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1090 nigel 63 .ti +5n
1091 nigel 75 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1092 nigel 63 .ti +5n
1093 nigel 75 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1094     .P
1095     The function \fBpcre_exec()\fP is called to match a subject string against a
1096     compiled pattern, which is passed in the \fIcode\fP argument. If the
1097 nigel 63 pattern has been studied, the result of the study should be passed in the
1098 nigel 77 \fIextra\fP argument. This function is the main matching facility of the
1099     library, and it operates in a Perl-like manner. For specialist use there is
1100     also an alternative matching function, which is described
1101     .\" HTML <a href="#dfamatch">
1102     .\" </a>
1103     below
1104     .\"
1105     in the section about the \fBpcre_dfa_exec()\fP function.
1106 nigel 75 .P
1107     In most applications, the pattern will have been compiled (and optionally
1108     studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1109     possible to save compiled patterns and study data, and then use them later
1110     in different processes, possibly even on different hosts. For a discussion
1111     about this, see the
1112     .\" HREF
1113     \fBpcreprecompile\fP
1114     .\"
1115     documentation.
1116     .P
1117     Here is an example of a simple call to \fBpcre_exec()\fP:
1118     .sp
1119 nigel 63 int rc;
1120     int ovector[30];
1121     rc = pcre_exec(
1122     re, /* result of pcre_compile() */
1123     NULL, /* we didn't study the pattern */
1124     "some string", /* the subject string */
1125     11, /* the length of the subject string */
1126     0, /* start at offset 0 in the subject */
1127     0, /* default options */
1128 nigel 75 ovector, /* vector of integers for substring information */
1129 nigel 77 30); /* number of elements (NOT size in bytes) */
1130 nigel 75 .
1131     .\" HTML <a name="extradata"></a>
1132     .SS "Extra data for \fBpcre_exec()\fR"
1133     .rs
1134     .sp
1135     If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1136     data block. The \fBpcre_study()\fP function returns such a block (when it
1137 nigel 63 doesn't return NULL), but you can also create one for yourself, and pass
1138 nigel 87 additional information in it. The \fBpcre_extra\fP block contains the following
1139     fields (not necessarily in this order):
1140 nigel 75 .sp
1141     unsigned long int \fIflags\fP;
1142     void *\fIstudy_data\fP;
1143     unsigned long int \fImatch_limit\fP;
1144 nigel 87 unsigned long int \fImatch_limit_recursion\fP;
1145 nigel 75 void *\fIcallout_data\fP;
1146     const unsigned char *\fItables\fP;
1147     .sp
1148     The \fIflags\fP field is a bitmap that specifies which of the other fields
1149 nigel 63 are set. The flag bits are:
1150 nigel 75 .sp
1151 nigel 63 PCRE_EXTRA_STUDY_DATA
1152     PCRE_EXTRA_MATCH_LIMIT
1153 nigel 87 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1154 nigel 63 PCRE_EXTRA_CALLOUT_DATA
1155 nigel 75 PCRE_EXTRA_TABLES
1156     .sp
1157     Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1158     \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1159     the appropriate flag bit. You should not set this yourself, but you may add to
1160     the block by setting the other fields and their corresponding flag bits.
1161     .P
1162     The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1163 nigel 63 vast amount of resources when running patterns that are not going to match,
1164     but which have a very large number of possibilities in their search trees. The
1165 nigel 75 classic example is the use of nested unlimited repeats.
1166     .P
1167     Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1168 nigel 87 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1169     number of times this function is called during a match, which has the effect of
1170     limiting the amount of backtracking that can take place. For patterns that are
1171     not anchored, the count restarts from zero for each position in the subject
1172     string.
1173 nigel 75 .P
1174 nigel 87 The default value for the limit can be set when PCRE is built; the default
1175 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1176 nigel 87 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1177     block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1178     the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1179     PCRE_ERROR_MATCHLIMIT.
1180 nigel 75 .P
1181 nigel 87 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1182     instead of limiting the total number of times that \fBmatch()\fP is called, it
1183     limits the depth of recursion. The recursion depth is a smaller number than the
1184     total number of calls, because not all calls to \fBmatch()\fP are recursive.
1185     This limit is of use only if it is set smaller than \fImatch_limit\fP.
1186     .P
1187     Limiting the recursion depth limits the amount of stack that can be used, or,
1188     when PCRE has been compiled to use memory on the heap instead of the stack, the
1189     amount of heap memory that can be used.
1190     .P
1191     The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1192     built; the default default is the same value as the default for
1193     \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1194     with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1195     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1196     is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1197     .P
1198 nigel 75 The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1199     which is described in the
1200     .\" HREF
1201     \fBpcrecallout\fP
1202     .\"
1203     documentation.
1204     .P
1205     The \fItables\fP field is used to pass a character tables pointer to
1206     \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1207     pattern. A non-NULL value is stored with the compiled pattern only if custom
1208     tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1209     If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1210     internal tables to be used. This facility is helpful when re-using patterns
1211     that have been saved after compiling with an external set of tables, because
1212     the external tables might be at a different address when \fBpcre_exec()\fP is
1213     called. See the
1214     .\" HREF
1215     \fBpcreprecompile\fP
1216     .\"
1217     documentation for a discussion of saving compiled patterns for later use.
1218     .
1219 ph10 226 .\" HTML <a name="execoptions"></a>
1220 nigel 75 .SS "Option bits for \fBpcre_exec()\fP"
1221     .rs
1222     .sp
1223     The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1224 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1225     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1226 nigel 75 .sp
1227     PCRE_ANCHORED
1228     .sp
1229     The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1230     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1231     to be anchored by virtue of its contents, it cannot be made unachored at
1232     matching time.
1233     .sp
1234 ph10 231 PCRE_BSR_ANYCRLF
1235     PCRE_BSR_UNICODE
1236     .sp
1237     These options (which are mutually exclusive) control what the \eR escape
1238     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1239     match any Unicode newline sequence. These options override the choice that was
1240     made or defaulted when the pattern was compiled.
1241     .sp
1242 nigel 91 PCRE_NEWLINE_CR
1243     PCRE_NEWLINE_LF
1244     PCRE_NEWLINE_CRLF
1245 ph10 150 PCRE_NEWLINE_ANYCRLF
1246 nigel 93 PCRE_NEWLINE_ANY
1247 nigel 91 .sp
1248     These options override the newline definition that was chosen or defaulted when
1249 nigel 93 the pattern was compiled. For details, see the description of
1250     \fBpcre_compile()\fP above. During matching, the newline choice affects the
1251     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1252     the way the match position is advanced after a match failure for an unanchored
1253 ph10 227 pattern.
1254 ph10 225 .P
1255 ph10 226 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1256     match attempt for an unanchored pattern fails when the current position is at a
1257 ph10 230 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1258 ph10 226 characters, the match position is advanced by two characters instead of one, in
1259     other words, to after the CRLF.
1260     .P
1261 ph10 227 The above rule is a compromise that makes the most common cases work as
1262     expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1263     set), it does not match the string "\er\enA" because, after failing at the
1264     start, it skips both the CR and the LF before retrying. However, the pattern
1265     [\er\en]A does match that string, because it contains an explicit CR or LF
1266 ph10 226 reference, and so advances only by one character after the first failure.
1267     .P
1268 ph10 231 An explicit match for CR of LF is either a literal appearance of one of those
1269     characters, or one of the \er or \en escape sequences. Implicit matches such as
1270 ph10 230 [^X] do not count, nor does \es (which includes CR and LF in the characters
1271     that it matches).
1272     .P
1273 ph10 226 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1274     valid newline sequence and explicit \er or \en escapes appear in the pattern.
1275 nigel 91 .sp
1276 nigel 63 PCRE_NOTBOL
1277 nigel 75 .sp
1278     This option specifies that first character of the subject string is not the
1279     beginning of a line, so the circumflex metacharacter should not match before
1280     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1281     never to match. This option affects only the behaviour of the circumflex
1282     metacharacter. It does not affect \eA.
1283     .sp
1284 nigel 63 PCRE_NOTEOL
1285 nigel 75 .sp
1286     This option specifies that the end of the subject string is not the end of a
1287     line, so the dollar metacharacter should not match it nor (except in multiline
1288     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1289     compile time) causes dollar never to match. This option affects only the
1290     behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1291     .sp
1292 nigel 63 PCRE_NOTEMPTY
1293 nigel 75 .sp
1294 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1295     there are alternatives in the pattern, they are tried. If all the alternatives
1296     match the empty string, the entire match fails. For example, if the pattern
1297 nigel 75 .sp
1298 nigel 63 a?b?
1299 nigel 75 .sp
1300 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1301     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1302     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1303 nigel 75 .P
1304 nigel 63 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1305 nigel 75 of a pattern match of the empty string within its \fBsplit()\fP function, and
1306 nigel 63 when using the /g modifier. It is possible to emulate Perl's behaviour after
1307     matching a null string by first trying the match again at the same offset with
1308 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1309     starting offset (see below) and trying an ordinary match again. There is some
1310     code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1311     .sp
1312     PCRE_NO_UTF8_CHECK
1313     .sp
1314     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1315     string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1316 ph10 211 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1317     start of a UTF-8 character. There is a discussion about the validity of UTF-8
1318     strings in the
1319     .\" HTML <a href="pcre.html#utf8strings">
1320     .\" </a>
1321     section on UTF-8 support
1322     .\"
1323     in the main
1324     .\" HREF
1325     \fBpcre\fP
1326     .\"
1327     page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1328     the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1329 ph10 209 PCRE_ERROR_BADUTF8_OFFSET is returned.
1330 nigel 75 .P
1331     If you already know that your subject is valid, and you want to skip these
1332     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1333     calling \fBpcre_exec()\fP. You might want to do this for the second and
1334     subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1335     all the matches in a single subject string. However, you should be sure that
1336     the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1337     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1338     subject, or a value of \fIstartoffset\fP that does not point to the start of a
1339     UTF-8 character, is undefined. Your program may crash.
1340     .sp
1341     PCRE_PARTIAL
1342     .sp
1343     This option turns on the partial matching feature. If the subject string fails
1344     to match the pattern, but at some point during the matching process the end of
1345     the subject was reached (that is, the subject partially matches the pattern and
1346     the failure to match occurred only because there were not enough subject
1347     characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1348     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1349     may appear in the pattern. These are discussed in the
1350     .\" HREF
1351     \fBpcrepartial\fP
1352     .\"
1353     documentation.
1354     .
1355     .SS "The string to be matched by \fBpcre_exec()\fP"
1356     .rs
1357     .sp
1358     The subject string is passed to \fBpcre_exec()\fP as a pointer in
1359     \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
1360     \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
1361     UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1362     bytes. When the starting offset is zero, the search for a match starts at the
1363     beginning of the subject, and this is by far the most common case.
1364     .P
1365 nigel 63 A non-zero starting offset is useful when searching for another match in the
1366 nigel 75 same subject by calling \fBpcre_exec()\fP again after a previous success.
1367     Setting \fIstartoffset\fP differs from just passing over a shortened string and
1368 nigel 63 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1369     lookbehind. For example, consider the pattern
1370 nigel 75 .sp
1371     \eBiss\eB
1372     .sp
1373     which finds occurrences of "iss" in the middle of words. (\eB matches only if
1374 nigel 63 the current position in the subject is not a word boundary.) When applied to
1375 nigel 75 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1376     occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1377     subject, namely "issipi", it does not match, because \eB is always false at the
1378 nigel 63 start of the subject, which is deemed to be a word boundary. However, if
1379 nigel 75 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1380 nigel 63 set to 4, it finds the second occurrence of "iss" because it is able to look
1381     behind the starting point to discover that it is preceded by a letter.
1382 nigel 75 .P
1383 nigel 63 If a non-zero starting offset is passed when the pattern is anchored, one
1384 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1385 nigel 63 pattern does not require the match to be at the start of the subject.
1386 nigel 75 .
1387     .SS "How \fBpcre_exec()\fP returns captured substrings"
1388     .rs
1389     .sp
1390 nigel 63 In general, a pattern matches a certain portion of the subject, and in
1391     addition, further substrings from the subject may be picked out by parts of the
1392     pattern. Following the usage in Jeffrey Friedl's book, this is called
1393     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1394     a fragment of a pattern that picks out a substring. PCRE supports several other
1395     kinds of parenthesized subpattern that do not cause substrings to be captured.
1396 nigel 75 .P
1397 nigel 63 Captured substrings are returned to the caller via a vector of integer offsets
1398 nigel 75 whose address is passed in \fIovector\fP. The number of elements in the vector
1399     is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:
1400     this argument is NOT the size of \fIovector\fP in bytes.
1401     .P
1402     The first two-thirds of the vector is used to pass back captured substrings,
1403     each substring using a pair of integers. The remaining third of the vector is
1404     used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1405     and is not available for passing back information. The length passed in
1406     \fIovecsize\fP should always be a multiple of three. If it is not, it is
1407     rounded down.
1408     .P
1409     When a match is successful, information about captured substrings is returned
1410     in pairs of integers, starting at the beginning of \fIovector\fP, and
1411 nigel 63 continuing up to two-thirds of its length at the most. The first element of a
1412     pair is set to the offset of the first character in a substring, and the second
1413     is set to the offset of the first character after the end of a substring. The
1414 nigel 75 first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
1415 nigel 63 subject string matched by the entire pattern. The next pair is used for the
1416 nigel 75 first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
1417 nigel 91 is one more than the highest numbered pair that has been set. For example, if
1418     two substrings have been captured, the returned value is 3. If there are no
1419     capturing subpatterns, the return value from a successful match is 1,
1420     indicating that just the first pair of offsets has been set.
1421 nigel 75 .P
1422 nigel 63 If a capturing subpattern is matched repeatedly, it is the last portion of the
1423 nigel 75 string that it matched that is returned.
1424     .P
1425     If the vector is too small to hold all the captured substring offsets, it is
1426     used as far as possible (up to two-thirds of its length), and the function
1427     returns a value of zero. In particular, if the substring offsets are not of
1428     interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1429     \fIovecsize\fP as zero. However, if the pattern contains back references and
1430     the \fIovector\fP is not big enough to remember the related substrings, PCRE
1431     has to get additional memory for use during matching. Thus it is usually
1432     advisable to supply an \fIovector\fP.
1433     .P
1434 nigel 91 The \fBpcre_info()\fP function can be used to find out how many capturing
1435 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1436 nigel 75 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1437     the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1438 nigel 91 .P
1439     It is possible for capturing subpattern number \fIn+1\fP to match some part of
1440     the subject when subpattern \fIn\fP has not been used at all. For example, if
1441     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1442     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1443     happens, both values in the offset pairs corresponding to unused subpatterns
1444     are set to -1.
1445     .P
1446     Offset values that correspond to unused subpatterns at the end of the
1447     expression are also set to -1. For example, if the string "abc" is matched
1448     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1449     return from the function is 2, because the highest used capturing subpattern
1450     number is 1. However, you can refer to the offsets for the second and third
1451     capturing subpatterns if you wish (assuming the vector is large enough, of
1452     course).
1453     .P
1454     Some convenience functions are provided for extracting the captured substrings
1455     as separate strings. These are described below.
1456 nigel 75 .
1457 nigel 77 .\" HTML <a name="errorlist"></a>
1458 nigel 91 .SS "Error return values from \fBpcre_exec()\fP"
1459 nigel 75 .rs
1460     .sp
1461     If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1462 nigel 63 defined in the header file:
1463 nigel 75 .sp
1464 nigel 63 PCRE_ERROR_NOMATCH (-1)
1465 nigel 75 .sp
1466 nigel 63 The subject string did not match the pattern.
1467 nigel 75 .sp
1468 nigel 63 PCRE_ERROR_NULL (-2)
1469 nigel 75 .sp
1470     Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1471     NULL and \fIovecsize\fP was not zero.
1472     .sp
1473 nigel 63 PCRE_ERROR_BADOPTION (-3)
1474 nigel 75 .sp
1475     An unrecognized bit was set in the \fIoptions\fP argument.
1476     .sp
1477 nigel 63 PCRE_ERROR_BADMAGIC (-4)
1478 nigel 75 .sp
1479 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1480 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1481     compiled in an environment of one endianness is run in an environment with the
1482     other endianness. This is the error that PCRE gives when the magic number is
1483     not present.
1484     .sp
1485 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1486 nigel 75 .sp
1487 nigel 63 While running the pattern match, an unknown item was encountered in the
1488     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1489     of the compiled pattern.
1490 nigel 75 .sp
1491 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1492 nigel 75 .sp
1493     If a pattern contains back references, but the \fIovector\fP that is passed to
1494     \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1495 nigel 63 gets a block of memory at the start of matching to use for this purpose. If the
1496 nigel 75 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1497     automatically freed at the end of matching.
1498     .sp
1499 nigel 63 PCRE_ERROR_NOSUBSTRING (-7)
1500 nigel 75 .sp
1501     This error is used by the \fBpcre_copy_substring()\fP,
1502     \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1503     below). It is never returned by \fBpcre_exec()\fP.
1504     .sp
1505 nigel 63 PCRE_ERROR_MATCHLIMIT (-8)
1506 nigel 75 .sp
1507 nigel 87 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1508     \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1509     above.
1510     .sp
1511 nigel 63 PCRE_ERROR_CALLOUT (-9)
1512 nigel 75 .sp
1513     This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1514 nigel 63 use by callout functions that want to yield a distinctive error code. See the
1515 nigel 75 .\" HREF
1516     \fBpcrecallout\fP
1517     .\"
1518     documentation for details.
1519     .sp
1520 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1521 nigel 75 .sp
1522 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1523 nigel 75 .sp
1524 nigel 73 PCRE_ERROR_BADUTF8_OFFSET (-11)
1525 nigel 75 .sp
1526 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1527 nigel 75 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1528     .sp
1529 nigel 77 PCRE_ERROR_PARTIAL (-12)
1530 nigel 75 .sp
1531     The subject string did not match, but it did match partially. See the
1532     .\" HREF
1533     \fBpcrepartial\fP
1534     .\"
1535     documentation for details of partial matching.
1536     .sp
1537 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1538 nigel 75 .sp
1539     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1540     are not supported for partial matching. See the
1541     .\" HREF
1542     \fBpcrepartial\fP
1543     .\"
1544     documentation for details of partial matching.
1545     .sp
1546 nigel 77 PCRE_ERROR_INTERNAL (-14)
1547 nigel 75 .sp
1548     An unexpected internal error has occurred. This error could be caused by a bug
1549     in PCRE or by overwriting of the compiled pattern.
1550     .sp
1551 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1552 nigel 75 .sp
1553     This error is given if the value of the \fIovecsize\fP argument is negative.
1554 nigel 93 .sp
1555     PCRE_ERROR_RECURSIONLIMIT (-21)
1556     .sp
1557     The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1558     field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1559     description above.
1560     .sp
1561     PCRE_ERROR_BADNEWLINE (-23)
1562     .sp
1563     An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1564     .P
1565 ph10 197 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1566 nigel 75 .
1567     .
1568     .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1569 nigel 63 .rs
1570     .sp
1571 nigel 75 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1572 nigel 63 .ti +5n
1573 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1574 nigel 63 .ti +5n
1575 nigel 75 .B int \fIbuffersize\fP);
1576 nigel 63 .PP
1577 nigel 75 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1578 nigel 63 .ti +5n
1579 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1580 nigel 63 .ti +5n
1581 nigel 75 .B const char **\fIstringptr\fP);
1582 nigel 63 .PP
1583 nigel 75 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1584 nigel 63 .ti +5n
1585 nigel 75 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1586 nigel 63 .PP
1587     Captured substrings can be accessed directly by using the offsets returned by
1588 nigel 75 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1589     \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1590     \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1591 nigel 63 as new, separate, zero-terminated strings. These functions identify substrings
1592     by number. The next section describes functions for extracting named
1593 nigel 91 substrings.
1594 nigel 75 .P
1595 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1596     further zero added on the end, but the result is not, of course, a C string.
1597     However, you can process such a string by referring to the length that is
1598     returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1599     Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1600     for handling strings containing binary zeros, because the end of the final
1601     string is not independently indicated.
1602     .P
1603 nigel 63 The first three arguments are the same for all three of these functions:
1604 nigel 75 \fIsubject\fP is the subject string that has just been successfully matched,
1605     \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1606     \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1607 nigel 63 captured by the match, including the substring that matched the entire regular
1608 nigel 75 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1609     than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1610     space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1611     number of elements in the vector divided by three.
1612     .P
1613     The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1614     extract a single substring, whose number is given as \fIstringnumber\fP. A
1615     value of zero extracts the substring that matched the entire pattern, whereas
1616     higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1617     the string is placed in \fIbuffer\fP, whose length is given by
1618     \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1619     obtained via \fBpcre_malloc\fP, and its address is returned via
1620     \fIstringptr\fP. The yield of the function is the length of the string, not
1621 nigel 93 including the terminating zero, or one of these error codes:
1622 nigel 75 .sp
1623 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1624 nigel 75 .sp
1625     The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1626     memory failed for \fBpcre_get_substring()\fP.
1627     .sp
1628 nigel 63 PCRE_ERROR_NOSUBSTRING (-7)
1629 nigel 75 .sp
1630     There is no substring whose number is \fIstringnumber\fP.
1631     .P
1632     The \fBpcre_get_substring_list()\fP function extracts all available substrings
1633 nigel 63 and builds a list of pointers to them. All this is done in a single block of
1634 nigel 75 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1635     is returned via \fIlistptr\fP, which is also the start of the list of string
1636 nigel 63 pointers. The end of the list is marked by a NULL pointer. The yield of the
1637 nigel 93 function is zero if all went well, or the error code
1638 nigel 75 .sp
1639 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1640 nigel 75 .sp
1641 nigel 63 if the attempt to get the memory block failed.
1642 nigel 75 .P
1643 nigel 63 When any of these functions encounter a substring that is unset, which can
1644 nigel 75 happen when capturing subpattern number \fIn+1\fP matches some part of the
1645     subject, but subpattern \fIn\fP has not been used at all, they return an empty
1646 nigel 63 string. This can be distinguished from a genuine zero-length substring by
1647 nigel 75 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1648 nigel 63 substrings.
1649 nigel 75 .P
1650     The two convenience functions \fBpcre_free_substring()\fP and
1651     \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1652     a previous call of \fBpcre_get_substring()\fP or
1653     \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1654     the function pointed to by \fBpcre_free\fP, which of course could be called
1655 nigel 63 directly from a C program. However, PCRE is used in some situations where it is
1656 nigel 91 linked via a special interface to another programming language that cannot use
1657 nigel 75 \fBpcre_free\fP directly; it is for these cases that the functions are
1658 nigel 63 provided.
1659 nigel 75 .
1660     .
1661     .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1662 nigel 63 .rs
1663     .sp
1664 nigel 75 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1665 nigel 63 .ti +5n
1666 nigel 75 .B const char *\fIname\fP);
1667 nigel 63 .PP
1668 nigel 75 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1669 nigel 63 .ti +5n
1670 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
1671     .ti +5n
1672     .B int \fIstringcount\fP, const char *\fIstringname\fP,
1673     .ti +5n
1674     .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1675 nigel 63 .PP
1676 nigel 75 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1677 nigel 63 .ti +5n
1678 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
1679 nigel 63 .ti +5n
1680 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1681 nigel 63 .ti +5n
1682 nigel 75 .B const char **\fIstringptr\fP);
1683 nigel 63 .PP
1684 nigel 75 To extract a substring by name, you first have to find associated number.
1685     For example, for this pattern
1686     .sp
1687 nigel 93 (a+)b(?<xxx>\ed+)...
1688 nigel 75 .sp
1689 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1690     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1691     calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1692     pattern, and the second is the name. The yield of the function is the
1693 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1694     that name.
1695     .P
1696     Given the number, you can extract the substring directly, or use one of the
1697     functions described in the previous section. For convenience, there are also
1698     two functions that do the whole job.
1699     .P
1700 nigel 91 Most of the arguments of \fBpcre_copy_named_substring()\fP and
1701     \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1702 nigel 75 functions that extract by number. As these are described in the previous
1703     section, they are not re-described here. There are just two differences:
1704     .P
1705 nigel 63 First, instead of a substring number, a substring name is given. Second, there
1706     is an extra argument, given at the start, which is a pointer to the compiled
1707     pattern. This is needed in order to gain access to the name-to-number
1708     translation table.
1709 nigel 75 .P
1710     These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1711 ph10 127 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1712     appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1713 ph10 128 the behaviour may not be what you want (see the next section).
1714 nigel 77 .
1715     .
1716 nigel 91 .SH "DUPLICATE SUBPATTERN NAMES"
1717     .rs
1718     .sp
1719     .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1720     .ti +5n
1721     .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1722     .PP
1723     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1724     are not required to be unique. Normally, patterns with duplicate names are such
1725     that in any one match, only one of the named subpatterns participates. An
1726     example is shown in the
1727     .\" HREF
1728     \fBpcrepattern\fP
1729     .\"
1730 ph10 208 documentation.
1731 ph10 203 .P
1732     When duplicates are present, \fBpcre_copy_named_substring()\fP and
1733     \fBpcre_get_named_substring()\fP return the first substring corresponding to
1734     the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1735     returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1736     returns one of the numbers that are associated with the name, but it is not
1737     defined which it is.
1738     .P
1739 nigel 91 If you want to get full details of all captured substrings for a given name,
1740     you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1741     argument is the compiled pattern, and the second is the name. The third and
1742     fourth are pointers to variables which are updated by the function. After it
1743     has run, they point to the first and last entries in the name-to-number table
1744     for the given name. The function itself returns the length of each entry, or
1745 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1746     described above in the section entitled \fIInformation about a pattern\fP.
1747     Given all the relevant entries for the name, you can extract each of their
1748     numbers, and hence the captured data, if any.
1749 nigel 91 .
1750     .
1751 nigel 77 .SH "FINDING ALL POSSIBLE MATCHES"
1752     .rs
1753     .sp
1754     The traditional matching function uses a similar algorithm to Perl, which stops
1755     when it finds the first match, starting at a given point in the subject. If you
1756     want to find all possible matches, or the longest possible match, consider
1757     using the alternative matching function (see below) instead. If you cannot use
1758     the alternative function, but still need to find all possible matches, you
1759     can kludge it up by making use of the callout facility, which is described in
1760     the
1761     .\" HREF
1762     \fBpcrecallout\fP
1763     .\"
1764     documentation.
1765 nigel 75 .P
1766 nigel 77 What you have to do is to insert a callout right at the end of the pattern.
1767     When your callout function is called, extract and save the current matched
1768     substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1769     other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1770     will yield PCRE_ERROR_NOMATCH.
1771     .
1772     .
1773     .\" HTML <a name="dfamatch"></a>
1774     .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1775     .rs
1776     .sp
1777     .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1778     .ti +5n
1779     .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1780     .ti +5n
1781     .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1782     .ti +5n
1783     .B int *\fIworkspace\fP, int \fIwscount\fP);
1784     .P
1785     The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1786 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1787     just once, and does not backtrack. This has different characteristics to the
1788     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1789     patterns are not supported. Nevertheless, there are times when this kind of
1790     matching can be useful. For a discussion of the two matching algorithms, see
1791     the
1792 nigel 77 .\" HREF
1793     \fBpcrematching\fP
1794     .\"
1795     documentation.
1796     .P
1797     The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1798     \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1799     different way, and this is described below. The other common arguments are used
1800     in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1801     here.
1802     .P
1803     The two additional arguments provide workspace for the function. The workspace
1804     vector should contain at least 20 elements. It is used for keeping track of
1805     multiple paths through the pattern tree. More workspace will be needed for
1806 nigel 91 patterns and subjects where there are a lot of potential matches.
1807 nigel 77 .P
1808 nigel 87 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1809 nigel 77 .sp
1810     int rc;
1811     int ovector[10];
1812     int wspace[20];
1813 nigel 87 rc = pcre_dfa_exec(
1814 nigel 77 re, /* result of pcre_compile() */
1815     NULL, /* we didn't study the pattern */
1816     "some string", /* the subject string */
1817     11, /* the length of the subject string */
1818     0, /* start at offset 0 in the subject */
1819     0, /* default options */
1820     ovector, /* vector of integers for substring information */
1821     10, /* number of elements (NOT size in bytes) */
1822     wspace, /* working space vector */
1823     20); /* number of elements (NOT size in bytes) */
1824     .
1825     .SS "Option bits for \fBpcre_dfa_exec()\fP"
1826     .rs
1827     .sp
1828     The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1829 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1830     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1831 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1832     the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1833     .sp
1834     PCRE_PARTIAL
1835     .sp
1836     This has the same general effect as it does for \fBpcre_exec()\fP, but the
1837     details are slightly different. When PCRE_PARTIAL is set for
1838     \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1839     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1840     complete matches, but there is still at least one matching possibility. The
1841     portion of the string that provided the partial match is set as the first
1842     matching string.
1843     .sp
1844     PCRE_DFA_SHORTEST
1845     .sp
1846     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1847 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1848     works, this is necessarily the shortest possible match at the first possible
1849     matching point in the subject string.
1850 nigel 77 .sp
1851     PCRE_DFA_RESTART
1852     .sp
1853     When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1854     a partial match, it is possible to call it again, with additional subject
1855     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1856     option requests this action; when it is set, the \fIworkspace\fP and
1857     \fIwscount\fP options must reference the same vector as before because data
1858     about the match so far is left in them after a partial match. There is more
1859     discussion of this facility in the
1860     .\" HREF
1861     \fBpcrepartial\fP
1862     .\"
1863     documentation.
1864     .
1865     .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1866     .rs
1867     .sp
1868     When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1869     substring in the subject. Note, however, that all the matches from one run of
1870     the function start at the same point in the subject. The shorter matches are
1871     all initial substrings of the longer matches. For example, if the pattern
1872     .sp
1873     <.*>
1874     .sp
1875     is matched against the string
1876     .sp
1877     This is <something> <something else> <something further> no more
1878     .sp
1879     the three matched strings are
1880     .sp
1881     <something>
1882     <something> <something else>
1883     <something> <something else> <something further>
1884     .sp
1885     On success, the yield of the function is a number greater than zero, which is
1886     the number of matched substrings. The substrings themselves are returned in
1887     \fIovector\fP. Each string uses two elements; the first is the offset to the
1888 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1889     the same start offset. (Space could have been saved by giving this only once,
1890     but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1891     returns data, even though the meaning of the strings is different.)
1892 nigel 77 .P
1893     The strings are returned in reverse order of length; that is, the longest
1894     matching string is given first. If there were too many matches to fit into
1895     \fIovector\fP, the yield of the function is zero, and the vector is filled with
1896     the longest matches.
1897     .
1898     .SS "Error returns from \fBpcre_dfa_exec()\fP"
1899     .rs
1900     .sp
1901     The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1902     Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1903     described
1904     .\" HTML <a href="#errorlist">
1905     .\" </a>
1906     above.
1907     .\"
1908     There are in addition the following errors that are specific to
1909     \fBpcre_dfa_exec()\fP:
1910     .sp
1911     PCRE_ERROR_DFA_UITEM (-16)
1912     .sp
1913     This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1914     that it does not support, for instance, the use of \eC or a back reference.
1915     .sp
1916     PCRE_ERROR_DFA_UCOND (-17)
1917     .sp
1918 nigel 93 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1919     uses a back reference for the condition, or a test for recursion in a specific
1920     group. These are not supported.
1921 nigel 77 .sp
1922     PCRE_ERROR_DFA_UMLIMIT (-18)
1923     .sp
1924     This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1925     block that contains a setting of the \fImatch_limit\fP field. This is not
1926     supported (it is meaningless).
1927     .sp
1928     PCRE_ERROR_DFA_WSSIZE (-19)
1929     .sp
1930     This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1931     \fIworkspace\fP vector.
1932     .sp
1933     PCRE_ERROR_DFA_RECURSE (-20)
1934     .sp
1935     When a recursive subpattern is processed, the matching function calls itself
1936     recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1937     error is given if the output vector is not large enough. This should be
1938     extremely rare, as a vector of size 1000 is used.
1939 nigel 93 .
1940     .
1941     .SH "SEE ALSO"
1942     .rs
1943     .sp
1944     \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1945     \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1946     \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1947 ph10 99 .
1948     .
1949     .SH AUTHOR
1950     .rs
1951     .sp
1952     .nf
1953     Philip Hazel
1954     University Computing Service
1955     Cambridge CB2 3QH, England.
1956     .fi
1957     .
1958     .
1959     .SH REVISION
1960     .rs
1961     .sp
1962     .nf
1963 ph10 312 Last updated: 23 January 2008
1964     Copyright (c) 1997-2008 University of Cambridge.
1965 ph10 99 .fi

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12