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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 376 - (hide annotations) (download)
Sun Mar 1 12:00:59 2009 UTC (5 years, 5 months ago) by ph10
File size: 85061 byte(s)
Bug fixes: forgot to commit these individually.

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 ph10 376 The output is a long 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 ph10 376 The output is a long integer that gives the default limit for the depth of
357 nigel 87 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 ph10 336 PCRE_JAVASCRIPT_COMPAT
553     .sp
554 ph10 345 If this option is set, PCRE's behaviour is changed in some ways so that it is
555 ph10 336 compatible with JavaScript rather than Perl. The changes are as follows:
556     .P
557     (1) A lone closing square bracket in a pattern causes a compile-time error,
558     because this is illegal in JavaScript (by default it is treated as a data
559     character). Thus, the pattern AB]CD becomes illegal when this option is set.
560     .P
561     (2) At run time, a back reference to an unset subpattern group matches an empty
562 ph10 345 string (by default this causes the current matching alternative to fail). A
563     pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
564     an "a" in the subject), whereas it fails by default, for Perl compatibility.
565 ph10 336 .sp
566 nigel 63 PCRE_MULTILINE
567 nigel 75 .sp
568     By default, PCRE treats the subject string as consisting of a single line of
569     characters (even if it actually contains newlines). The "start of line"
570 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
571     line" metacharacter ($) matches only at the end of the string, or before a
572     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
573     Perl.
574 nigel 75 .P
575 nigel 63 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
576 nigel 91 match immediately following or immediately before internal newlines in the
577     subject string, respectively, as well as at the very start and end. This is
578     equivalent to Perl's /m option, and it can be changed within a pattern by a
579     (?m) option setting. If there are no newlines in a subject string, or no
580 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
581 nigel 75 .sp
582 nigel 91 PCRE_NEWLINE_CR
583     PCRE_NEWLINE_LF
584     PCRE_NEWLINE_CRLF
585 ph10 150 PCRE_NEWLINE_ANYCRLF
586 nigel 93 PCRE_NEWLINE_ANY
587 nigel 91 .sp
588     These options override the default newline definition that was chosen when PCRE
589     was built. Setting the first or the second specifies that a newline is
590 nigel 93 indicated by a single character (CR or LF, respectively). Setting
591     PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
592 ph10 149 CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
593     preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
594     that any Unicode newline sequence should be recognized. The Unicode newline
595     sequences are the three just mentioned, plus the single characters VT (vertical
596     tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
597     separator, U+2028), and PS (paragraph separator, U+2029). The last two are
598     recognized only in UTF-8 mode.
599 nigel 91 .P
600 nigel 93 The newline setting in the options word uses three bits that are treated
601 ph10 149 as a number, giving eight possibilities. Currently only six are used (default
602     plus the five values above). This means that if you set more than one newline
603 nigel 93 option, the combination may or may not be sensible. For example,
604     PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
605 ph10 149 other combinations may yield unused numbers and cause an error.
606 nigel 93 .P
607     The only time that a line break is specially recognized when compiling a
608     pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
609     class is encountered. This indicates a comment that lasts until after the next
610     line break sequence. In other circumstances, line break sequences are treated
611     as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
612     as whitespace characters and are therefore ignored.
613     .P
614     The newline option that is set at compile time becomes the default that is used
615     for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
616 nigel 91 .sp
617 nigel 63 PCRE_NO_AUTO_CAPTURE
618 nigel 75 .sp
619 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
620     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
621     were followed by ?: but named parentheses can still be used for capturing (and
622     they acquire numbers in the usual way). There is no equivalent of this option
623     in Perl.
624 nigel 75 .sp
625 nigel 63 PCRE_UNGREEDY
626 nigel 75 .sp
627 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
628     greedy by default, but become greedy if followed by "?". It is not compatible
629     with Perl. It can also be set by a (?U) option setting within the pattern.
630 nigel 75 .sp
631 nigel 63 PCRE_UTF8
632 nigel 75 .sp
633 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
634     of UTF-8 characters instead of single-byte character strings. However, it is
635 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
636 nigel 63 of this option provokes an error. Details of how this option changes the
637     behaviour of PCRE are given in the
638     .\" HTML <a href="pcre.html#utf8support">
639     .\" </a>
640     section on UTF-8 support
641     .\"
642     in the main
643     .\" HREF
644 nigel 75 \fBpcre\fP
645 nigel 63 .\"
646     page.
647 nigel 75 .sp
648 nigel 71 PCRE_NO_UTF8_CHECK
649 nigel 75 .sp
650 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
651 ph10 211 automatically checked. There is a discussion about the
652     .\" HTML <a href="pcre.html#utf8strings">
653     .\" </a>
654     validity of UTF-8 strings
655     .\"
656     in the main
657     .\" HREF
658     \fBpcre\fP
659     .\"
660     page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
661     returns an error. If you already know that your pattern is valid, and you want
662     to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
663     option. When it is set, the effect of passing an invalid UTF-8 string as a
664     pattern is undefined. It may cause your program to crash. Note that this option
665     can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
666     the UTF-8 validity checking of subject strings.
667 nigel 75 .
668     .
669 nigel 77 .SH "COMPILATION ERROR CODES"
670     .rs
671     .sp
672     The following table lists the error codes than may be returned by
673     \fBpcre_compile2()\fP, along with the error messages that may be returned by
674 nigel 93 both compiling functions. As PCRE has developed, some error codes have fallen
675     out of use. To avoid confusion, they have not been re-used.
676 nigel 77 .sp
677     0 no error
678     1 \e at end of pattern
679     2 \ec at end of pattern
680     3 unrecognized character follows \e
681     4 numbers out of order in {} quantifier
682     5 number too big in {} quantifier
683     6 missing terminating ] for character class
684     7 invalid escape sequence in character class
685     8 range out of order in character class
686     9 nothing to repeat
687 nigel 93 10 [this code is not in use]
688 nigel 77 11 internal error: unexpected repeat
689 ph10 290 12 unrecognized character after (? or (?-
690 nigel 77 13 POSIX named classes are supported only within a class
691     14 missing )
692     15 reference to non-existent subpattern
693     16 erroffset passed as NULL
694     17 unknown option bit(s) set
695     18 missing ) after comment
696 nigel 93 19 [this code is not in use]
697 ph10 290 20 regular expression is too large
698 nigel 77 21 failed to get memory
699     22 unmatched parentheses
700     23 internal error: code overflow
701     24 unrecognized character after (?<
702     25 lookbehind assertion is not fixed length
703 nigel 91 26 malformed number or name after (?(
704 nigel 77 27 conditional group contains more than two branches
705     28 assertion expected after (?(
706 ph10 181 29 (?R or (?[+-]digits must be followed by )
707 nigel 77 30 unknown POSIX class name
708     31 POSIX collating elements are not supported
709     32 this version of PCRE is not compiled with PCRE_UTF8 support
710 nigel 93 33 [this code is not in use]
711 nigel 77 34 character value in \ex{...} sequence is too large
712     35 invalid condition (?(0)
713     36 \eC not allowed in lookbehind assertion
714     37 PCRE does not support \eL, \el, \eN, \eU, or \eu
715     38 number after (?C is > 255
716     39 closing ) for (?C expected
717     40 recursive call could loop indefinitely
718     41 unrecognized character after (?P
719 nigel 93 42 syntax error in subpattern name (missing terminator)
720 nigel 91 43 two named subpatterns have the same name
721 nigel 77 44 invalid UTF-8 string
722     45 support for \eP, \ep, and \eX has not been compiled
723     46 malformed \eP or \ep sequence
724     47 unknown property name after \eP or \ep
725 nigel 91 48 subpattern name is too long (maximum 32 characters)
726 ph10 290 49 too many named subpatterns (maximum 10000)
727 ph10 202 50 [this code is not in use]
728 nigel 91 51 octal value is greater than \e377 (not in UTF-8 mode)
729 nigel 93 52 internal error: overran compiling workspace
730     53 internal error: previously-checked referenced subpattern not found
731     54 DEFINE group contains more than one branch
732     55 repeating a DEFINE group is not allowed
733 ph10 231 56 inconsistent NEWLINE options
734 ph10 345 57 \eg is not followed by a braced, angle-bracketed, or quoted
735     name/number or by a plain number
736 ph10 336 58 a numbered reference must not be zero
737 ph10 290 59 (*VERB) with an argument is not supported
738     60 (*VERB) not recognized
739     61 number is too big
740     62 subpattern name expected
741 ph10 292 63 digit expected after (?+
742 ph10 336 64 ] is an invalid data character in JavaScript compatibility mode
743 ph10 290 .sp
744 ph10 292 The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
745 ph10 290 be used if the limits were changed when PCRE was built.
746 nigel 77 .
747     .
748 nigel 75 .SH "STUDYING A PATTERN"
749 nigel 63 .rs
750     .sp
751 nigel 77 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
752 nigel 63 .ti +5n
753 nigel 75 .B const char **\fIerrptr\fP);
754 nigel 63 .PP
755 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
756     more time analyzing it in order to speed up the time taken for matching. The
757     function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
758     argument. If studying the pattern produces additional information that will
759     help speed up matching, \fBpcre_study()\fP returns a pointer to a
760     \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
761     results of the study.
762     .P
763     The returned value from \fBpcre_study()\fP can be passed directly to
764     \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
765 nigel 63 fields that can be set by the caller before the block is passed; these are
766 nigel 75 described
767     .\" HTML <a href="#extradata">
768     .\" </a>
769     below
770     .\"
771     in the section on matching a pattern.
772     .P
773 nigel 77 If studying the pattern does not produce any additional information
774 nigel 75 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
775     wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
776     own \fBpcre_extra\fP block.
777     .P
778     The second argument of \fBpcre_study()\fP contains option bits. At present, no
779     options are defined, and this argument should always be zero.
780     .P
781     The third argument for \fBpcre_study()\fP is a pointer for an error message. If
782 nigel 63 studying succeeds (even if no data is returned), the variable it points to is
783 nigel 87 set to NULL. Otherwise it is set to point to a textual error message. This is a
784     static string that is part of the library. You must not try to free it. You
785     should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
786     sure that it has run successfully.
787 nigel 75 .P
788     This is a typical call to \fBpcre_study\fP():
789     .sp
790 nigel 63 pcre_extra *pe;
791     pe = pcre_study(
792     re, /* result of pcre_compile() */
793     0, /* no options exist */
794     &error); /* set to NULL or points to a message */
795 nigel 75 .sp
796 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
797     not have a single fixed starting character. A bitmap of possible starting
798 nigel 75 bytes is created.
799     .
800     .
801 nigel 63 .\" HTML <a name="localesupport"></a>
802 nigel 75 .SH "LOCALE SUPPORT"
803 nigel 63 .rs
804     .sp
805 ph10 139 PCRE handles caseless matching, and determines whether characters are letters,
806 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
807 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
808 nigel 75 less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
809     can be tested with \ep if PCRE is built with Unicode character property
810 ph10 142 support. The use of locales with Unicode is discouraged. If you are handling
811     characters with codes greater than 128, you should either use UTF-8 and
812 ph10 139 Unicode, or use locales, but not try to mix the two.
813 nigel 75 .P
814 ph10 139 PCRE contains an internal set of tables that are used when the final argument
815     of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
816 ph10 142 Normally, the internal tables recognize only ASCII characters. However, when
817 ph10 139 PCRE is built, it is possible to cause the internal tables to be rebuilt in the
818     default "C" locale of the local system, which may cause them to be different.
819 nigel 75 .P
820 ph10 139 The internal tables can always be overridden by tables supplied by the
821     application that calls PCRE. These may be created in a different locale from
822     the default. As more and more applications change to using Unicode, the need
823     for this locale support is expected to die away.
824     .P
825 nigel 75 External tables are built by calling the \fBpcre_maketables()\fP function,
826     which has no arguments, in the relevant locale. The result can then be passed
827     to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
828     example, to build and use tables that are appropriate for the French locale
829     (where accented characters with values greater than 128 are treated as letters),
830     the following code could be used:
831     .sp
832     setlocale(LC_CTYPE, "fr_FR");
833 nigel 63 tables = pcre_maketables();
834     re = pcre_compile(..., tables);
835 nigel 75 .sp
836 ph10 142 The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
837 ph10 139 are using Windows, the name for the French locale is "french".
838     .P
839 nigel 75 When \fBpcre_maketables()\fP runs, the tables are built in memory that is
840     obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
841     that the memory containing the tables remains available for as long as it is
842     needed.
843     .P
844     The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
845     pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
846     and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
847     pattern, compilation, studying and matching all happen in the same locale, but
848     different patterns can be compiled in different locales.
849     .P
850     It is possible to pass a table pointer or NULL (indicating the use of the
851     internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
852     this facility could be used to match a pattern in a different locale from the
853     one in which it was compiled. Passing table pointers at run time is discussed
854     below in the section on matching a pattern.
855     .
856     .
857     .SH "INFORMATION ABOUT A PATTERN"
858 nigel 63 .rs
859     .sp
860 nigel 75 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
861 nigel 63 .ti +5n
862 nigel 75 .B int \fIwhat\fP, void *\fIwhere\fP);
863 nigel 63 .PP
864 nigel 75 The \fBpcre_fullinfo()\fP function returns information about a compiled
865     pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
866 nigel 63 nevertheless retained for backwards compability (and is documented below).
867 nigel 75 .P
868     The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
869     pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
870 nigel 63 the pattern was not studied. The third argument specifies which piece of
871     information is required, and the fourth argument is a pointer to a variable
872     to receive the data. The yield of the function is zero for success, or one of
873     the following negative numbers:
874 nigel 75 .sp
875     PCRE_ERROR_NULL the argument \fIcode\fP was NULL
876     the argument \fIwhere\fP was NULL
877 nigel 63 PCRE_ERROR_BADMAGIC the "magic number" was not found
878 nigel 75 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
879     .sp
880     The "magic number" is placed at the start of each compiled pattern as an simple
881     check against passing an arbitrary memory pointer. Here is a typical call of
882     \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
883     .sp
884 nigel 63 int rc;
885 nigel 91 size_t length;
886 nigel 63 rc = pcre_fullinfo(
887     re, /* result of pcre_compile() */
888     pe, /* result of pcre_study(), or NULL */
889     PCRE_INFO_SIZE, /* what is required */
890     &length); /* where to put the data */
891 nigel 75 .sp
892     The possible values for the third argument are defined in \fBpcre.h\fP, and are
893 nigel 63 as follows:
894 nigel 75 .sp
895 nigel 63 PCRE_INFO_BACKREFMAX
896 nigel 75 .sp
897 nigel 63 Return the number of the highest back reference in the pattern. The fourth
898 nigel 75 argument should point to an \fBint\fP variable. Zero is returned if there are
899 nigel 63 no back references.
900 nigel 75 .sp
901 nigel 63 PCRE_INFO_CAPTURECOUNT
902 nigel 75 .sp
903 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
904 nigel 75 should point to an \fBint\fP variable.
905     .sp
906 nigel 77 PCRE_INFO_DEFAULT_TABLES
907 nigel 75 .sp
908     Return a pointer to the internal default character tables within PCRE. The
909     fourth argument should point to an \fBunsigned char *\fP variable. This
910     information call is provided for internal use by the \fBpcre_study()\fP
911     function. External callers can cause PCRE to use its internal tables by passing
912     a NULL table pointer.
913     .sp
914 nigel 63 PCRE_INFO_FIRSTBYTE
915 nigel 75 .sp
916 nigel 63 Return information about the first byte of any matched string, for a
917 nigel 91 non-anchored pattern. The fourth argument should point to an \fBint\fP
918     variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
919     still recognized for backwards compatibility.)
920 nigel 75 .P
921     If there is a fixed first byte, for example, from a pattern such as
922 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
923 nigel 75 .sp
924 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
925     starts with "^", or
926 nigel 75 .sp
927 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
928     (if it were set, the pattern would be anchored),
929 nigel 75 .sp
930 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
931     subject string or after any newline within the string. Otherwise -2 is
932     returned. For anchored patterns, -2 is returned.
933 nigel 75 .sp
934 nigel 63 PCRE_INFO_FIRSTTABLE
935 nigel 75 .sp
936 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
937     table indicating a fixed set of bytes for the first byte in any matching
938     string, a pointer to the table is returned. Otherwise NULL is returned. The
939 nigel 75 fourth argument should point to an \fBunsigned char *\fP variable.
940     .sp
941 ph10 226 PCRE_INFO_HASCRORLF
942     .sp
943 ph10 227 Return 1 if the pattern contains any explicit matches for CR or LF characters,
944 ph10 243 otherwise 0. The fourth argument should point to an \fBint\fP variable. An
945 ph10 231 explicit match is either a literal CR or LF character, or \er or \en.
946 ph10 226 .sp
947 ph10 169 PCRE_INFO_JCHANGED
948     .sp
949 ph10 278 Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
950     0. The fourth argument should point to an \fBint\fP variable. (?J) and
951     (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
952 ph10 169 .sp
953 nigel 63 PCRE_INFO_LASTLITERAL
954 nigel 75 .sp
955 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
956     string, other than at its start, if such a byte has been recorded. The fourth
957 nigel 75 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
958 nigel 65 returned. For anchored patterns, a last literal byte is recorded only if it
959     follows something of variable length. For example, for the pattern
960 nigel 75 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
961 nigel 65 is -1.
962 nigel 75 .sp
963 nigel 63 PCRE_INFO_NAMECOUNT
964     PCRE_INFO_NAMEENTRYSIZE
965     PCRE_INFO_NAMETABLE
966 nigel 75 .sp
967 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
968     names are just an additional way of identifying the parentheses, which still
969 nigel 91 acquire numbers. Several convenience functions such as
970     \fBpcre_get_named_substring()\fP are provided for extracting captured
971     substrings by name. It is also possible to extract the data directly, by first
972     converting the name to a number in order to access the correct pointers in the
973     output vector (described with \fBpcre_exec()\fP below). To do the conversion,
974     you need to use the name-to-number map, which is described by these three
975     values.
976 nigel 75 .P
977 nigel 63 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
978     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
979 nigel 75 entry; both of these return an \fBint\fP value. The entry size depends on the
980 nigel 63 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
981 nigel 75 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
982 nigel 63 are the number of the capturing parenthesis, most significant byte first. The
983     rest of the entry is the corresponding name, zero terminated. The names are in
984 nigel 91 alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
985     their parentheses numbers. For example, consider the following pattern (assume
986 nigel 63 PCRE_EXTENDED is set, so white space - including newlines - is ignored):
987 nigel 75 .sp
988     .\" JOIN
989 nigel 93 (?<date> (?<year>(\ed\ed)?\ed\ed) -
990     (?<month>\ed\ed) - (?<day>\ed\ed) )
991 nigel 75 .sp
992 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
993     in the table is eight bytes long. The table is as follows, with non-printing
994 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
995     .sp
996 nigel 63 00 01 d a t e 00 ??
997     00 05 d a y 00 ?? ??
998     00 04 m o n t h 00
999     00 02 y e a r 00 ??
1000 nigel 75 .sp
1001     When writing code to extract data from named subpatterns using the
1002 nigel 91 name-to-number map, remember that the length of the entries is likely to be
1003 nigel 75 different for each compiled pattern.
1004     .sp
1005 ph10 169 PCRE_INFO_OKPARTIAL
1006     .sp
1007 ph10 172 Return 1 if the pattern can be used for partial matching, otherwise 0. The
1008 ph10 169 fourth argument should point to an \fBint\fP variable. The
1009     .\" HREF
1010     \fBpcrepartial\fP
1011     .\"
1012     documentation lists the restrictions that apply to patterns when partial
1013     matching is used.
1014     .sp
1015 nigel 63 PCRE_INFO_OPTIONS
1016 nigel 75 .sp
1017 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
1018 nigel 75 argument should point to an \fBunsigned long int\fP variable. These option bits
1019     are those specified in the call to \fBpcre_compile()\fP, modified by any
1020 ph10 196 top-level option settings at the start of the pattern itself. In other words,
1021     they are the options that will be in force when matching starts. For example,
1022     if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1023     result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1024 nigel 75 .P
1025 nigel 63 A pattern is automatically anchored by PCRE if all of its top-level
1026     alternatives begin with one of the following:
1027 nigel 75 .sp
1028 nigel 63 ^ unless PCRE_MULTILINE is set
1029 nigel 75 \eA always
1030     \eG always
1031     .\" JOIN
1032 nigel 63 .* if PCRE_DOTALL is set and there are no back
1033     references to the subpattern in which .* appears
1034 nigel 75 .sp
1035 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1036 nigel 75 \fBpcre_fullinfo()\fP.
1037     .sp
1038 nigel 63 PCRE_INFO_SIZE
1039 nigel 75 .sp
1040 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
1041 nigel 75 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1042     place the compiled data. The fourth argument should point to a \fBsize_t\fP
1043 nigel 63 variable.
1044 nigel 75 .sp
1045 nigel 63 PCRE_INFO_STUDYSIZE
1046 nigel 75 .sp
1047     Return the size of the data block pointed to by the \fIstudy_data\fP field in
1048     a \fBpcre_extra\fP block. That is, it is the value that was passed to
1049     \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1050     created by \fBpcre_study()\fP. The fourth argument should point to a
1051     \fBsize_t\fP variable.
1052     .
1053     .
1054     .SH "OBSOLETE INFO FUNCTION"
1055 nigel 63 .rs
1056     .sp
1057 nigel 75 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1058     .B *\fIfirstcharptr\fP);
1059 nigel 63 .PP
1060 nigel 75 The \fBpcre_info()\fP function is now obsolete because its interface is too
1061 nigel 63 restrictive to return all the available data about a compiled pattern. New
1062 nigel 75 programs should use \fBpcre_fullinfo()\fP instead. The yield of
1063     \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1064 nigel 63 following negative numbers:
1065 nigel 75 .sp
1066     PCRE_ERROR_NULL the argument \fIcode\fP was NULL
1067 nigel 63 PCRE_ERROR_BADMAGIC the "magic number" was not found
1068 nigel 75 .sp
1069     If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1070 nigel 63 pattern was compiled is placed in the integer it points to (see
1071     PCRE_INFO_OPTIONS above).
1072 nigel 75 .P
1073     If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1074 nigel 63 it is used to pass back information about the first character of any matched
1075     string (see PCRE_INFO_FIRSTBYTE above).
1076 nigel 75 .
1077     .
1078 nigel 77 .SH "REFERENCE COUNTS"
1079 nigel 63 .rs
1080     .sp
1081 nigel 77 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1082     .PP
1083     The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1084     data block that contains a compiled pattern. It is provided for the benefit of
1085     applications that operate in an object-oriented manner, where different parts
1086     of the application may be using the same compiled pattern, but you want to free
1087     the block when they are all done.
1088     .P
1089     When a pattern is compiled, the reference count field is initialized to zero.
1090     It is changed only by calling this function, whose action is to add the
1091     \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1092     function is the new value. However, the value of the count is constrained to
1093     lie between 0 and 65535, inclusive. If the new value is outside these limits,
1094     it is forced to the appropriate limit value.
1095     .P
1096     Except when it is zero, the reference count is not correctly preserved if a
1097     pattern is compiled on one host and then transferred to a host whose byte-order
1098     is different. (This seems a highly unlikely scenario.)
1099     .
1100     .
1101     .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1102     .rs
1103     .sp
1104 nigel 75 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1105 nigel 63 .ti +5n
1106 nigel 75 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1107 nigel 63 .ti +5n
1108 nigel 75 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1109     .P
1110     The function \fBpcre_exec()\fP is called to match a subject string against a
1111     compiled pattern, which is passed in the \fIcode\fP argument. If the
1112 nigel 63 pattern has been studied, the result of the study should be passed in the
1113 nigel 77 \fIextra\fP argument. This function is the main matching facility of the
1114     library, and it operates in a Perl-like manner. For specialist use there is
1115     also an alternative matching function, which is described
1116     .\" HTML <a href="#dfamatch">
1117     .\" </a>
1118     below
1119     .\"
1120     in the section about the \fBpcre_dfa_exec()\fP function.
1121 nigel 75 .P
1122     In most applications, the pattern will have been compiled (and optionally
1123     studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1124     possible to save compiled patterns and study data, and then use them later
1125     in different processes, possibly even on different hosts. For a discussion
1126     about this, see the
1127     .\" HREF
1128     \fBpcreprecompile\fP
1129     .\"
1130     documentation.
1131     .P
1132     Here is an example of a simple call to \fBpcre_exec()\fP:
1133     .sp
1134 nigel 63 int rc;
1135     int ovector[30];
1136     rc = pcre_exec(
1137     re, /* result of pcre_compile() */
1138     NULL, /* we didn't study the pattern */
1139     "some string", /* the subject string */
1140     11, /* the length of the subject string */
1141     0, /* start at offset 0 in the subject */
1142     0, /* default options */
1143 nigel 75 ovector, /* vector of integers for substring information */
1144 nigel 77 30); /* number of elements (NOT size in bytes) */
1145 nigel 75 .
1146     .\" HTML <a name="extradata"></a>
1147     .SS "Extra data for \fBpcre_exec()\fR"
1148     .rs
1149     .sp
1150     If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1151     data block. The \fBpcre_study()\fP function returns such a block (when it
1152 nigel 63 doesn't return NULL), but you can also create one for yourself, and pass
1153 nigel 87 additional information in it. The \fBpcre_extra\fP block contains the following
1154     fields (not necessarily in this order):
1155 nigel 75 .sp
1156     unsigned long int \fIflags\fP;
1157     void *\fIstudy_data\fP;
1158     unsigned long int \fImatch_limit\fP;
1159 nigel 87 unsigned long int \fImatch_limit_recursion\fP;
1160 nigel 75 void *\fIcallout_data\fP;
1161     const unsigned char *\fItables\fP;
1162     .sp
1163     The \fIflags\fP field is a bitmap that specifies which of the other fields
1164 nigel 63 are set. The flag bits are:
1165 nigel 75 .sp
1166 nigel 63 PCRE_EXTRA_STUDY_DATA
1167     PCRE_EXTRA_MATCH_LIMIT
1168 nigel 87 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1169 nigel 63 PCRE_EXTRA_CALLOUT_DATA
1170 nigel 75 PCRE_EXTRA_TABLES
1171     .sp
1172     Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1173     \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1174     the appropriate flag bit. You should not set this yourself, but you may add to
1175     the block by setting the other fields and their corresponding flag bits.
1176     .P
1177     The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1178 nigel 63 vast amount of resources when running patterns that are not going to match,
1179     but which have a very large number of possibilities in their search trees. The
1180 nigel 75 classic example is the use of nested unlimited repeats.
1181     .P
1182     Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1183 nigel 87 (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1184     number of times this function is called during a match, which has the effect of
1185     limiting the amount of backtracking that can take place. For patterns that are
1186     not anchored, the count restarts from zero for each position in the subject
1187     string.
1188 nigel 75 .P
1189 nigel 87 The default value for the limit can be set when PCRE is built; the default
1190 nigel 63 default is 10 million, which handles all but the most extreme cases. You can
1191 nigel 87 override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1192     block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1193     the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1194     PCRE_ERROR_MATCHLIMIT.
1195 nigel 75 .P
1196 nigel 87 The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1197     instead of limiting the total number of times that \fBmatch()\fP is called, it
1198     limits the depth of recursion. The recursion depth is a smaller number than the
1199     total number of calls, because not all calls to \fBmatch()\fP are recursive.
1200     This limit is of use only if it is set smaller than \fImatch_limit\fP.
1201     .P
1202     Limiting the recursion depth limits the amount of stack that can be used, or,
1203     when PCRE has been compiled to use memory on the heap instead of the stack, the
1204     amount of heap memory that can be used.
1205     .P
1206     The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1207     built; the default default is the same value as the default for
1208     \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1209     with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1210     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1211     is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1212     .P
1213 nigel 75 The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1214     which is described in the
1215     .\" HREF
1216     \fBpcrecallout\fP
1217     .\"
1218     documentation.
1219     .P
1220     The \fItables\fP field is used to pass a character tables pointer to
1221     \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1222     pattern. A non-NULL value is stored with the compiled pattern only if custom
1223     tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1224     If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1225     internal tables to be used. This facility is helpful when re-using patterns
1226     that have been saved after compiling with an external set of tables, because
1227     the external tables might be at a different address when \fBpcre_exec()\fP is
1228     called. See the
1229     .\" HREF
1230     \fBpcreprecompile\fP
1231     .\"
1232     documentation for a discussion of saving compiled patterns for later use.
1233     .
1234 ph10 226 .\" HTML <a name="execoptions"></a>
1235 nigel 75 .SS "Option bits for \fBpcre_exec()\fP"
1236     .rs
1237     .sp
1238     The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1239 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1240     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1241 nigel 75 .sp
1242     PCRE_ANCHORED
1243     .sp
1244     The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1245     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1246     to be anchored by virtue of its contents, it cannot be made unachored at
1247     matching time.
1248     .sp
1249 ph10 231 PCRE_BSR_ANYCRLF
1250     PCRE_BSR_UNICODE
1251     .sp
1252     These options (which are mutually exclusive) control what the \eR escape
1253     sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1254     match any Unicode newline sequence. These options override the choice that was
1255     made or defaulted when the pattern was compiled.
1256     .sp
1257 nigel 91 PCRE_NEWLINE_CR
1258     PCRE_NEWLINE_LF
1259     PCRE_NEWLINE_CRLF
1260 ph10 150 PCRE_NEWLINE_ANYCRLF
1261 nigel 93 PCRE_NEWLINE_ANY
1262 nigel 91 .sp
1263     These options override the newline definition that was chosen or defaulted when
1264 nigel 93 the pattern was compiled. For details, see the description of
1265     \fBpcre_compile()\fP above. During matching, the newline choice affects the
1266     behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1267     the way the match position is advanced after a match failure for an unanchored
1268 ph10 227 pattern.
1269 ph10 225 .P
1270 ph10 226 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1271     match attempt for an unanchored pattern fails when the current position is at a
1272 ph10 230 CRLF sequence, and the pattern contains no explicit matches for CR or LF
1273 ph10 226 characters, the match position is advanced by two characters instead of one, in
1274     other words, to after the CRLF.
1275     .P
1276 ph10 227 The above rule is a compromise that makes the most common cases work as
1277     expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1278     set), it does not match the string "\er\enA" because, after failing at the
1279     start, it skips both the CR and the LF before retrying. However, the pattern
1280     [\er\en]A does match that string, because it contains an explicit CR or LF
1281 ph10 226 reference, and so advances only by one character after the first failure.
1282     .P
1283 ph10 231 An explicit match for CR of LF is either a literal appearance of one of those
1284     characters, or one of the \er or \en escape sequences. Implicit matches such as
1285 ph10 230 [^X] do not count, nor does \es (which includes CR and LF in the characters
1286     that it matches).
1287     .P
1288 ph10 226 Notwithstanding the above, anomalous effects may still occur when CRLF is a
1289     valid newline sequence and explicit \er or \en escapes appear in the pattern.
1290 nigel 91 .sp
1291 nigel 63 PCRE_NOTBOL
1292 nigel 75 .sp
1293     This option specifies that first character of the subject string is not the
1294     beginning of a line, so the circumflex metacharacter should not match before
1295     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1296     never to match. This option affects only the behaviour of the circumflex
1297     metacharacter. It does not affect \eA.
1298     .sp
1299 nigel 63 PCRE_NOTEOL
1300 nigel 75 .sp
1301     This option specifies that the end of the subject string is not the end of a
1302     line, so the dollar metacharacter should not match it nor (except in multiline
1303     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1304     compile time) causes dollar never to match. This option affects only the
1305     behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1306     .sp
1307 nigel 63 PCRE_NOTEMPTY
1308 nigel 75 .sp
1309 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1310     there are alternatives in the pattern, they are tried. If all the alternatives
1311     match the empty string, the entire match fails. For example, if the pattern
1312 nigel 75 .sp
1313 nigel 63 a?b?
1314 nigel 75 .sp
1315 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1316     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1317     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1318 nigel 75 .P
1319 nigel 63 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1320 nigel 75 of a pattern match of the empty string within its \fBsplit()\fP function, and
1321 nigel 63 when using the /g modifier. It is possible to emulate Perl's behaviour after
1322     matching a null string by first trying the match again at the same offset with
1323 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1324     starting offset (see below) and trying an ordinary match again. There is some
1325     code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1326     .sp
1327     PCRE_NO_UTF8_CHECK
1328     .sp
1329     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1330     string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1331 ph10 211 The value of \fIstartoffset\fP is also checked to ensure that it points to the
1332     start of a UTF-8 character. There is a discussion about the validity of UTF-8
1333     strings in the
1334     .\" HTML <a href="pcre.html#utf8strings">
1335     .\" </a>
1336     section on UTF-8 support
1337     .\"
1338     in the main
1339     .\" HREF
1340     \fBpcre\fP
1341     .\"
1342     page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1343     the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1344 ph10 209 PCRE_ERROR_BADUTF8_OFFSET is returned.
1345 nigel 75 .P
1346     If you already know that your subject is valid, and you want to skip these
1347     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1348     calling \fBpcre_exec()\fP. You might want to do this for the second and
1349     subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1350     all the matches in a single subject string. However, you should be sure that
1351     the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1352     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1353     subject, or a value of \fIstartoffset\fP that does not point to the start of a
1354     UTF-8 character, is undefined. Your program may crash.
1355     .sp
1356     PCRE_PARTIAL
1357     .sp
1358     This option turns on the partial matching feature. If the subject string fails
1359     to match the pattern, but at some point during the matching process the end of
1360     the subject was reached (that is, the subject partially matches the pattern and
1361     the failure to match occurred only because there were not enough subject
1362     characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1363     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1364     may appear in the pattern. These are discussed in the
1365     .\" HREF
1366     \fBpcrepartial\fP
1367     .\"
1368     documentation.
1369     .
1370     .SS "The string to be matched by \fBpcre_exec()\fP"
1371     .rs
1372     .sp
1373     The subject string is passed to \fBpcre_exec()\fP as a pointer in
1374 ph10 368 \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1375     in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of
1376     a UTF-8 character. Unlike the pattern string, the subject may contain binary
1377     zero bytes. When the starting offset is zero, the search for a match starts at
1378     the beginning of the subject, and this is by far the most common case.
1379 nigel 75 .P
1380 nigel 63 A non-zero starting offset is useful when searching for another match in the
1381 nigel 75 same subject by calling \fBpcre_exec()\fP again after a previous success.
1382     Setting \fIstartoffset\fP differs from just passing over a shortened string and
1383 nigel 63 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1384     lookbehind. For example, consider the pattern
1385 nigel 75 .sp
1386     \eBiss\eB
1387     .sp
1388     which finds occurrences of "iss" in the middle of words. (\eB matches only if
1389 nigel 63 the current position in the subject is not a word boundary.) When applied to
1390 nigel 75 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1391     occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1392     subject, namely "issipi", it does not match, because \eB is always false at the
1393 nigel 63 start of the subject, which is deemed to be a word boundary. However, if
1394 nigel 75 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1395 nigel 63 set to 4, it finds the second occurrence of "iss" because it is able to look
1396     behind the starting point to discover that it is preceded by a letter.
1397 nigel 75 .P
1398 nigel 63 If a non-zero starting offset is passed when the pattern is anchored, one
1399 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1400 nigel 63 pattern does not require the match to be at the start of the subject.
1401 nigel 75 .
1402     .SS "How \fBpcre_exec()\fP returns captured substrings"
1403     .rs
1404     .sp
1405 nigel 63 In general, a pattern matches a certain portion of the subject, and in
1406     addition, further substrings from the subject may be picked out by parts of the
1407     pattern. Following the usage in Jeffrey Friedl's book, this is called
1408     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1409     a fragment of a pattern that picks out a substring. PCRE supports several other
1410     kinds of parenthesized subpattern that do not cause substrings to be captured.
1411 nigel 75 .P
1412 ph10 368 Captured substrings are returned to the caller via a vector of integers whose
1413     address is passed in \fIovector\fP. The number of elements in the vector is
1414     passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1415     argument is NOT the size of \fIovector\fP in bytes.
1416 nigel 75 .P
1417     The first two-thirds of the vector is used to pass back captured substrings,
1418     each substring using a pair of integers. The remaining third of the vector is
1419     used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1420 ph10 368 and is not available for passing back information. The number passed in
1421 nigel 75 \fIovecsize\fP should always be a multiple of three. If it is not, it is
1422     rounded down.
1423     .P
1424     When a match is successful, information about captured substrings is returned
1425     in pairs of integers, starting at the beginning of \fIovector\fP, and
1426 ph10 371 continuing up to two-thirds of its length at the most. The first element of
1427 ph10 368 each pair is set to the byte offset of the first character in a substring, and
1428     the second is set to the byte offset of the first character after the end of a
1429     substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1430     mode. They are not character counts.
1431 nigel 75 .P
1432 ph10 368 The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1433     portion of the subject string matched by the entire pattern. The next pair is
1434     used for the first capturing subpattern, and so on. The value returned by
1435     \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1436     For example, if two substrings have been captured, the returned value is 3. If
1437     there are no capturing subpatterns, the return value from a successful match is
1438     1, indicating that just the first pair of offsets has been set.
1439     .P
1440 nigel 63 If a capturing subpattern is matched repeatedly, it is the last portion of the
1441 nigel 75 string that it matched that is returned.
1442     .P
1443     If the vector is too small to hold all the captured substring offsets, it is
1444     used as far as possible (up to two-thirds of its length), and the function
1445 ph10 368 returns a value of zero. If the substring offsets are not of interest,
1446     \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1447 nigel 75 \fIovecsize\fP as zero. However, if the pattern contains back references and
1448     the \fIovector\fP is not big enough to remember the related substrings, PCRE
1449     has to get additional memory for use during matching. Thus it is usually
1450     advisable to supply an \fIovector\fP.
1451     .P
1452 nigel 91 The \fBpcre_info()\fP function can be used to find out how many capturing
1453 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1454 nigel 75 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1455     the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1456 nigel 91 .P
1457     It is possible for capturing subpattern number \fIn+1\fP to match some part of
1458     the subject when subpattern \fIn\fP has not been used at all. For example, if
1459     the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1460     function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1461     happens, both values in the offset pairs corresponding to unused subpatterns
1462     are set to -1.
1463     .P
1464     Offset values that correspond to unused subpatterns at the end of the
1465     expression are also set to -1. For example, if the string "abc" is matched
1466     against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1467     return from the function is 2, because the highest used capturing subpattern
1468     number is 1. However, you can refer to the offsets for the second and third
1469     capturing subpatterns if you wish (assuming the vector is large enough, of
1470     course).
1471     .P
1472     Some convenience functions are provided for extracting the captured substrings
1473     as separate strings. These are described below.
1474 nigel 75 .
1475 nigel 77 .\" HTML <a name="errorlist"></a>
1476 nigel 91 .SS "Error return values from \fBpcre_exec()\fP"
1477 nigel 75 .rs
1478     .sp
1479     If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1480 nigel 63 defined in the header file:
1481 nigel 75 .sp
1482 nigel 63 PCRE_ERROR_NOMATCH (-1)
1483 nigel 75 .sp
1484 nigel 63 The subject string did not match the pattern.
1485 nigel 75 .sp
1486 nigel 63 PCRE_ERROR_NULL (-2)
1487 nigel 75 .sp
1488     Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1489     NULL and \fIovecsize\fP was not zero.
1490     .sp
1491 nigel 63 PCRE_ERROR_BADOPTION (-3)
1492 nigel 75 .sp
1493     An unrecognized bit was set in the \fIoptions\fP argument.
1494     .sp
1495 nigel 63 PCRE_ERROR_BADMAGIC (-4)
1496 nigel 75 .sp
1497 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1498 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1499     compiled in an environment of one endianness is run in an environment with the
1500     other endianness. This is the error that PCRE gives when the magic number is
1501     not present.
1502     .sp
1503 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1504 nigel 75 .sp
1505 nigel 63 While running the pattern match, an unknown item was encountered in the
1506     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1507     of the compiled pattern.
1508 nigel 75 .sp
1509 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1510 nigel 75 .sp
1511     If a pattern contains back references, but the \fIovector\fP that is passed to
1512     \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1513 nigel 63 gets a block of memory at the start of matching to use for this purpose. If the
1514 nigel 75 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1515     automatically freed at the end of matching.
1516     .sp
1517 nigel 63 PCRE_ERROR_NOSUBSTRING (-7)
1518 nigel 75 .sp
1519     This error is used by the \fBpcre_copy_substring()\fP,
1520     \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1521     below). It is never returned by \fBpcre_exec()\fP.
1522     .sp
1523 nigel 63 PCRE_ERROR_MATCHLIMIT (-8)
1524 nigel 75 .sp
1525 nigel 87 The backtracking limit, as specified by the \fImatch_limit\fP field in a
1526     \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1527     above.
1528     .sp
1529 nigel 63 PCRE_ERROR_CALLOUT (-9)
1530 nigel 75 .sp
1531     This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1532 nigel 63 use by callout functions that want to yield a distinctive error code. See the
1533 nigel 75 .\" HREF
1534     \fBpcrecallout\fP
1535     .\"
1536     documentation for details.
1537     .sp
1538 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1539 nigel 75 .sp
1540 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1541 nigel 75 .sp
1542 nigel 73 PCRE_ERROR_BADUTF8_OFFSET (-11)
1543 nigel 75 .sp
1544 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1545 nigel 75 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1546     .sp
1547 nigel 77 PCRE_ERROR_PARTIAL (-12)
1548 nigel 75 .sp
1549     The subject string did not match, but it did match partially. See the
1550     .\" HREF
1551     \fBpcrepartial\fP
1552     .\"
1553     documentation for details of partial matching.
1554     .sp
1555 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1556 nigel 75 .sp
1557     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1558     are not supported for partial matching. See the
1559     .\" HREF
1560     \fBpcrepartial\fP
1561     .\"
1562     documentation for details of partial matching.
1563     .sp
1564 nigel 77 PCRE_ERROR_INTERNAL (-14)
1565 nigel 75 .sp
1566     An unexpected internal error has occurred. This error could be caused by a bug
1567     in PCRE or by overwriting of the compiled pattern.
1568     .sp
1569 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1570 nigel 75 .sp
1571     This error is given if the value of the \fIovecsize\fP argument is negative.
1572 nigel 93 .sp
1573     PCRE_ERROR_RECURSIONLIMIT (-21)
1574     .sp
1575     The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1576     field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1577     description above.
1578     .sp
1579     PCRE_ERROR_BADNEWLINE (-23)
1580     .sp
1581     An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1582     .P
1583 ph10 197 Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1584 nigel 75 .
1585     .
1586     .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1587 nigel 63 .rs
1588     .sp
1589 nigel 75 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1590 nigel 63 .ti +5n
1591 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1592 nigel 63 .ti +5n
1593 nigel 75 .B int \fIbuffersize\fP);
1594 nigel 63 .PP
1595 nigel 75 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1596 nigel 63 .ti +5n
1597 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1598 nigel 63 .ti +5n
1599 nigel 75 .B const char **\fIstringptr\fP);
1600 nigel 63 .PP
1601 nigel 75 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1602 nigel 63 .ti +5n
1603 nigel 75 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1604 nigel 63 .PP
1605     Captured substrings can be accessed directly by using the offsets returned by
1606 nigel 75 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1607     \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1608     \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1609 nigel 63 as new, separate, zero-terminated strings. These functions identify substrings
1610     by number. The next section describes functions for extracting named
1611 nigel 91 substrings.
1612 nigel 75 .P
1613 nigel 91 A substring that contains a binary zero is correctly extracted and has a
1614     further zero added on the end, but the result is not, of course, a C string.
1615     However, you can process such a string by referring to the length that is
1616     returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1617     Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1618     for handling strings containing binary zeros, because the end of the final
1619     string is not independently indicated.
1620     .P
1621 nigel 63 The first three arguments are the same for all three of these functions:
1622 nigel 75 \fIsubject\fP is the subject string that has just been successfully matched,
1623     \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1624     \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1625 nigel 63 captured by the match, including the substring that matched the entire regular
1626 nigel 75 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1627     than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1628     space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1629     number of elements in the vector divided by three.
1630     .P
1631     The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1632     extract a single substring, whose number is given as \fIstringnumber\fP. A
1633     value of zero extracts the substring that matched the entire pattern, whereas
1634     higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1635     the string is placed in \fIbuffer\fP, whose length is given by
1636     \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1637     obtained via \fBpcre_malloc\fP, and its address is returned via
1638     \fIstringptr\fP. The yield of the function is the length of the string, not
1639 nigel 93 including the terminating zero, or one of these error codes:
1640 nigel 75 .sp
1641 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1642 nigel 75 .sp
1643     The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1644     memory failed for \fBpcre_get_substring()\fP.
1645     .sp
1646 nigel 63 PCRE_ERROR_NOSUBSTRING (-7)
1647 nigel 75 .sp
1648     There is no substring whose number is \fIstringnumber\fP.
1649     .P
1650     The \fBpcre_get_substring_list()\fP function extracts all available substrings
1651 nigel 63 and builds a list of pointers to them. All this is done in a single block of
1652 nigel 75 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1653     is returned via \fIlistptr\fP, which is also the start of the list of string
1654 nigel 63 pointers. The end of the list is marked by a NULL pointer. The yield of the
1655 nigel 93 function is zero if all went well, or the error code
1656 nigel 75 .sp
1657 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1658 nigel 75 .sp
1659 nigel 63 if the attempt to get the memory block failed.
1660 nigel 75 .P
1661 nigel 63 When any of these functions encounter a substring that is unset, which can
1662 nigel 75 happen when capturing subpattern number \fIn+1\fP matches some part of the
1663     subject, but subpattern \fIn\fP has not been used at all, they return an empty
1664 nigel 63 string. This can be distinguished from a genuine zero-length substring by
1665 nigel 75 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1666 nigel 63 substrings.
1667 nigel 75 .P
1668     The two convenience functions \fBpcre_free_substring()\fP and
1669     \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1670     a previous call of \fBpcre_get_substring()\fP or
1671     \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1672     the function pointed to by \fBpcre_free\fP, which of course could be called
1673 nigel 63 directly from a C program. However, PCRE is used in some situations where it is
1674 nigel 91 linked via a special interface to another programming language that cannot use
1675 nigel 75 \fBpcre_free\fP directly; it is for these cases that the functions are
1676 nigel 63 provided.
1677 nigel 75 .
1678     .
1679     .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1680 nigel 63 .rs
1681     .sp
1682 nigel 75 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1683 nigel 63 .ti +5n
1684 nigel 75 .B const char *\fIname\fP);
1685 nigel 63 .PP
1686 nigel 75 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1687 nigel 63 .ti +5n
1688 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
1689     .ti +5n
1690     .B int \fIstringcount\fP, const char *\fIstringname\fP,
1691     .ti +5n
1692     .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1693 nigel 63 .PP
1694 nigel 75 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1695 nigel 63 .ti +5n
1696 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
1697 nigel 63 .ti +5n
1698 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1699 nigel 63 .ti +5n
1700 nigel 75 .B const char **\fIstringptr\fP);
1701 nigel 63 .PP
1702 nigel 75 To extract a substring by name, you first have to find associated number.
1703     For example, for this pattern
1704     .sp
1705 nigel 93 (a+)b(?<xxx>\ed+)...
1706 nigel 75 .sp
1707 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to be
1708     unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1709     calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1710     pattern, and the second is the name. The yield of the function is the
1711 nigel 75 subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1712     that name.
1713     .P
1714     Given the number, you can extract the substring directly, or use one of the
1715     functions described in the previous section. For convenience, there are also
1716     two functions that do the whole job.
1717     .P
1718 nigel 91 Most of the arguments of \fBpcre_copy_named_substring()\fP and
1719     \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1720 nigel 75 functions that extract by number. As these are described in the previous
1721     section, they are not re-described here. There are just two differences:
1722     .P
1723 nigel 63 First, instead of a substring number, a substring name is given. Second, there
1724     is an extra argument, given at the start, which is a pointer to the compiled
1725     pattern. This is needed in order to gain access to the name-to-number
1726     translation table.
1727 nigel 75 .P
1728     These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1729 ph10 127 then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1730     appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1731 ph10 128 the behaviour may not be what you want (see the next section).
1732 nigel 77 .
1733     .
1734 nigel 91 .SH "DUPLICATE SUBPATTERN NAMES"
1735     .rs
1736     .sp
1737     .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1738     .ti +5n
1739     .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1740     .PP
1741     When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1742     are not required to be unique. Normally, patterns with duplicate names are such
1743     that in any one match, only one of the named subpatterns participates. An
1744     example is shown in the
1745     .\" HREF
1746     \fBpcrepattern\fP
1747     .\"
1748 ph10 208 documentation.
1749 ph10 203 .P
1750     When duplicates are present, \fBpcre_copy_named_substring()\fP and
1751     \fBpcre_get_named_substring()\fP return the first substring corresponding to
1752     the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1753     returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1754     returns one of the numbers that are associated with the name, but it is not
1755     defined which it is.
1756     .P
1757 nigel 91 If you want to get full details of all captured substrings for a given name,
1758     you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1759     argument is the compiled pattern, and the second is the name. The third and
1760     fourth are pointers to variables which are updated by the function. After it
1761     has run, they point to the first and last entries in the name-to-number table
1762     for the given name. The function itself returns the length of each entry, or
1763 nigel 93 PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1764     described above in the section entitled \fIInformation about a pattern\fP.
1765     Given all the relevant entries for the name, you can extract each of their
1766     numbers, and hence the captured data, if any.
1767 nigel 91 .
1768     .
1769 nigel 77 .SH "FINDING ALL POSSIBLE MATCHES"
1770     .rs
1771     .sp
1772     The traditional matching function uses a similar algorithm to Perl, which stops
1773     when it finds the first match, starting at a given point in the subject. If you
1774     want to find all possible matches, or the longest possible match, consider
1775     using the alternative matching function (see below) instead. If you cannot use
1776     the alternative function, but still need to find all possible matches, you
1777     can kludge it up by making use of the callout facility, which is described in
1778     the
1779     .\" HREF
1780     \fBpcrecallout\fP
1781     .\"
1782     documentation.
1783 nigel 75 .P
1784 nigel 77 What you have to do is to insert a callout right at the end of the pattern.
1785     When your callout function is called, extract and save the current matched
1786     substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1787     other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1788     will yield PCRE_ERROR_NOMATCH.
1789     .
1790     .
1791     .\" HTML <a name="dfamatch"></a>
1792     .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1793     .rs
1794     .sp
1795     .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1796     .ti +5n
1797     .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1798     .ti +5n
1799     .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1800     .ti +5n
1801     .B int *\fIworkspace\fP, int \fIwscount\fP);
1802     .P
1803     The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1804 nigel 93 a compiled pattern, using a matching algorithm that scans the subject string
1805     just once, and does not backtrack. This has different characteristics to the
1806     normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1807     patterns are not supported. Nevertheless, there are times when this kind of
1808     matching can be useful. For a discussion of the two matching algorithms, see
1809     the
1810 nigel 77 .\" HREF
1811     \fBpcrematching\fP
1812     .\"
1813     documentation.
1814     .P
1815     The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1816     \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1817     different way, and this is described below. The other common arguments are used
1818     in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1819     here.
1820     .P
1821     The two additional arguments provide workspace for the function. The workspace
1822     vector should contain at least 20 elements. It is used for keeping track of
1823     multiple paths through the pattern tree. More workspace will be needed for
1824 nigel 91 patterns and subjects where there are a lot of potential matches.
1825 nigel 77 .P
1826 nigel 87 Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1827 nigel 77 .sp
1828     int rc;
1829     int ovector[10];
1830     int wspace[20];
1831 nigel 87 rc = pcre_dfa_exec(
1832 nigel 77 re, /* result of pcre_compile() */
1833     NULL, /* we didn't study the pattern */
1834     "some string", /* the subject string */
1835     11, /* the length of the subject string */
1836     0, /* start at offset 0 in the subject */
1837     0, /* default options */
1838     ovector, /* vector of integers for substring information */
1839     10, /* number of elements (NOT size in bytes) */
1840     wspace, /* working space vector */
1841     20); /* number of elements (NOT size in bytes) */
1842     .
1843     .SS "Option bits for \fBpcre_dfa_exec()\fP"
1844     .rs
1845     .sp
1846     The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1847 nigel 91 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1848     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1849 nigel 77 PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1850     the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1851     .sp
1852     PCRE_PARTIAL
1853     .sp
1854     This has the same general effect as it does for \fBpcre_exec()\fP, but the
1855     details are slightly different. When PCRE_PARTIAL is set for
1856     \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1857     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1858     complete matches, but there is still at least one matching possibility. The
1859     portion of the string that provided the partial match is set as the first
1860     matching string.
1861     .sp
1862     PCRE_DFA_SHORTEST
1863     .sp
1864     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1865 nigel 93 soon as it has found one match. Because of the way the alternative algorithm
1866     works, this is necessarily the shortest possible match at the first possible
1867     matching point in the subject string.
1868 nigel 77 .sp
1869     PCRE_DFA_RESTART
1870     .sp
1871     When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1872     a partial match, it is possible to call it again, with additional subject
1873     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1874     option requests this action; when it is set, the \fIworkspace\fP and
1875     \fIwscount\fP options must reference the same vector as before because data
1876     about the match so far is left in them after a partial match. There is more
1877     discussion of this facility in the
1878     .\" HREF
1879     \fBpcrepartial\fP
1880     .\"
1881     documentation.
1882     .
1883     .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1884     .rs
1885     .sp
1886     When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1887     substring in the subject. Note, however, that all the matches from one run of
1888     the function start at the same point in the subject. The shorter matches are
1889     all initial substrings of the longer matches. For example, if the pattern
1890     .sp
1891     <.*>
1892     .sp
1893     is matched against the string
1894     .sp
1895     This is <something> <something else> <something further> no more
1896     .sp
1897     the three matched strings are
1898     .sp
1899     <something>
1900     <something> <something else>
1901     <something> <something else> <something further>
1902     .sp
1903     On success, the yield of the function is a number greater than zero, which is
1904     the number of matched substrings. The substrings themselves are returned in
1905     \fIovector\fP. Each string uses two elements; the first is the offset to the
1906 nigel 93 start, and the second is the offset to the end. In fact, all the strings have
1907     the same start offset. (Space could have been saved by giving this only once,
1908     but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1909     returns data, even though the meaning of the strings is different.)
1910 nigel 77 .P
1911     The strings are returned in reverse order of length; that is, the longest
1912     matching string is given first. If there were too many matches to fit into
1913     \fIovector\fP, the yield of the function is zero, and the vector is filled with
1914     the longest matches.
1915     .
1916     .SS "Error returns from \fBpcre_dfa_exec()\fP"
1917     .rs
1918     .sp
1919     The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1920     Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1921     described
1922     .\" HTML <a href="#errorlist">
1923     .\" </a>
1924     above.
1925     .\"
1926     There are in addition the following errors that are specific to
1927     \fBpcre_dfa_exec()\fP:
1928     .sp
1929     PCRE_ERROR_DFA_UITEM (-16)
1930     .sp
1931     This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1932     that it does not support, for instance, the use of \eC or a back reference.
1933     .sp
1934     PCRE_ERROR_DFA_UCOND (-17)
1935     .sp
1936 nigel 93 This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1937     uses a back reference for the condition, or a test for recursion in a specific
1938     group. These are not supported.
1939 nigel 77 .sp
1940     PCRE_ERROR_DFA_UMLIMIT (-18)
1941     .sp
1942     This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1943     block that contains a setting of the \fImatch_limit\fP field. This is not
1944     supported (it is meaningless).
1945     .sp
1946     PCRE_ERROR_DFA_WSSIZE (-19)
1947     .sp
1948     This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1949     \fIworkspace\fP vector.
1950     .sp
1951     PCRE_ERROR_DFA_RECURSE (-20)
1952     .sp
1953     When a recursive subpattern is processed, the matching function calls itself
1954     recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1955     error is given if the output vector is not large enough. This should be
1956     extremely rare, as a vector of size 1000 is used.
1957 nigel 93 .
1958     .
1959     .SH "SEE ALSO"
1960     .rs
1961     .sp
1962     \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1963     \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1964     \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1965 ph10 99 .
1966     .
1967     .SH AUTHOR
1968     .rs
1969     .sp
1970     .nf
1971     Philip Hazel
1972     University Computing Service
1973     Cambridge CB2 3QH, England.
1974     .fi
1975     .
1976     .
1977     .SH REVISION
1978     .rs
1979     .sp
1980     .nf
1981 ph10 376 Last updated: 28 February 2009
1982     Copyright (c) 1997-2009 University of Cambridge.
1983 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