ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 79 - (hide annotations) (download)
Sat Feb 24 21:40:52 2007 UTC (8 years, 2 months ago) by nigel
File size: 68188 byte(s)
Load pcre-6.1 into code/trunk.

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     .br
11 nigel 75 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
12 nigel 63 .ti +5n
13 nigel 75 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
14 nigel 63 .ti +5n
15 nigel 75 .B const unsigned char *\fItableptr\fP);
16 nigel 63 .PP
17     .br
18 nigel 77 .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
19     .ti +5n
20     .B int *\fIerrorcodeptr\fP,
21     .ti +5n
22     .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
23     .ti +5n
24     .B const unsigned char *\fItableptr\fP);
25     .PP
26     .br
27 nigel 75 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
28 nigel 63 .ti +5n
29 nigel 75 .B const char **\fIerrptr\fP);
30 nigel 63 .PP
31     .br
32 nigel 75 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
33 nigel 63 .ti +5n
34 nigel 75 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
35 nigel 63 .ti +5n
36 nigel 75 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
37 nigel 63 .PP
38     .br
39 nigel 77 .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
40     .ti +5n
41     .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
42     .ti +5n
43     .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
44     .ti +5n
45     .B int *\fIworkspace\fP, int \fIwscount\fP);
46     .PP
47     .br
48 nigel 75 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
49 nigel 63 .ti +5n
50 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
51 nigel 63 .ti +5n
52 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
53 nigel 63 .ti +5n
54 nigel 75 .B char *\fIbuffer\fP, int \fIbuffersize\fP);
55 nigel 63 .PP
56     .br
57 nigel 75 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
58 nigel 63 .ti +5n
59 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
60 nigel 63 .ti +5n
61 nigel 75 .B int \fIbuffersize\fP);
62 nigel 63 .PP
63     .br
64 nigel 75 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
65 nigel 63 .ti +5n
66 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
67 nigel 63 .ti +5n
68 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
69 nigel 63 .ti +5n
70 nigel 75 .B const char **\fIstringptr\fP);
71 nigel 63 .PP
72     .br
73 nigel 75 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
74 nigel 63 .ti +5n
75 nigel 75 .B const char *\fIname\fP);
76 nigel 63 .PP
77     .br
78 nigel 75 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
79 nigel 63 .ti +5n
80 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP,
81 nigel 63 .ti +5n
82 nigel 75 .B const char **\fIstringptr\fP);
83 nigel 63 .PP
84     .br
85 nigel 75 .B int pcre_get_substring_list(const char *\fIsubject\fP,
86 nigel 63 .ti +5n
87 nigel 75 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
88 nigel 63 .PP
89     .br
90 nigel 75 .B void pcre_free_substring(const char *\fIstringptr\fP);
91 nigel 63 .PP
92     .br
93 nigel 75 .B void pcre_free_substring_list(const char **\fIstringptr\fP);
94 nigel 63 .PP
95     .br
96     .B const unsigned char *pcre_maketables(void);
97     .PP
98     .br
99 nigel 75 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
100 nigel 63 .ti +5n
101 nigel 75 .B int \fIwhat\fP, void *\fIwhere\fP);
102 nigel 63 .PP
103     .br
104 nigel 75 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
105     .B *\fIfirstcharptr\fP);
106 nigel 63 .PP
107     .br
108 nigel 77 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
109     .PP
110     .br
111 nigel 75 .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
112 nigel 63 .PP
113     .br
114     .B char *pcre_version(void);
115     .PP
116     .br
117     .B void *(*pcre_malloc)(size_t);
118     .PP
119     .br
120     .B void (*pcre_free)(void *);
121     .PP
122     .br
123 nigel 73 .B void *(*pcre_stack_malloc)(size_t);
124     .PP
125     .br
126     .B void (*pcre_stack_free)(void *);
127     .PP
128     .br
129 nigel 63 .B int (*pcre_callout)(pcre_callout_block *);
130 nigel 75 .
131     .
133 nigel 63 .rs
134     .sp
135 nigel 77 PCRE has its own native API, which is described in this document. There is
136     also a set of wrapper functions that correspond to the POSIX regular expression
137     API. These are described in the
138 nigel 75 .\" HREF
139     \fBpcreposix\fP
140     .\"
141 nigel 77 documentation. Both of these APIs define a set of C function calls. A C++
142     wrapper is distributed with PCRE. It is documented in the
143     .\" HREF
144     \fBpcrecpp\fP
145     .\"
146     page.
147 nigel 75 .P
148 nigel 77 The native API C function prototypes are defined in the header file
149     \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
150     It can normally be accessed by adding \fB-lpcre\fP to the command for linking
151     an application that uses PCRE. The header file defines the macros PCRE_MAJOR
152     and PCRE_MINOR to contain the major and minor release numbers for the library.
153 nigel 75 Applications can use these to include support for different releases of PCRE.
154     .P
155 nigel 77 The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
156     and \fBpcre_exec()\fP are used for compiling and matching regular expressions
157     in a Perl-compatible manner. A sample program that demonstrates the simplest
158     way of using them is provided in the file called \fIpcredemo.c\fP in the source
159     distribution. The
160 nigel 75 .\" HREF
161     \fBpcresample\fP
162     .\"
163     documentation describes how to run it.
164     .P
165 nigel 77 A second matching function, \fBpcre_dfa_exec()\fP, which is not
166     Perl-compatible, is also provided. This uses a different algorithm for the
167     matching. This allows it to find all possible matches (at a given point in the
168     subject), not just one. However, this algorithm does not return captured
169     substrings. A description of the two matching algorithms and their advantages
170     and disadvantages is given in the
171     .\" HREF
172     \fBpcrematching\fP
173     .\"
174     documentation.
175     .P
176 nigel 75 In addition to the main compiling and matching functions, there are convenience
177 nigel 77 functions for extracting captured substrings from a subject string that is
178     matched by \fBpcre_exec()\fP. They are:
179 nigel 75 .sp
180     \fBpcre_copy_substring()\fP
181     \fBpcre_copy_named_substring()\fP
182     \fBpcre_get_substring()\fP
183     \fBpcre_get_named_substring()\fP
184     \fBpcre_get_substring_list()\fP
185     \fBpcre_get_stringnumber()\fP
186     .sp
187     \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
188 nigel 63 provided, to free the memory used for extracted strings.
189 nigel 75 .P
190     The function \fBpcre_maketables()\fP is used to build a set of character tables
191 nigel 77 in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
192     or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
193     specialist use. Most commonly, no special tables are passed, in which case
194     internal tables that are generated when PCRE is built are used.
195 nigel 75 .P
196     The function \fBpcre_fullinfo()\fP is used to find out information about a
197     compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
198 nigel 63 some of the available information, but is retained for backwards compatibility.
199 nigel 75 The function \fBpcre_version()\fP returns a pointer to a string containing the
200 nigel 63 version of PCRE and its date of release.
201 nigel 75 .P
202 nigel 77 The function \fBpcre_refcount()\fP maintains a reference count in a data block
203     containing a compiled pattern. This is provided for the benefit of
204     object-oriented applications.
205     .P
206 nigel 75 The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
207     the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
208 nigel 63 respectively. PCRE calls the memory management functions via these variables,
209     so a calling program can replace them if it wishes to intercept the calls. This
210     should be done before calling any PCRE functions.
211 nigel 75 .P
212     The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
213 nigel 73 indirections to memory management functions. These special functions are used
214     only when PCRE is compiled to use the heap for remembering data, instead of
215 nigel 77 recursive function calls, when running the \fBpcre_exec()\fP function. This is
216     a non-standard way of building PCRE, for use in environments that have limited
217     stacks. Because of the greater use of memory management, it runs more slowly.
218     Separate functions are provided so that special-purpose external code can be
219     used for this case. When used, these functions are always called in a
220     stack-like manner (last obtained, first freed), and always for memory blocks of
221     the same size.
222 nigel 75 .P
223     The global variable \fBpcre_callout\fP initially contains NULL. It can be set
224 nigel 63 by the caller to a "callout" function, which PCRE will then call at specified
225 nigel 75 points during a matching operation. Details are given in the
226     .\" HREF
227     \fBpcrecallout\fP
228     .\"
229 nigel 63 documentation.
230 nigel 75 .
231     .
233     .rs
234     .sp
235     The PCRE functions can be used in multi-threading applications, with the
236 nigel 75 proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
237     \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
238     callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
239     .P
240 nigel 63 The compiled form of a regular expression is not altered during matching, so
241     the same compiled pattern can safely be used by several threads at once.
242 nigel 75 .
243     .
245 nigel 63 .rs
246     .sp
247 nigel 75 The compiled form of a regular expression can be saved and re-used at a later
248     time, possibly by a different program, and even on a host other than the one on
249     which it was compiled. Details are given in the
250     .\" HREF
251     \fBpcreprecompile\fP
252     .\"
253     documentation.
254     .
255     .
257     .rs
258     .sp
259     .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
260 nigel 63 .PP
261 nigel 75 The function \fBpcre_config()\fP makes it possible for a PCRE client to
262 nigel 63 discover which optional features have been compiled into the PCRE library. The
263     .\" HREF
264 nigel 75 \fBpcrebuild\fP
265 nigel 63 .\"
266     documentation has more details about these optional features.
267 nigel 75 .P
268     The first argument for \fBpcre_config()\fP is an integer, specifying which
269 nigel 63 information is required; the second argument is a pointer to a variable into
270     which the information is placed. The following information is available:
271 nigel 75 .sp
272 nigel 63 PCRE_CONFIG_UTF8
273 nigel 75 .sp
274 nigel 63 The output is an integer that is set to one if UTF-8 support is available;
275     otherwise it is set to zero.
276 nigel 75 .sp
278     .sp
279     The output is an integer that is set to one if support for Unicode character
280     properties is available; otherwise it is set to zero.
281     .sp
283 nigel 75 .sp
284 nigel 63 The output is an integer that is set to the value of the code that is used for
285     the newline character. It is either linefeed (10) or carriage return (13), and
286     should normally be the standard character for your operating system.
287 nigel 75 .sp
289 nigel 75 .sp
290 nigel 63 The output is an integer that contains the number of bytes used for internal
291     linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
292     allow larger regular expressions to be compiled, at the expense of slower
293     matching. The default value of 2 is sufficient for all but the most massive
294     patterns, since it allows the compiled pattern to be up to 64K in size.
295 nigel 75 .sp
297 nigel 75 .sp
298 nigel 63 The output is an integer that contains the threshold above which the POSIX
299 nigel 75 interface uses \fBmalloc()\fP for output vectors. Further details are given in
300     the
301     .\" HREF
302     \fBpcreposix\fP
303     .\"
304     documentation.
305     .sp
307 nigel 75 .sp
308 nigel 63 The output is an integer that gives the default limit for the number of
309 nigel 75 internal matching function calls in a \fBpcre_exec()\fP execution. Further
310     details are given with \fBpcre_exec()\fP below.
311     .sp
313 nigel 75 .sp
314 nigel 77 The output is an integer that is set to one if internal recursion when running
315     \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
316     to remember their state. This is the usual way that PCRE is compiled. The
317     output is zero if PCRE was compiled to use blocks of data on the heap instead
318     of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
319     \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
320     avoiding the use of the stack.
321 nigel 75 .
322     .
324 nigel 63 .rs
325     .sp
326 nigel 75 .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
327 nigel 63 .ti +5n
328 nigel 75 .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
329 nigel 63 .ti +5n
330 nigel 75 .B const unsigned char *\fItableptr\fP);
331 nigel 77 .sp
332     .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
333     .ti +5n
334     .B int *\fIerrorcodeptr\fP,
335     .ti +5n
336     .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
337     .ti +5n
338     .B const unsigned char *\fItableptr\fP);
339 nigel 75 .P
340 nigel 77 Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
341     called to compile a pattern into an internal form. The only difference between
342     the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
343     \fIerrorcodeptr\fP, via which a numerical error code can be returned.
344 nigel 75 .P
345 nigel 77 The pattern is a C string terminated by a binary zero, and is passed in the
346     \fIpattern\fP argument. A pointer to a single block of memory that is obtained
347     via \fBpcre_malloc\fP is returned. This contains the compiled code and related
348     data. The \fBpcre\fP type is defined for the returned block; this is a typedef
349     for a structure whose contents are not externally defined. It is up to the
350     caller to free the memory when it is no longer required.
351     .P
352 nigel 63 Although the compiled code of a PCRE regex is relocatable, that is, it does not
353 nigel 75 depend on memory location, the complete \fBpcre\fP data block is not
354     fully relocatable, because it may contain a copy of the \fItableptr\fP
355     argument, which is an address (see below).
356     .P
357     The \fIoptions\fP argument contains independent bits that affect the
358     compilation. It should be zero if no options are required. The available
359     options are described below. Some of them, in particular, those that are
360     compatible with Perl, can also be set and unset from within the pattern (see
361     the detailed description in the
362     .\" HREF
363     \fBpcrepattern\fP
364     .\"
365     documentation). For these options, the contents of the \fIoptions\fP argument
366     specifies their initial settings at the start of compilation and execution. The
367     PCRE_ANCHORED option can be set at the time of matching as well as at compile
368     time.
369     .P
370     If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
371     Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
372     NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
373 nigel 63 error message. The offset from the start of the pattern to the character where
374     the error was discovered is placed in the variable pointed to by
375 nigel 75 \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
376     .P
377 nigel 77 If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
378     \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
379     returned via this argument in the event of an error. This is in addition to the
380     textual error message. Error codes and messages are listed below.
381     .P
382 nigel 75 If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
383     character tables that are built when PCRE is compiled, using the default C
384     locale. Otherwise, \fItableptr\fP must be an address that is the result of a
385     call to \fBpcre_maketables()\fP. This value is stored with the compiled
386     pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
387     passed to it. For more discussion, see the section on locale support below.
388     .P
389     This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
390     .sp
391 nigel 63 pcre *re;
392     const char *error;
393     int erroffset;
394     re = pcre_compile(
395     "^A.*Z", /* the pattern */
396     0, /* default options */
397     &error, /* for error message */
398     &erroffset, /* for error offset */
399     NULL); /* use default character tables */
400 nigel 75 .sp
401     The following names for option bits are defined in the \fBpcre.h\fP header
402     file:
403     .sp
404 nigel 63 PCRE_ANCHORED
405 nigel 75 .sp
406 nigel 63 If this bit is set, the pattern is forced to be "anchored", that is, it is
407 nigel 75 constrained to match only at the first matching point in the string that is
408 nigel 63 being searched (the "subject string"). This effect can also be achieved by
409     appropriate constructs in the pattern itself, which is the only way to do it in
410     Perl.
411 nigel 75 .sp
413     .sp
414     If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
415     all with number 255, before each pattern item. For discussion of the callout
416     facility, see the
417     .\" HREF
418     \fBpcrecallout\fP
419     .\"
420     documentation.
421     .sp
422 nigel 63 PCRE_CASELESS
423 nigel 75 .sp
424 nigel 63 If this bit is set, letters in the pattern match both upper and lower case
425     letters. It is equivalent to Perl's /i option, and it can be changed within a
426 nigel 77 pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
427     concept of case for characters whose values are less than 128, so caseless
428     matching is always possible. For characters with higher values, the concept of
429     case is supported if PCRE is compiled with Unicode property support, but not
430     otherwise. If you want to use caseless matching for characters 128 and above,
431     you must ensure that PCRE is compiled with Unicode property support as well as
432     with UTF-8 support.
433 nigel 75 .sp
435 nigel 75 .sp
436 nigel 63 If this bit is set, a dollar metacharacter in the pattern matches only at the
437     end of the subject string. Without this option, a dollar also matches
438     immediately before the final character if it is a newline (but not before any
439     other newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is
440     set. There is no equivalent to this option in Perl, and no way to set it within
441     a pattern.
442 nigel 75 .sp
443 nigel 63 PCRE_DOTALL
444 nigel 75 .sp
445 nigel 63 If this bit is set, a dot metacharater in the pattern matches all characters,
446     including newlines. Without it, newlines are excluded. This option is
447     equivalent to Perl's /s option, and it can be changed within a pattern by a
448     (?s) option setting. A negative class such as [^a] always matches a newline
449     character, independent of the setting of this option.
450 nigel 75 .sp
451 nigel 63 PCRE_EXTENDED
452 nigel 75 .sp
453 nigel 63 If this bit is set, whitespace data characters in the pattern are totally
454     ignored except when escaped or inside a character class. Whitespace does not
455     include the VT character (code 11). In addition, characters between an
456     unescaped # outside a character class and the next newline character,
457     inclusive, are also ignored. This is equivalent to Perl's /x option, and it can
458     be changed within a pattern by a (?x) option setting.
459 nigel 75 .P
460 nigel 63 This option makes it possible to include comments inside complicated patterns.
461     Note, however, that this applies only to data characters. Whitespace characters
462     may never appear within special character sequences in a pattern, for example
463     within the sequence (?( which introduces a conditional subpattern.
464 nigel 75 .sp
465 nigel 63 PCRE_EXTRA
466 nigel 75 .sp
467 nigel 63 This option was invented in order to turn on additional functionality of PCRE
468     that is incompatible with Perl, but it is currently of very little use. When
469     set, any backslash in a pattern that is followed by a letter that has no
470     special meaning causes an error, thus reserving these combinations for future
471     expansion. By default, as in Perl, a backslash followed by a letter with no
472     special meaning is treated as a literal. There are at present no other features
473     controlled by this option. It can also be set by a (?X) option setting within a
474     pattern.
475 nigel 75 .sp
476 nigel 77 PCRE_FIRSTLINE
477     .sp
478     If this option is set, an unanchored pattern is required to match before or at
479     the first newline character in the subject string, though the matched text may
480     continue over the newline.
481     .sp
482 nigel 63 PCRE_MULTILINE
483 nigel 75 .sp
484     By default, PCRE treats the subject string as consisting of a single line of
485     characters (even if it actually contains newlines). The "start of line"
486 nigel 63 metacharacter (^) matches only at the start of the string, while the "end of
487     line" metacharacter ($) matches only at the end of the string, or before a
488     terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
489     Perl.
490 nigel 75 .P
491 nigel 63 When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
492     match immediately following or immediately before any newline in the subject
493     string, respectively, as well as at the very start and end. This is equivalent
494     to Perl's /m option, and it can be changed within a pattern by a (?m) option
495 nigel 75 setting. If there are no "\en" characters in a subject string, or no
496 nigel 63 occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
497 nigel 75 .sp
499 nigel 75 .sp
500 nigel 63 If this option is set, it disables the use of numbered capturing parentheses in
501     the pattern. Any opening parenthesis that is not followed by ? behaves as if it
502     were followed by ?: but named parentheses can still be used for capturing (and
503     they acquire numbers in the usual way). There is no equivalent of this option
504     in Perl.
505 nigel 75 .sp
506 nigel 63 PCRE_UNGREEDY
507 nigel 75 .sp
508 nigel 63 This option inverts the "greediness" of the quantifiers so that they are not
509     greedy by default, but become greedy if followed by "?". It is not compatible
510     with Perl. It can also be set by a (?U) option setting within the pattern.
511 nigel 75 .sp
512 nigel 63 PCRE_UTF8
513 nigel 75 .sp
514 nigel 63 This option causes PCRE to regard both the pattern and the subject as strings
515     of UTF-8 characters instead of single-byte character strings. However, it is
516 nigel 75 available only when PCRE is built to include UTF-8 support. If not, the use
517 nigel 63 of this option provokes an error. Details of how this option changes the
518     behaviour of PCRE are given in the
519     .\" HTML <a href="pcre.html#utf8support">
520     .\" </a>
521     section on UTF-8 support
522     .\"
523     in the main
524     .\" HREF
525 nigel 75 \fBpcre\fP
526 nigel 63 .\"
527     page.
528 nigel 75 .sp
529 nigel 71 PCRE_NO_UTF8_CHECK
530 nigel 75 .sp
531 nigel 71 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
532     automatically checked. If an invalid UTF-8 sequence of bytes is found,
533 nigel 75 \fBpcre_compile()\fP returns an error. If you already know that your pattern is
534 nigel 71 valid, and you want to skip this check for performance reasons, you can set the
535     PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
536     UTF-8 string as a pattern is undefined. It may cause your program to crash.
537 nigel 77 Note that this option can also be passed to \fBpcre_exec()\fP and
538     \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject
539     strings.
540 nigel 75 .
541     .
543     .rs
544     .sp
545     The following table lists the error codes than may be returned by
546     \fBpcre_compile2()\fP, along with the error messages that may be returned by
547     both compiling functions.
548     .sp
549     0 no error
550     1 \e at end of pattern
551     2 \ec at end of pattern
552     3 unrecognized character follows \e
553     4 numbers out of order in {} quantifier
554     5 number too big in {} quantifier
555     6 missing terminating ] for character class
556     7 invalid escape sequence in character class
557     8 range out of order in character class
558     9 nothing to repeat
559     10 operand of unlimited repeat could match the empty string
560     11 internal error: unexpected repeat
561     12 unrecognized character after (?
562     13 POSIX named classes are supported only within a class
563     14 missing )
564     15 reference to non-existent subpattern
565     16 erroffset passed as NULL
566     17 unknown option bit(s) set
567     18 missing ) after comment
568     19 parentheses nested too deeply
569     20 regular expression too large
570     21 failed to get memory
571     22 unmatched parentheses
572     23 internal error: code overflow
573     24 unrecognized character after (?<
574     25 lookbehind assertion is not fixed length
575     26 malformed number after (?(
576     27 conditional group contains more than two branches
577     28 assertion expected after (?(
578     29 (?R or (?digits must be followed by )
579     30 unknown POSIX class name
580     31 POSIX collating elements are not supported
581     32 this version of PCRE is not compiled with PCRE_UTF8 support
582     33 spare error
583     34 character value in \ex{...} sequence is too large
584     35 invalid condition (?(0)
585     36 \eC not allowed in lookbehind assertion
586     37 PCRE does not support \eL, \el, \eN, \eU, or \eu
587     38 number after (?C is > 255
588     39 closing ) for (?C expected
589     40 recursive call could loop indefinitely
590     41 unrecognized character after (?P
591     42 syntax error after (?P
592     43 two named groups have the same name
593     44 invalid UTF-8 string
594     45 support for \eP, \ep, and \eX has not been compiled
595     46 malformed \eP or \ep sequence
596     47 unknown property name after \eP or \ep
597     .
598     .
599 nigel 75 .SH "STUDYING A PATTERN"
600 nigel 63 .rs
601     .sp
602 nigel 77 .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
603 nigel 63 .ti +5n
604 nigel 75 .B const char **\fIerrptr\fP);
605 nigel 63 .PP
606 nigel 75 If a compiled pattern is going to be used several times, it is worth spending
607     more time analyzing it in order to speed up the time taken for matching. The
608     function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
609     argument. If studying the pattern produces additional information that will
610     help speed up matching, \fBpcre_study()\fP returns a pointer to a
611     \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
612     results of the study.
613     .P
614     The returned value from \fBpcre_study()\fP can be passed directly to
615     \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
616 nigel 63 fields that can be set by the caller before the block is passed; these are
617 nigel 75 described
618     .\" HTML <a href="#extradata">
619     .\" </a>
620     below
621     .\"
622     in the section on matching a pattern.
623     .P
624 nigel 77 If studying the pattern does not produce any additional information
625 nigel 75 \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
626     wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
627     own \fBpcre_extra\fP block.
628     .P
629     The second argument of \fBpcre_study()\fP contains option bits. At present, no
630     options are defined, and this argument should always be zero.
631     .P
632     The third argument for \fBpcre_study()\fP is a pointer for an error message. If
633 nigel 63 studying succeeds (even if no data is returned), the variable it points to is
634     set to NULL. Otherwise it points to a textual error message. You should
635 nigel 75 therefore test the error pointer for NULL after calling \fBpcre_study()\fP, to
636 nigel 63 be sure that it has run successfully.
637 nigel 75 .P
638     This is a typical call to \fBpcre_study\fP():
639     .sp
640 nigel 63 pcre_extra *pe;
641     pe = pcre_study(
642     re, /* result of pcre_compile() */
643     0, /* no options exist */
644     &error); /* set to NULL or points to a message */
645 nigel 75 .sp
646 nigel 63 At present, studying a pattern is useful only for non-anchored patterns that do
647     not have a single fixed starting character. A bitmap of possible starting
648 nigel 75 bytes is created.
649     .
650     .
651 nigel 63 .\" HTML <a name="localesupport"></a>
652 nigel 75 .SH "LOCALE SUPPORT"
653 nigel 63 .rs
654     .sp
655 nigel 77 PCRE handles caseless matching, and determines whether characters are letters
656 nigel 75 digits, or whatever, by reference to a set of tables, indexed by character
657 nigel 77 value. When running in UTF-8 mode, this applies only to characters with codes
658 nigel 75 less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
659     can be tested with \ep if PCRE is built with Unicode character property
660 nigel 77 support.
661 nigel 75 .P
662     An internal set of tables is created in the default C locale when PCRE is
663     built. This is used when the final argument of \fBpcre_compile()\fP is NULL,
664     and is sufficient for many applications. An alternative set of tables can,
665     however, be supplied. These may be created in a different locale from the
666     default. As more and more applications change to using Unicode, the need for
667     this locale support is expected to die away.
668     .P
669     External tables are built by calling the \fBpcre_maketables()\fP function,
670     which has no arguments, in the relevant locale. The result can then be passed
671     to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
672     example, to build and use tables that are appropriate for the French locale
673     (where accented characters with values greater than 128 are treated as letters),
674     the following code could be used:
675     .sp
676     setlocale(LC_CTYPE, "fr_FR");
677 nigel 63 tables = pcre_maketables();
678     re = pcre_compile(..., tables);
679 nigel 75 .sp
680     When \fBpcre_maketables()\fP runs, the tables are built in memory that is
681     obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
682     that the memory containing the tables remains available for as long as it is
683     needed.
684     .P
685     The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
686     pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
687     and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
688     pattern, compilation, studying and matching all happen in the same locale, but
689     different patterns can be compiled in different locales.
690     .P
691     It is possible to pass a table pointer or NULL (indicating the use of the
692     internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
693     this facility could be used to match a pattern in a different locale from the
694     one in which it was compiled. Passing table pointers at run time is discussed
695     below in the section on matching a pattern.
696     .
697     .
699 nigel 63 .rs
700     .sp
701 nigel 75 .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
702 nigel 63 .ti +5n
703 nigel 75 .B int \fIwhat\fP, void *\fIwhere\fP);
704 nigel 63 .PP
705 nigel 75 The \fBpcre_fullinfo()\fP function returns information about a compiled
706     pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
707 nigel 63 nevertheless retained for backwards compability (and is documented below).
708 nigel 75 .P
709     The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
710     pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
711 nigel 63 the pattern was not studied. The third argument specifies which piece of
712     information is required, and the fourth argument is a pointer to a variable
713     to receive the data. The yield of the function is zero for success, or one of
714     the following negative numbers:
715 nigel 75 .sp
716     PCRE_ERROR_NULL the argument \fIcode\fP was NULL
717     the argument \fIwhere\fP was NULL
718 nigel 63 PCRE_ERROR_BADMAGIC the "magic number" was not found
719 nigel 75 PCRE_ERROR_BADOPTION the value of \fIwhat\fP was invalid
720     .sp
721     The "magic number" is placed at the start of each compiled pattern as an simple
722     check against passing an arbitrary memory pointer. Here is a typical call of
723     \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
724     .sp
725 nigel 63 int rc;
726     unsigned long int length;
727     rc = pcre_fullinfo(
728     re, /* result of pcre_compile() */
729     pe, /* result of pcre_study(), or NULL */
730     PCRE_INFO_SIZE, /* what is required */
731     &length); /* where to put the data */
732 nigel 75 .sp
733     The possible values for the third argument are defined in \fBpcre.h\fP, and are
734 nigel 63 as follows:
735 nigel 75 .sp
737 nigel 75 .sp
738 nigel 63 Return the number of the highest back reference in the pattern. The fourth
739 nigel 75 argument should point to an \fBint\fP variable. Zero is returned if there are
740 nigel 63 no back references.
741 nigel 75 .sp
743 nigel 75 .sp
744 nigel 63 Return the number of capturing subpatterns in the pattern. The fourth argument
745 nigel 75 should point to an \fBint\fP variable.
746     .sp
748 nigel 75 .sp
749     Return a pointer to the internal default character tables within PCRE. The
750     fourth argument should point to an \fBunsigned char *\fP variable. This
751     information call is provided for internal use by the \fBpcre_study()\fP
752     function. External callers can cause PCRE to use its internal tables by passing
753     a NULL table pointer.
754     .sp
756 nigel 75 .sp
757 nigel 63 Return information about the first byte of any matched string, for a
758     non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the
759     old name is still recognized for backwards compatibility.)
760 nigel 75 .P
761     If there is a fixed first byte, for example, from a pattern such as
762     (cat|cow|coyote), it is returned in the integer pointed to by \fIwhere\fP.
763     Otherwise, if either
764     .sp
765 nigel 63 (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
766     starts with "^", or
767 nigel 75 .sp
768 nigel 63 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
769     (if it were set, the pattern would be anchored),
770 nigel 75 .sp
771 nigel 63 -1 is returned, indicating that the pattern matches only at the start of a
772     subject string or after any newline within the string. Otherwise -2 is
773     returned. For anchored patterns, -2 is returned.
774 nigel 75 .sp
776 nigel 75 .sp
777 nigel 63 If the pattern was studied, and this resulted in the construction of a 256-bit
778     table indicating a fixed set of bytes for the first byte in any matching
779     string, a pointer to the table is returned. Otherwise NULL is returned. The
780 nigel 75 fourth argument should point to an \fBunsigned char *\fP variable.
781     .sp
783 nigel 75 .sp
784 nigel 65 Return the value of the rightmost literal byte that must exist in any matched
785     string, other than at its start, if such a byte has been recorded. The fourth
786 nigel 75 argument should point to an \fBint\fP variable. If there is no such byte, -1 is
787 nigel 65 returned. For anchored patterns, a last literal byte is recorded only if it
788     follows something of variable length. For example, for the pattern
789 nigel 75 /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
790 nigel 65 is -1.
791 nigel 75 .sp
795 nigel 75 .sp
796 nigel 63 PCRE supports the use of named as well as numbered capturing parentheses. The
797     names are just an additional way of identifying the parentheses, which still
798 nigel 75 acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP
799     is provided for extracting an individual captured substring by name. It is also
800     possible to extract the data directly, by first converting the name to a number
801     in order to access the correct pointers in the output vector (described with
802     \fBpcre_exec()\fP below). To do the conversion, you need to use the
803     name-to-number map, which is described by these three values.
804     .P
805 nigel 63 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
806     the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
807 nigel 75 entry; both of these return an \fBint\fP value. The entry size depends on the
808 nigel 63 length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
809 nigel 75 entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
810 nigel 63 are the number of the capturing parenthesis, most significant byte first. The
811     rest of the entry is the corresponding name, zero terminated. The names are in
812     alphabetical order. For example, consider the following pattern (assume
813     PCRE_EXTENDED is set, so white space - including newlines - is ignored):
814 nigel 75 .sp
815     .\" JOIN
816     (?P<date> (?P<year>(\ed\ed)?\ed\ed) -
817     (?P<month>\ed\ed) - (?P<day>\ed\ed) )
818     .sp
819 nigel 63 There are four named subpatterns, so the table has four entries, and each entry
820     in the table is eight bytes long. The table is as follows, with non-printing
821 nigel 75 bytes shows in hexadecimal, and undefined bytes shown as ??:
822     .sp
823 nigel 63 00 01 d a t e 00 ??
824     00 05 d a y 00 ?? ??
825     00 04 m o n t h 00
826     00 02 y e a r 00 ??
827 nigel 75 .sp
828     When writing code to extract data from named subpatterns using the
829     name-to-number map, remember that the length of each entry is likely to be
830     different for each compiled pattern.
831     .sp
832 nigel 63 PCRE_INFO_OPTIONS
833 nigel 75 .sp
834 nigel 63 Return a copy of the options with which the pattern was compiled. The fourth
835 nigel 75 argument should point to an \fBunsigned long int\fP variable. These option bits
836     are those specified in the call to \fBpcre_compile()\fP, modified by any
837 nigel 63 top-level option settings within the pattern itself.
838 nigel 75 .P
839 nigel 63 A pattern is automatically anchored by PCRE if all of its top-level
840     alternatives begin with one of the following:
841 nigel 75 .sp
842 nigel 63 ^ unless PCRE_MULTILINE is set
843 nigel 75 \eA always
844     \eG always
845     .\" JOIN
846 nigel 63 .* if PCRE_DOTALL is set and there are no back
847     references to the subpattern in which .* appears
848 nigel 75 .sp
849 nigel 63 For such patterns, the PCRE_ANCHORED bit is set in the options returned by
850 nigel 75 \fBpcre_fullinfo()\fP.
851     .sp
852 nigel 63 PCRE_INFO_SIZE
853 nigel 75 .sp
854 nigel 63 Return the size of the compiled pattern, that is, the value that was passed as
855 nigel 75 the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
856     place the compiled data. The fourth argument should point to a \fBsize_t\fP
857 nigel 63 variable.
858 nigel 75 .sp
860 nigel 75 .sp
861     Return the size of the data block pointed to by the \fIstudy_data\fP field in
862     a \fBpcre_extra\fP block. That is, it is the value that was passed to
863     \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
864     created by \fBpcre_study()\fP. The fourth argument should point to a
865     \fBsize_t\fP variable.
866     .
867     .
869 nigel 63 .rs
870     .sp
871 nigel 75 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
872     .B *\fIfirstcharptr\fP);
873 nigel 63 .PP
874 nigel 75 The \fBpcre_info()\fP function is now obsolete because its interface is too
875 nigel 63 restrictive to return all the available data about a compiled pattern. New
876 nigel 75 programs should use \fBpcre_fullinfo()\fP instead. The yield of
877     \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
878 nigel 63 following negative numbers:
879 nigel 75 .sp
880     PCRE_ERROR_NULL the argument \fIcode\fP was NULL
881 nigel 63 PCRE_ERROR_BADMAGIC the "magic number" was not found
882 nigel 75 .sp
883     If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
884 nigel 63 pattern was compiled is placed in the integer it points to (see
885     PCRE_INFO_OPTIONS above).
886 nigel 75 .P
887     If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
888 nigel 63 it is used to pass back information about the first character of any matched
889     string (see PCRE_INFO_FIRSTBYTE above).
890 nigel 75 .
891     .
892 nigel 77 .SH "REFERENCE COUNTS"
893 nigel 63 .rs
894     .sp
895 nigel 77 .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
896     .PP
897     The \fBpcre_refcount()\fP function is used to maintain a reference count in the
898     data block that contains a compiled pattern. It is provided for the benefit of
899     applications that operate in an object-oriented manner, where different parts
900     of the application may be using the same compiled pattern, but you want to free
901     the block when they are all done.
902     .P
903     When a pattern is compiled, the reference count field is initialized to zero.
904     It is changed only by calling this function, whose action is to add the
905     \fIadjust\fP value (which may be positive or negative) to it. The yield of the
906     function is the new value. However, the value of the count is constrained to
907     lie between 0 and 65535, inclusive. If the new value is outside these limits,
908     it is forced to the appropriate limit value.
909     .P
910     Except when it is zero, the reference count is not correctly preserved if a
911     pattern is compiled on one host and then transferred to a host whose byte-order
912     is different. (This seems a highly unlikely scenario.)
913     .
914     .
916     .rs
917     .sp
918 nigel 75 .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
919 nigel 63 .ti +5n
920 nigel 75 .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
921 nigel 63 .ti +5n
922 nigel 75 .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
923     .P
924     The function \fBpcre_exec()\fP is called to match a subject string against a
925     compiled pattern, which is passed in the \fIcode\fP argument. If the
926 nigel 63 pattern has been studied, the result of the study should be passed in the
927 nigel 77 \fIextra\fP argument. This function is the main matching facility of the
928     library, and it operates in a Perl-like manner. For specialist use there is
929     also an alternative matching function, which is described
930     .\" HTML <a href="#dfamatch">
931     .\" </a>
932     below
933     .\"
934     in the section about the \fBpcre_dfa_exec()\fP function.
935 nigel 75 .P
936     In most applications, the pattern will have been compiled (and optionally
937     studied) in the same process that calls \fBpcre_exec()\fP. However, it is
938     possible to save compiled patterns and study data, and then use them later
939     in different processes, possibly even on different hosts. For a discussion
940     about this, see the
941     .\" HREF
942     \fBpcreprecompile\fP
943     .\"
944     documentation.
945     .P
946     Here is an example of a simple call to \fBpcre_exec()\fP:
947     .sp
948 nigel 63 int rc;
949     int ovector[30];
950     rc = pcre_exec(
951     re, /* result of pcre_compile() */
952     NULL, /* we didn't study the pattern */
953     "some string", /* the subject string */
954     11, /* the length of the subject string */
955     0, /* start at offset 0 in the subject */
956     0, /* default options */
957 nigel 75 ovector, /* vector of integers for substring information */
958 nigel 77 30); /* number of elements (NOT size in bytes) */
959 nigel 75 .
960     .\" HTML <a name="extradata"></a>
961     .SS "Extra data for \fBpcre_exec()\fR"
962     .rs
963     .sp
964     If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
965     data block. The \fBpcre_study()\fP function returns such a block (when it
966 nigel 63 doesn't return NULL), but you can also create one for yourself, and pass
967 nigel 75 additional information in it. The fields in a \fBpcre_extra\fP block are as
968     follows:
969     .sp
970     unsigned long int \fIflags\fP;
971     void *\fIstudy_data\fP;
972     unsigned long int \fImatch_limit\fP;
973     void *\fIcallout_data\fP;
974     const unsigned char *\fItables\fP;
975     .sp
976     The \fIflags\fP field is a bitmap that specifies which of the other fields
977 nigel 63 are set. The flag bits are:
978 nigel 75 .sp
982 nigel 75 PCRE_EXTRA_TABLES
983     .sp
984     Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
985     \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
986     the appropriate flag bit. You should not set this yourself, but you may add to
987     the block by setting the other fields and their corresponding flag bits.
988     .P
989     The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
990 nigel 63 vast amount of resources when running patterns that are not going to match,
991     but which have a very large number of possibilities in their search trees. The
992 nigel 75 classic example is the use of nested unlimited repeats.
993     .P
994     Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
995     (sometimes recursively). The limit is imposed on the number of times this
996     function is called during a match, which has the effect of limiting the amount
997     of recursion and backtracking that can take place. For patterns that are not
998     anchored, the count starts from zero for each position in the subject string.
999     .P
1000 nigel 63 The default limit for the library can be set when PCRE is built; the default
1001     default is 10 million, which handles all but the most extreme cases. You can
1002 nigel 75 reduce the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP block
1003     in which \fImatch_limit\fP is set to a smaller value, and
1004     PCRE_EXTRA_MATCH_LIMIT is set in the \fIflags\fP field. If the limit is
1005     exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.
1006     .P
1007     The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1008     which is described in the
1009     .\" HREF
1010     \fBpcrecallout\fP
1011     .\"
1012     documentation.
1013     .P
1014     The \fItables\fP field is used to pass a character tables pointer to
1015     \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1016     pattern. A non-NULL value is stored with the compiled pattern only if custom
1017     tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1018     If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1019     internal tables to be used. This facility is helpful when re-using patterns
1020     that have been saved after compiling with an external set of tables, because
1021     the external tables might be at a different address when \fBpcre_exec()\fP is
1022     called. See the
1023     .\" HREF
1024     \fBpcreprecompile\fP
1025     .\"
1026     documentation for a discussion of saving compiled patterns for later use.
1027     .
1028     .SS "Option bits for \fBpcre_exec()\fP"
1029     .rs
1030     .sp
1031     The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1032     zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,
1034     .sp
1036     .sp
1037     The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1038     matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1039     to be anchored by virtue of its contents, it cannot be made unachored at
1040     matching time.
1041     .sp
1042 nigel 63 PCRE_NOTBOL
1043 nigel 75 .sp
1044     This option specifies that first character of the subject string is not the
1045     beginning of a line, so the circumflex metacharacter should not match before
1046     it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1047     never to match. This option affects only the behaviour of the circumflex
1048     metacharacter. It does not affect \eA.
1049     .sp
1050 nigel 63 PCRE_NOTEOL
1051 nigel 75 .sp
1052     This option specifies that the end of the subject string is not the end of a
1053     line, so the dollar metacharacter should not match it nor (except in multiline
1054     mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1055     compile time) causes dollar never to match. This option affects only the
1056     behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1057     .sp
1058 nigel 63 PCRE_NOTEMPTY
1059 nigel 75 .sp
1060 nigel 63 An empty string is not considered to be a valid match if this option is set. If
1061     there are alternatives in the pattern, they are tried. If all the alternatives
1062     match the empty string, the entire match fails. For example, if the pattern
1063 nigel 75 .sp
1064 nigel 63 a?b?
1065 nigel 75 .sp
1066 nigel 63 is applied to a string not beginning with "a" or "b", it matches the empty
1067     string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1068     valid, so PCRE searches further into the string for occurrences of "a" or "b".
1069 nigel 75 .P
1070 nigel 63 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1071 nigel 75 of a pattern match of the empty string within its \fBsplit()\fP function, and
1072 nigel 63 when using the /g modifier. It is possible to emulate Perl's behaviour after
1073     matching a null string by first trying the match again at the same offset with
1074 nigel 75 PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1075     starting offset (see below) and trying an ordinary match again. There is some
1076     code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1077     .sp
1079     .sp
1080     When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1081     string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1082     The value of \fIstartoffset\fP is also checked to ensure that it points to the
1083     start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1084     \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP
1085     contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1086     .P
1087     If you already know that your subject is valid, and you want to skip these
1088     checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1089     calling \fBpcre_exec()\fP. You might want to do this for the second and
1090     subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1091     all the matches in a single subject string. However, you should be sure that
1092     the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1093     PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1094     subject, or a value of \fIstartoffset\fP that does not point to the start of a
1095     UTF-8 character, is undefined. Your program may crash.
1096     .sp
1098     .sp
1099     This option turns on the partial matching feature. If the subject string fails
1100     to match the pattern, but at some point during the matching process the end of
1101     the subject was reached (that is, the subject partially matches the pattern and
1102     the failure to match occurred only because there were not enough subject
1103     characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1104     PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1105     may appear in the pattern. These are discussed in the
1106     .\" HREF
1107     \fBpcrepartial\fP
1108     .\"
1109     documentation.
1110     .
1111     .SS "The string to be matched by \fBpcre_exec()\fP"
1112     .rs
1113     .sp
1114     The subject string is passed to \fBpcre_exec()\fP as a pointer in
1115     \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
1116     \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
1117     UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1118     bytes. When the starting offset is zero, the search for a match starts at the
1119     beginning of the subject, and this is by far the most common case.
1120     .P
1121 nigel 63 A non-zero starting offset is useful when searching for another match in the
1122 nigel 75 same subject by calling \fBpcre_exec()\fP again after a previous success.
1123     Setting \fIstartoffset\fP differs from just passing over a shortened string and
1124 nigel 63 setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1125     lookbehind. For example, consider the pattern
1126 nigel 75 .sp
1127     \eBiss\eB
1128     .sp
1129     which finds occurrences of "iss" in the middle of words. (\eB matches only if
1130 nigel 63 the current position in the subject is not a word boundary.) When applied to
1131 nigel 75 the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1132     occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1133     subject, namely "issipi", it does not match, because \eB is always false at the
1134 nigel 63 start of the subject, which is deemed to be a word boundary. However, if
1135 nigel 75 \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1136 nigel 63 set to 4, it finds the second occurrence of "iss" because it is able to look
1137     behind the starting point to discover that it is preceded by a letter.
1138 nigel 75 .P
1139 nigel 63 If a non-zero starting offset is passed when the pattern is anchored, one
1140 nigel 75 attempt to match at the given offset is made. This can only succeed if the
1141 nigel 63 pattern does not require the match to be at the start of the subject.
1142 nigel 75 .
1143     .SS "How \fBpcre_exec()\fP returns captured substrings"
1144     .rs
1145     .sp
1146 nigel 63 In general, a pattern matches a certain portion of the subject, and in
1147     addition, further substrings from the subject may be picked out by parts of the
1148     pattern. Following the usage in Jeffrey Friedl's book, this is called
1149     "capturing" in what follows, and the phrase "capturing subpattern" is used for
1150     a fragment of a pattern that picks out a substring. PCRE supports several other
1151     kinds of parenthesized subpattern that do not cause substrings to be captured.
1152 nigel 75 .P
1153 nigel 63 Captured substrings are returned to the caller via a vector of integer offsets
1154 nigel 75 whose address is passed in \fIovector\fP. The number of elements in the vector
1155     is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:
1156     this argument is NOT the size of \fIovector\fP in bytes.
1157     .P
1158     The first two-thirds of the vector is used to pass back captured substrings,
1159     each substring using a pair of integers. The remaining third of the vector is
1160     used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1161     and is not available for passing back information. The length passed in
1162     \fIovecsize\fP should always be a multiple of three. If it is not, it is
1163     rounded down.
1164     .P
1165     When a match is successful, information about captured substrings is returned
1166     in pairs of integers, starting at the beginning of \fIovector\fP, and
1167 nigel 63 continuing up to two-thirds of its length at the most. The first element of a
1168     pair is set to the offset of the first character in a substring, and the second
1169     is set to the offset of the first character after the end of a substring. The
1170 nigel 75 first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
1171 nigel 63 subject string matched by the entire pattern. The next pair is used for the
1172 nigel 75 first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
1173 nigel 63 is the number of pairs that have been set. If there are no capturing
1174     subpatterns, the return value from a successful match is 1, indicating that
1175     just the first pair of offsets has been set.
1176 nigel 75 .P
1177 nigel 63 Some convenience functions are provided for extracting the captured substrings
1178     as separate strings. These are described in the following section.
1179 nigel 75 .P
1180     It is possible for an capturing subpattern number \fIn+1\fP to match some
1181     part of the subject when subpattern \fIn\fP has not been used at all. For
1182 nigel 63 example, if the string "abc" is matched against the pattern (a|(z))(bc)
1183     subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset
1184     values corresponding to the unused subpattern are set to -1.
1185 nigel 75 .P
1186 nigel 63 If a capturing subpattern is matched repeatedly, it is the last portion of the
1187 nigel 75 string that it matched that is returned.
1188     .P
1189     If the vector is too small to hold all the captured substring offsets, it is
1190     used as far as possible (up to two-thirds of its length), and the function
1191     returns a value of zero. In particular, if the substring offsets are not of
1192     interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1193     \fIovecsize\fP as zero. However, if the pattern contains back references and
1194     the \fIovector\fP is not big enough to remember the related substrings, PCRE
1195     has to get additional memory for use during matching. Thus it is usually
1196     advisable to supply an \fIovector\fP.
1197     .P
1198     Note that \fBpcre_info()\fP can be used to find out how many capturing
1199 nigel 63 subpatterns there are in a compiled pattern. The smallest size for
1200 nigel 75 \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1201     the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1202     .
1203 nigel 77 .\" HTML <a name="errorlist"></a>
1204 nigel 75 .SS "Return values from \fBpcre_exec()\fP"
1205     .rs
1206     .sp
1207     If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1208 nigel 63 defined in the header file:
1209 nigel 75 .sp
1210 nigel 63 PCRE_ERROR_NOMATCH (-1)
1211 nigel 75 .sp
1212 nigel 63 The subject string did not match the pattern.
1213 nigel 75 .sp
1214 nigel 63 PCRE_ERROR_NULL (-2)
1215 nigel 75 .sp
1216     Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1217     NULL and \fIovecsize\fP was not zero.
1218     .sp
1219 nigel 63 PCRE_ERROR_BADOPTION (-3)
1220 nigel 75 .sp
1221     An unrecognized bit was set in the \fIoptions\fP argument.
1222     .sp
1223 nigel 63 PCRE_ERROR_BADMAGIC (-4)
1224 nigel 75 .sp
1225 nigel 63 PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1226 nigel 75 the case when it is passed a junk pointer and to detect when a pattern that was
1227     compiled in an environment of one endianness is run in an environment with the
1228     other endianness. This is the error that PCRE gives when the magic number is
1229     not present.
1230     .sp
1231 nigel 63 PCRE_ERROR_UNKNOWN_NODE (-5)
1232 nigel 75 .sp
1233 nigel 63 While running the pattern match, an unknown item was encountered in the
1234     compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1235     of the compiled pattern.
1236 nigel 75 .sp
1237 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1238 nigel 75 .sp
1239     If a pattern contains back references, but the \fIovector\fP that is passed to
1240     \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1241 nigel 63 gets a block of memory at the start of matching to use for this purpose. If the
1242 nigel 75 call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1243     automatically freed at the end of matching.
1244     .sp
1245 nigel 63 PCRE_ERROR_NOSUBSTRING (-7)
1246 nigel 75 .sp
1247     This error is used by the \fBpcre_copy_substring()\fP,
1248     \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1249     below). It is never returned by \fBpcre_exec()\fP.
1250     .sp
1251 nigel 63 PCRE_ERROR_MATCHLIMIT (-8)
1252 nigel 75 .sp
1253     The recursion and backtracking limit, as specified by the \fImatch_limit\fP
1254     field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1255 nigel 63 description above.
1256 nigel 75 .sp
1257 nigel 63 PCRE_ERROR_CALLOUT (-9)
1258 nigel 75 .sp
1259     This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1260 nigel 63 use by callout functions that want to yield a distinctive error code. See the
1261 nigel 75 .\" HREF
1262     \fBpcrecallout\fP
1263     .\"
1264     documentation for details.
1265     .sp
1266 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1267 nigel 75 .sp
1268 nigel 71 A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1269 nigel 75 .sp
1270 nigel 73 PCRE_ERROR_BADUTF8_OFFSET (-11)
1271 nigel 75 .sp
1272 nigel 73 The UTF-8 byte sequence that was passed as a subject was valid, but the value
1273 nigel 75 of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1274     .sp
1275 nigel 77 PCRE_ERROR_PARTIAL (-12)
1276 nigel 75 .sp
1277     The subject string did not match, but it did match partially. See the
1278     .\" HREF
1279     \fBpcrepartial\fP
1280     .\"
1281     documentation for details of partial matching.
1282     .sp
1283 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1284 nigel 75 .sp
1285     The PCRE_PARTIAL option was used with a compiled pattern containing items that
1286     are not supported for partial matching. See the
1287     .\" HREF
1288     \fBpcrepartial\fP
1289     .\"
1290     documentation for details of partial matching.
1291     .sp
1292 nigel 77 PCRE_ERROR_INTERNAL (-14)
1293 nigel 75 .sp
1294     An unexpected internal error has occurred. This error could be caused by a bug
1295     in PCRE or by overwriting of the compiled pattern.
1296     .sp
1297 nigel 77 PCRE_ERROR_BADCOUNT (-15)
1298 nigel 75 .sp
1299     This error is given if the value of the \fIovecsize\fP argument is negative.
1300     .
1301     .
1303 nigel 63 .rs
1304     .sp
1305 nigel 75 .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1306 nigel 63 .ti +5n
1307 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1308 nigel 63 .ti +5n
1309 nigel 75 .B int \fIbuffersize\fP);
1310 nigel 63 .PP
1311     .br
1312 nigel 75 .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1313 nigel 63 .ti +5n
1314 nigel 75 .B int \fIstringcount\fP, int \fIstringnumber\fP,
1315 nigel 63 .ti +5n
1316 nigel 75 .B const char **\fIstringptr\fP);
1317 nigel 63 .PP
1318     .br
1319 nigel 75 .B int pcre_get_substring_list(const char *\fIsubject\fP,
1320 nigel 63 .ti +5n
1321 nigel 75 .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1322 nigel 63 .PP
1323     Captured substrings can be accessed directly by using the offsets returned by
1324 nigel 75 \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1325     \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1326     \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1327 nigel 63 as new, separate, zero-terminated strings. These functions identify substrings
1328     by number. The next section describes functions for extracting named
1329     substrings. A substring that contains a binary zero is correctly extracted and
1330     has a further zero added on the end, but the result is not, of course,
1331     a C string.
1332 nigel 75 .P
1333 nigel 63 The first three arguments are the same for all three of these functions:
1334 nigel 75 \fIsubject\fP is the subject string that has just been successfully matched,
1335     \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1336     \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1337 nigel 63 captured by the match, including the substring that matched the entire regular
1338 nigel 75 expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1339     than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1340     space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1341     number of elements in the vector divided by three.
1342     .P
1343     The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1344     extract a single substring, whose number is given as \fIstringnumber\fP. A
1345     value of zero extracts the substring that matched the entire pattern, whereas
1346     higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1347     the string is placed in \fIbuffer\fP, whose length is given by
1348     \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1349     obtained via \fBpcre_malloc\fP, and its address is returned via
1350     \fIstringptr\fP. The yield of the function is the length of the string, not
1351 nigel 63 including the terminating zero, or one of
1352 nigel 75 .sp
1353 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1354 nigel 75 .sp
1355     The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1356     memory failed for \fBpcre_get_substring()\fP.
1357     .sp
1358 nigel 63 PCRE_ERROR_NOSUBSTRING (-7)
1359 nigel 75 .sp
1360     There is no substring whose number is \fIstringnumber\fP.
1361     .P
1362     The \fBpcre_get_substring_list()\fP function extracts all available substrings
1363 nigel 63 and builds a list of pointers to them. All this is done in a single block of
1364 nigel 75 memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1365     is returned via \fIlistptr\fP, which is also the start of the list of string
1366 nigel 63 pointers. The end of the list is marked by a NULL pointer. The yield of the
1367     function is zero if all went well, or
1368 nigel 75 .sp
1369 nigel 63 PCRE_ERROR_NOMEMORY (-6)
1370 nigel 75 .sp
1371 nigel 63 if the attempt to get the memory block failed.
1372 nigel 75 .P
1373 nigel 63 When any of these functions encounter a substring that is unset, which can
1374 nigel 75 happen when capturing subpattern number \fIn+1\fP matches some part of the
1375     subject, but subpattern \fIn\fP has not been used at all, they return an empty
1376 nigel 63 string. This can be distinguished from a genuine zero-length substring by
1377 nigel 75 inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1378 nigel 63 substrings.
1379 nigel 75 .P
1380     The two convenience functions \fBpcre_free_substring()\fP and
1381     \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1382     a previous call of \fBpcre_get_substring()\fP or
1383     \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1384     the function pointed to by \fBpcre_free\fP, which of course could be called
1385 nigel 63 directly from a C program. However, PCRE is used in some situations where it is
1386     linked via a special interface to another programming language which cannot use
1387 nigel 75 \fBpcre_free\fP directly; it is for these cases that the functions are
1388 nigel 63 provided.
1389 nigel 75 .
1390     .
1392 nigel 63 .rs
1393     .sp
1394 nigel 75 .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1395 nigel 63 .ti +5n
1396 nigel 75 .B const char *\fIname\fP);
1397 nigel 63 .PP
1398     .br
1399 nigel 75 .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1400 nigel 63 .ti +5n
1401 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
1402     .ti +5n
1403     .B int \fIstringcount\fP, const char *\fIstringname\fP,
1404     .ti +5n
1405     .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1406 nigel 63 .PP
1407     .br
1408 nigel 75 .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1409 nigel 63 .ti +5n
1410 nigel 75 .B const char *\fIsubject\fP, int *\fIovector\fP,
1411 nigel 63 .ti +5n
1412 nigel 75 .B int \fIstringcount\fP, const char *\fIstringname\fP,
1413 nigel 63 .ti +5n
1414 nigel 75 .B const char **\fIstringptr\fP);
1415 nigel 63 .PP
1416 nigel 75 To extract a substring by name, you first have to find associated number.
1417     For example, for this pattern
1418     .sp
1419 nigel 79 (a+)b(?P<xxx>\ed+)...
1420 nigel 75 .sp
1421     the number of the subpattern called "xxx" is 2. You can find the number from
1422     the name by calling \fBpcre_get_stringnumber()\fP. The first argument is the
1423     compiled pattern, and the second is the name. The yield of the function is the
1424     subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1425     that name.
1426     .P
1427     Given the number, you can extract the substring directly, or use one of the
1428     functions described in the previous section. For convenience, there are also
1429     two functions that do the whole job.
1430     .P
1431     Most of the arguments of \fIpcre_copy_named_substring()\fP and
1432     \fIpcre_get_named_substring()\fP are the same as those for the similarly named
1433     functions that extract by number. As these are described in the previous
1434     section, they are not re-described here. There are just two differences:
1435     .P
1436 nigel 63 First, instead of a substring number, a substring name is given. Second, there
1437     is an extra argument, given at the start, which is a pointer to the compiled
1438     pattern. This is needed in order to gain access to the name-to-number
1439     translation table.
1440 nigel 75 .P
1441     These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1442     then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as
1443 nigel 63 appropriate.
1444 nigel 77 .
1445     .
1447     .rs
1448     .sp
1449     The traditional matching function uses a similar algorithm to Perl, which stops
1450     when it finds the first match, starting at a given point in the subject. If you
1451     want to find all possible matches, or the longest possible match, consider
1452     using the alternative matching function (see below) instead. If you cannot use
1453     the alternative function, but still need to find all possible matches, you
1454     can kludge it up by making use of the callout facility, which is described in
1455     the
1456     .\" HREF
1457     \fBpcrecallout\fP
1458     .\"
1459     documentation.
1460 nigel 75 .P
1461 nigel 77 What you have to do is to insert a callout right at the end of the pattern.
1462     When your callout function is called, extract and save the current matched
1463     substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1464     other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1465     will yield PCRE_ERROR_NOMATCH.
1466     .
1467     .
1468     .\" HTML <a name="dfamatch"></a>
1470     .rs
1471     .sp
1472     .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1473     .ti +5n
1474     .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1475     .ti +5n
1476     .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1477     .ti +5n
1478     .B int *\fIworkspace\fP, int \fIwscount\fP);
1479     .P
1480     The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1481     a compiled pattern, using a "DFA" matching algorithm. This has different
1482     characteristics to the normal algorithm, and is not compatible with Perl. Some
1483     of the features of PCRE patterns are not supported. Nevertheless, there are
1484     times when this kind of matching can be useful. For a discussion of the two
1485     matching algorithms, see the
1486     .\" HREF
1487     \fBpcrematching\fP
1488     .\"
1489     documentation.
1490     .P
1491     The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1492     \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1493     different way, and this is described below. The other common arguments are used
1494     in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1495     here.
1496     .P
1497     The two additional arguments provide workspace for the function. The workspace
1498     vector should contain at least 20 elements. It is used for keeping track of
1499     multiple paths through the pattern tree. More workspace will be needed for
1500     patterns and subjects where there are a lot of possible matches.
1501     .P
1502     Here is an example of a simple call to \fBpcre_exec()\fP:
1503     .sp
1504     int rc;
1505     int ovector[10];
1506     int wspace[20];
1507     rc = pcre_exec(
1508     re, /* result of pcre_compile() */
1509     NULL, /* we didn't study the pattern */
1510     "some string", /* the subject string */
1511     11, /* the length of the subject string */
1512     0, /* start at offset 0 in the subject */
1513     0, /* default options */
1514     ovector, /* vector of integers for substring information */
1515     10, /* number of elements (NOT size in bytes) */
1516     wspace, /* working space vector */
1517     20); /* number of elements (NOT size in bytes) */
1518     .
1519     .SS "Option bits for \fBpcre_dfa_exec()\fP"
1520     .rs
1521     .sp
1522     The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1523     zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,
1525     PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1526     the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1527     .sp
1529     .sp
1530     This has the same general effect as it does for \fBpcre_exec()\fP, but the
1531     details are slightly different. When PCRE_PARTIAL is set for
1532     \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1533     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1534     complete matches, but there is still at least one matching possibility. The
1535     portion of the string that provided the partial match is set as the first
1536     matching string.
1537     .sp
1539     .sp
1540     Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1541     soon as it has found one match. Because of the way the DFA algorithm works,
1542     this is necessarily the shortest possible match at the first possible matching
1543     point in the subject string.
1544     .sp
1546     .sp
1547     When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1548     a partial match, it is possible to call it again, with additional subject
1549     characters, and have it continue with the same match. The PCRE_DFA_RESTART
1550     option requests this action; when it is set, the \fIworkspace\fP and
1551     \fIwscount\fP options must reference the same vector as before because data
1552     about the match so far is left in them after a partial match. There is more
1553     discussion of this facility in the
1554     .\" HREF
1555     \fBpcrepartial\fP
1556     .\"
1557     documentation.
1558     .
1559     .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1560     .rs
1561     .sp
1562     When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1563     substring in the subject. Note, however, that all the matches from one run of
1564     the function start at the same point in the subject. The shorter matches are
1565     all initial substrings of the longer matches. For example, if the pattern
1566     .sp
1567     <.*>
1568     .sp
1569     is matched against the string
1570     .sp
1571     This is <something> <something else> <something further> no more
1572     .sp
1573     the three matched strings are
1574     .sp
1575     <something>
1576     <something> <something else>
1577     <something> <something else> <something further>
1578     .sp
1579     On success, the yield of the function is a number greater than zero, which is
1580     the number of matched substrings. The substrings themselves are returned in
1581     \fIovector\fP. Each string uses two elements; the first is the offset to the
1582     start, and the second is the offset to the end. All the strings have the same
1583     start offset. (Space could have been saved by giving this only once, but it was
1584     decided to retain some compatibility with the way \fBpcre_exec()\fP returns
1585     data, even though the meaning of the strings is different.)
1586     .P
1587     The strings are returned in reverse order of length; that is, the longest
1588     matching string is given first. If there were too many matches to fit into
1589     \fIovector\fP, the yield of the function is zero, and the vector is filled with
1590     the longest matches.
1591     .
1592     .SS "Error returns from \fBpcre_dfa_exec()\fP"
1593     .rs
1594     .sp
1595     The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1596     Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1597     described
1598     .\" HTML <a href="#errorlist">
1599     .\" </a>
1600     above.
1601     .\"
1602     There are in addition the following errors that are specific to
1603     \fBpcre_dfa_exec()\fP:
1604     .sp
1605     PCRE_ERROR_DFA_UITEM (-16)
1606     .sp
1607     This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1608     that it does not support, for instance, the use of \eC or a back reference.
1609     .sp
1610     PCRE_ERROR_DFA_UCOND (-17)
1611     .sp
1612     This return is given if \fBpcre_dfa_exec()\fP encounters a condition item in a
1613     pattern that uses a back reference for the condition. This is not supported.
1614     .sp
1616     .sp
1617     This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1618     block that contains a setting of the \fImatch_limit\fP field. This is not
1619     supported (it is meaningless).
1620     .sp
1621     PCRE_ERROR_DFA_WSSIZE (-19)
1622     .sp
1623     This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1624     \fIworkspace\fP vector.
1625     .sp
1627     .sp
1628     When a recursive subpattern is processed, the matching function calls itself
1629     recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1630     error is given if the output vector is not large enough. This should be
1631     extremely rare, as a vector of size 1000 is used.
1632     .P
1633 nigel 63 .in 0
1634 nigel 77 Last updated: 16 May 2005
1635 nigel 63 .br
1636 nigel 77 Copyright (c) 1997-2005 University of Cambridge.

ViewVC Help
Powered by ViewVC 1.1.12