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

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

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

revision 65 by nigel, Sat Feb 24 21:40:08 2007 UTC revision 426 by ph10, Wed Aug 26 15:38:32 2009 UTC
# Line 1  Line 1 
1  .TH PCRE 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH SYNOPSIS OF PCRE API  .SH "PCRE NATIVE API"
5  .rs  .rs
6  .sp  .sp
7  .B #include <pcre.h>  .B #include <pcre.h>
8  .PP  .PP
9  .SM  .SM
10  .br  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
 .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  
11  .ti +5n  .ti +5n
12  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13  .ti +5n  .ti +5n
14  .B const unsigned char *\fItableptr\fR);  .B const unsigned char *\fItableptr\fP);
15  .PP  .PP
16  .br  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
 .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  
17  .ti +5n  .ti +5n
18  .B const char **\fIerrptr\fR);  .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    .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25    .ti +5n
26    .B const char **\fIerrptr\fP);
27    .PP
28    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
29    .ti +5n
30    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31    .ti +5n
32    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33  .PP  .PP
34  .br  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .ti +5n
36    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37  .ti +5n  .ti +5n
38  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39  .ti +5n  .ti +5n
40  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int *\fIworkspace\fP, int \fIwscount\fP);
41  .PP  .PP
42  .br  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
 .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  
43  .ti +5n  .ti +5n
44  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
45  .ti +5n  .ti +5n
46  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
47  .ti +5n  .ti +5n
48  .B char *\fIbuffer\fR, int \fIbuffersize\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49  .PP  .PP
50  .br  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
 .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  
51  .ti +5n  .ti +5n
52  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53  .ti +5n  .ti +5n
54  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
55  .PP  .PP
56  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
57  .B int pcre_get_named_substring(const pcre *\fIcode\fR,  .ti +5n
58    .B const char *\fIsubject\fP, int *\fIovector\fP,
59  .ti +5n  .ti +5n
60  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
61  .ti +5n  .ti +5n
62  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char **\fIstringptr\fP);
63    .PP
64    .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65  .ti +5n  .ti +5n
66  .B const char **\fIstringptr\fR);  .B const char *\fIname\fP);
67  .PP  .PP
68  .br  .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
 .B int pcre_get_stringnumber(const pcre *\fIcode\fR,  
69  .ti +5n  .ti +5n
70  .B const char *\fIname\fR);  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71  .PP  .PP
72  .br  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
 .B int pcre_get_substring(const char *\fIsubject\fR, int *\fIovector\fR,  
73  .ti +5n  .ti +5n
74  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
75  .ti +5n  .ti +5n
76  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
77  .PP  .PP
78  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
79  .ti +5n  .ti +5n
80  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81  .PP  .PP
82  .br  .B void pcre_free_substring(const char *\fIstringptr\fP);
 .B void pcre_free_substring(const char *\fIstringptr\fR);  
83  .PP  .PP
84  .br  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
 .B void pcre_free_substring_list(const char **\fIstringptr\fR);  
85  .PP  .PP
 .br  
86  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
87  .PP  .PP
88  .br  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
 .B int pcre_fullinfo(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  
89  .ti +5n  .ti +5n
90  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
91    .PP
92    .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93    .B *\fIfirstcharptr\fP);
94  .PP  .PP
95  .br  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
 .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  
 .B *\fIfirstcharptr\fR);  
96  .PP  .PP
97  .br  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
 .B int pcre_config(int \fIwhat\fR, void *\fIwhere\fR);  
98  .PP  .PP
 .br  
99  .B char *pcre_version(void);  .B char *pcre_version(void);
100  .PP  .PP
 .br  
101  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
102  .PP  .PP
 .br  
103  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
104  .PP  .PP
105  .br  .B void *(*pcre_stack_malloc)(size_t);
106    .PP
107    .B void (*pcre_stack_free)(void *);
108    .PP
109  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
110    .
111  .SH PCRE API  .
112    .SH "PCRE API OVERVIEW"
113  .rs  .rs
114  .sp  .sp
115  PCRE has its own native API, which is described in this document. There is also  PCRE has its own native API, which is described in this document. There are
116  a set of wrapper functions that correspond to the POSIX regular expression API.  also some wrapper functions that correspond to the POSIX regular expression
117  These are described in the \fBpcreposix\fR documentation.  API. These are described in the
118    .\" HREF
119  The native API function prototypes are defined in the header file \fBpcre.h\fR,  \fBpcreposix\fP
120  and on Unix systems the library itself is called \fBlibpcre.a\fR, so can be  .\"
121  accessed by adding \fB-lpcre\fR to the command for linking an application which  documentation. Both of these APIs define a set of C function calls. A C++
122  calls it. The header file defines the macros PCRE_MAJOR and PCRE_MINOR to  wrapper is distributed with PCRE. It is documented in the
123  contain the major and minor release numbers for the library. Applications can  .\" HREF
124  use these to include support for different releases.  \fBpcrecpp\fP
125    .\"
126  The functions \fBpcre_compile()\fR, \fBpcre_study()\fR, and \fBpcre_exec()\fR  page.
127  are used for compiling and matching regular expressions. A sample program that  .P
128  demonstrates the simplest way of using them is given in the file  The native API C function prototypes are defined in the header file
129  \fIpcredemo.c\fR. The \fBpcresample\fR documentation describes how to run it.  \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  There are convenience functions for extracting captured substrings from a  an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132  matched subject string. They are:  and PCRE_MINOR to contain the major and minor release numbers for the library.
133    Applications can use these to include support for different releases of PCRE.
134    \fBpcre_copy_substring()\fR  .P
135    \fBpcre_copy_named_substring()\fR  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136    \fBpcre_get_substring()\fR  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137    \fBpcre_get_named_substring()\fR  in a Perl-compatible manner. A sample program that demonstrates the simplest
138    \fBpcre_get_substring_list()\fR  way of using them is provided in the file called \fIpcredemo.c\fP in the source
139    distribution. The
140  \fBpcre_free_substring()\fR and \fBpcre_free_substring_list()\fR are also  .\" HREF
141    \fBpcresample\fP
142    .\"
143    documentation describes how to compile and run it.
144    .P
145    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    matching. The alternative algorithm finds all possible matches (at a given
148    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    .\" HREF
152    \fBpcrematching\fP
153    .\"
154    documentation.
155    .P
156    In addition to the main compiling and matching functions, there are convenience
157    functions for extracting captured substrings from a subject string that is
158    matched by \fBpcre_exec()\fP. They are:
159    .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      \fBpcre_get_stringtable_entries()\fP
167    .sp
168    \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
169  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
170    .P
171  The function \fBpcre_maketables()\fR is used (optionally) to build a set of  The function \fBpcre_maketables()\fP is used to build a set of character tables
172  character tables in the current locale for passing to \fBpcre_compile()\fR.  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  The function \fBpcre_fullinfo()\fR is used to find out information about a  specialist use. Most commonly, no special tables are passed, in which case
175  compiled pattern; \fBpcre_info()\fR is an obsolete version which returns only  internal tables that are generated when PCRE is built are used.
176    .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  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
180  The function \fBpcre_version()\fR returns a pointer to a string containing the  The function \fBpcre_version()\fP returns a pointer to a string containing the
181  version of PCRE and its date of release.  version of PCRE and its date of release.
182    .P
183  The global variables \fBpcre_malloc\fR and \fBpcre_free\fR initially contain  The function \fBpcre_refcount()\fP maintains a reference count in a data block
184  the entry points of the standard \fBmalloc()\fR and \fBfree()\fR functions  containing a compiled pattern. This is provided for the benefit of
185    object-oriented applications.
186    .P
187    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  respectively. PCRE calls the memory management functions via these variables,  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  so a calling program can replace them if it wishes to intercept the calls. This
191  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
192    .P
193  The global variable \fBpcre_callout\fR initially contains NULL. It can be set  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
194    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    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    .P
212    The global variable \fBpcre_callout\fP initially contains NULL. It can be set
213  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
214  points during a matching operation. Details are given in the \fBpcrecallout\fR  points during a matching operation. Details are given in the
215    .\" HREF
216    \fBpcrecallout\fP
217    .\"
218  documentation.  documentation.
219    .
220    .
221    .\" HTML <a name="newlines"></a>
222    .SH NEWLINES
223    .rs
224    .sp
225    PCRE supports five different conventions for indicating line breaks in
226    strings: a single CR (carriage return) character, a single LF (linefeed)
227    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    .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    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    In the PCRE documentation the word "newline" is used to mean "the character or
248    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    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    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    .
263    .
264  .SH MULTITHREADING  .SH MULTITHREADING
265  .rs  .rs
266  .sp  .sp
267  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
268  proviso that the memory management functions pointed to by \fBpcre_malloc\fR  proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
269  and \fBpcre_free\fR, and the callout function pointed to by \fBpcre_callout\fR,  \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
270  are shared by all threads.  callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
271    .P
272  The compiled form of a regular expression is not altered during matching, so  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.  the same compiled pattern can safely be used by several threads at once.
274    .
275  .SH CHECKING BUILD-TIME OPTIONS  .
276    .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
277  .rs  .rs
278  .sp  .sp
279  .B int pcre_config(int \fIwhat\fR, void *\fIwhere\fR);  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    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    .
289    .
290    .SH "CHECKING BUILD-TIME OPTIONS"
291    .rs
292    .sp
293    .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
294  .PP  .PP
295  The function \fBpcre_config()\fR makes it possible for a PCRE client to  The function \fBpcre_config()\fP makes it possible for a PCRE client to
296  discover which optional features have been compiled into the PCRE library. The  discover which optional features have been compiled into the PCRE library. The
297  .\" HREF  .\" HREF
298  \fBpcrebuild\fR  \fBpcrebuild\fP
299  .\"  .\"
300  documentation has more details about these optional features.  documentation has more details about these optional features.
301    .P
302  The first argument for \fBpcre_config()\fR is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
303  information is required; the second argument is a pointer to a variable into  information is required; the second argument is a pointer to a variable into
304  which the information is placed. The following information is available:  which the information is placed. The following information is available:
305    .sp
306    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
307    .sp
308  The output is an integer that is set to one if UTF-8 support is available;  The output is an integer that is set to one if UTF-8 support is available;
309  otherwise it is set to zero.  otherwise it is set to zero.
310    .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    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
317    .sp
318  The output is an integer that is set to the value of the code that is used for  The output is an integer whose value specifies the default character sequence
319  the newline character. It is either linefeed (10) or carriage return (13), and  that is recognized as meaning "newline". The four values that are supported
320  should normally be the standard character for your operating system.  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
321    Though they are derived from ASCII, the same values are returned in EBCDIC
322    environments. The default should normally correspond to the standard sequence
323    for your operating system.
324    .sp
325      PCRE_CONFIG_BSR
326    .sp
327    The output is an integer whose value indicates what character sequences the \eR
328    escape sequence matches by default. A value of 0 means that \eR matches any
329    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
330    or CRLF. The default can be overridden when a pattern is compiled or matched.
331    .sp
332    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
333    .sp
334  The output is an integer that contains the number of bytes used for internal  The output is an integer that contains the number of bytes used for internal
335  linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values  linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values
336  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
337  matching. The default value of 2 is sufficient for all but the most massive  matching. The default value of 2 is sufficient for all but the most massive
338  patterns, since it allows the compiled pattern to be up to 64K in size.  patterns, since it allows the compiled pattern to be up to 64K in size.
339    .sp
340    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
341    .sp
342  The output is an integer that contains the threshold above which the POSIX  The output is an integer that contains the threshold above which the POSIX
343  interface uses \fBmalloc()\fR for output vectors. Further details are given in  interface uses \fBmalloc()\fP for output vectors. Further details are given in
344  the \fBpcreposix\fR documentation.  the
345    .\" HREF
346    \fBpcreposix\fP
347    .\"
348    documentation.
349    .sp
350    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
351    .sp
352  The output is an integer that gives the default limit for the number of  The output is a long integer that gives the default limit for the number of
353  internal matching function calls in a \fBpcre_exec()\fR execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
354  details are given with \fBpcre_exec()\fR below.  details are given with \fBpcre_exec()\fP below.
355    .sp
356  .SH COMPILING A PATTERN    PCRE_CONFIG_MATCH_LIMIT_RECURSION
357  .rs  .sp
358  .sp  The output is a long integer that gives the default limit for the depth of
359  .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  recursion when calling the internal matching function in a \fBpcre_exec()\fP
360  .ti +5n  execution. Further details are given with \fBpcre_exec()\fP below.
361  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .sp
362  .ti +5n    PCRE_CONFIG_STACKRECURSE
363  .B const unsigned char *\fItableptr\fR);  .sp
364  .PP  The output is an integer that is set to one if internal recursion when running
365    \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
366  The function \fBpcre_compile()\fR is called to compile a pattern into an  to remember their state. This is the usual way that PCRE is compiled. The
367  internal form. The pattern is a C string terminated by a binary zero, and  output is zero if PCRE was compiled to use blocks of data on the heap instead
368  is passed in the argument \fIpattern\fR. A pointer to a single block of memory  of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
369  that is obtained via \fBpcre_malloc\fR is returned. This contains the compiled  \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
370  code and related data. The \fBpcre\fR type is defined for the returned block;  avoiding the use of the stack.
371  this is a typedef for a structure whose contents are not externally defined. It  .
372  is up to the caller to free the memory when it is no longer required.  .
373    .SH "COMPILING A PATTERN"
374    .rs
375    .sp
376    .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
377    .ti +5n
378    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
379    .ti +5n
380    .B const unsigned char *\fItableptr\fP);
381    .sp
382    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
383    .ti +5n
384    .B int *\fIerrorcodeptr\fP,
385    .ti +5n
386    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
387    .ti +5n
388    .B const unsigned char *\fItableptr\fP);
389    .P
390    Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
391    called to compile a pattern into an internal form. The only difference between
392    the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
393    \fIerrorcodeptr\fP, via which a numerical error code can be returned.
394    .P
395    The pattern is a C string terminated by a binary zero, and is passed in the
396    \fIpattern\fP argument. A pointer to a single block of memory that is obtained
397    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
398    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
399    for a structure whose contents are not externally defined. It is up to the
400    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
401    .P
402  Although the compiled code of a PCRE regex is relocatable, that is, it does not  Although the compiled code of a PCRE regex is relocatable, that is, it does not
403  depend on memory location, the complete \fBpcre\fR data block is not  depend on memory location, the complete \fBpcre\fP data block is not
404  fully relocatable, because it contains a copy of the \fItableptr\fR argument,  fully relocatable, because it may contain a copy of the \fItableptr\fP
405  which is an address (see below).  argument, which is an address (see below).
406    .P
407  The \fIoptions\fR argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
408  compilation. It should be zero if no options are required. Some of the options,  compilation. It should be zero if no options are required. The available
409  in particular, those that are compatible with Perl, can also be set and unset  options are described below. Some of them (in particular, those that are
410  from within the pattern (see the detailed description of regular expressions  compatible with Perl, but also some others) can also be set and unset from
411  in the \fBpcrepattern\fR documentation). For these options, the contents of the  within the pattern (see the detailed description in the
412  \fIoptions\fR argument specifies their initial settings at the start of  .\" HREF
413  compilation and execution. The PCRE_ANCHORED option can be set at the time of  \fBpcrepattern\fP
414  matching as well as at compile time.  .\"
415    documentation). For those options that can be different in different parts of
416  If \fIerrptr\fR is NULL, \fBpcre_compile()\fR returns NULL immediately.  the pattern, the contents of the \fIoptions\fP argument specifies their initial
417  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fR returns  settings at the start of compilation and execution. The PCRE_ANCHORED and
418  NULL, and sets the variable pointed to by \fIerrptr\fR to point to a textual  PCRE_NEWLINE_\fIxxx\fP options can be set at the time of matching as well as at
419  error message. The offset from the start of the pattern to the character where  compile time.
420  the error was discovered is placed in the variable pointed to by  .P
421  \fIerroffset\fR, which must not be NULL. If it is, an immediate error is given.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
422    Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
423  If the final argument, \fItableptr\fR, is NULL, PCRE uses a default set of  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
424  character tables which are built when it is compiled, using the default C  error message. This is a static string that is part of the library. You must
425  locale. Otherwise, \fItableptr\fR must be the result of a call to  not try to free it. The offset from the start of the pattern to the character
426  \fBpcre_maketables()\fR. See the section on locale support below.  where the error was discovered is placed in the variable pointed to by
427    \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
428  This code fragment shows a typical straightforward call to \fBpcre_compile()\fR:  .P
429    If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
430    \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
431    returned via this argument in the event of an error. This is in addition to the
432    textual error message. Error codes and messages are listed below.
433    .P
434    If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
435    character tables that are built when PCRE is compiled, using the default C
436    locale. Otherwise, \fItableptr\fP must be an address that is the result of a
437    call to \fBpcre_maketables()\fP. This value is stored with the compiled
438    pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
439    passed to it. For more discussion, see the section on locale support below.
440    .P
441    This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
442    .sp
443    pcre *re;    pcre *re;
444    const char *error;    const char *error;
445    int erroffset;    int erroffset;
# Line 265  This code fragment shows a typical strai Line 449  This code fragment shows a typical strai
449      &error,           /* for error message */      &error,           /* for error message */
450      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
451      NULL);            /* use default character tables */      NULL);            /* use default character tables */
452    .sp
453  The following option bits are defined:  The following names for option bits are defined in the \fBpcre.h\fP header
454    file:
455    .sp
456    PCRE_ANCHORED    PCRE_ANCHORED
457    .sp
458  If this bit is set, the pattern is forced to be "anchored", that is, it is  If this bit is set, the pattern is forced to be "anchored", that is, it is
459  constrained to match only at the first matching point in the string which is  constrained to match only at the first matching point in the string that is
460  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
461  appropriate constructs in the pattern itself, which is the only way to do it in  appropriate constructs in the pattern itself, which is the only way to do it in
462  Perl.  Perl.
463    .sp
464      PCRE_AUTO_CALLOUT
465    .sp
466    If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
467    all with number 255, before each pattern item. For discussion of the callout
468    facility, see the
469    .\" HREF
470    \fBpcrecallout\fP
471    .\"
472    documentation.
473    .sp
474      PCRE_BSR_ANYCRLF
475      PCRE_BSR_UNICODE
476    .sp
477    These options (which are mutually exclusive) control what the \eR escape
478    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
479    match any Unicode newline sequence. The default is specified when PCRE is
480    built. It can be overridden from within the pattern, or by setting an option
481    when a compiled pattern is matched.
482    .sp
483    PCRE_CASELESS    PCRE_CASELESS
484    .sp
485  If this bit is set, letters in the pattern match both upper and lower case  If this bit is set, letters in the pattern match both upper and lower case
486  letters. It is equivalent to Perl's /i option, and it can be changed within a  letters. It is equivalent to Perl's /i option, and it can be changed within a
487  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
488    concept of case for characters whose values are less than 128, so caseless
489    matching is always possible. For characters with higher values, the concept of
490    case is supported if PCRE is compiled with Unicode property support, but not
491    otherwise. If you want to use caseless matching for characters 128 and above,
492    you must ensure that PCRE is compiled with Unicode property support as well as
493    with UTF-8 support.
494    .sp
495    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
496    .sp
497  If this bit is set, a dollar metacharacter in the pattern matches only at the  If this bit is set, a dollar metacharacter in the pattern matches only at the
498  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
499  immediately before the final character if it is a newline (but not before any  immediately before a newline at the end of the string (but not before any other
500  other newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is  newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
501  set. There is no equivalent to this option in Perl, and no way to set it within  There is no equivalent to this option in Perl, and no way to set it within a
502  a pattern.  pattern.
503    .sp
504    PCRE_DOTALL    PCRE_DOTALL
505    .sp
506  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharater in the pattern matches all characters,
507  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
508  equivalent to Perl's /s option, and it can be changed within a pattern by a  the current position is at a newline. This option is equivalent to Perl's /s
509  (?s) option setting. A negative class such as [^a] always matches a newline  option, and it can be changed within a pattern by a (?s) option setting. A
510  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
511    the setting of this option.
512    .sp
513      PCRE_DUPNAMES
514    .sp
515    If this bit is set, names used to identify capturing subpatterns need not be
516    unique. This can be helpful for certain types of pattern when it is known that
517    only one instance of the named subpattern can ever be matched. There are more
518    details of named subpatterns below; see also the
519    .\" HREF
520    \fBpcrepattern\fP
521    .\"
522    documentation.
523    .sp
524    PCRE_EXTENDED    PCRE_EXTENDED
525    .sp
526  If this bit is set, whitespace data characters in the pattern are totally  If this bit is set, whitespace data characters in the pattern are totally
527  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
528  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
529  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
530  inclusive, are also ignored. This is equivalent to Perl's /x option, and it can  ignored. This is equivalent to Perl's /x option, and it can be changed within a
531  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
532    .P
533  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
534  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
535  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
536  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
537    .sp
538    PCRE_EXTRA    PCRE_EXTRA
539    .sp
540  This option was invented in order to turn on additional functionality of PCRE  This option was invented in order to turn on additional functionality of PCRE
541  that is incompatible with Perl, but it is currently of very little use. When  that is incompatible with Perl, but it is currently of very little use. When
542  set, any backslash in a pattern that is followed by a letter that has no  set, any backslash in a pattern that is followed by a letter that has no
543  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
544  expansion. By default, as in Perl, a backslash followed by a letter with no  expansion. By default, as in Perl, a backslash followed by a letter with no
545  special meaning is treated as a literal. There are at present no other features  special meaning is treated as a literal. (Perl can, however, be persuaded to
546  controlled by this option. It can also be set by a (?X) option setting within a  give a warning for this.) There are at present no other features controlled by
547  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
548    .sp
549      PCRE_FIRSTLINE
550    .sp
551    If this option is set, an unanchored pattern is required to match before or at
552    the first newline in the subject string, though the matched text may continue
553    over the newline.
554    .sp
555      PCRE_JAVASCRIPT_COMPAT
556    .sp
557    If this option is set, PCRE's behaviour is changed in some ways so that it is
558    compatible with JavaScript rather than Perl. The changes are as follows:
559    .P
560    (1) A lone closing square bracket in a pattern causes a compile-time error,
561    because this is illegal in JavaScript (by default it is treated as a data
562    character). Thus, the pattern AB]CD becomes illegal when this option is set.
563    .P
564    (2) At run time, a back reference to an unset subpattern group matches an empty
565    string (by default this causes the current matching alternative to fail). A
566    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
567    an "a" in the subject), whereas it fails by default, for Perl compatibility.
568    .sp
569    PCRE_MULTILINE    PCRE_MULTILINE
570    .sp
571  By default, PCRE treats the subject string as consisting of a single "line" of  By default, PCRE treats the subject string as consisting of a single line of
572  characters (even if it actually contains several newlines). The "start of line"  characters (even if it actually contains newlines). The "start of line"
573  metacharacter (^) matches only at the start of the string, while the "end of  metacharacter (^) matches only at the start of the string, while the "end of
574  line" metacharacter ($) matches only at the end of the string, or before a  line" metacharacter ($) matches only at the end of the string, or before a
575  terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as  terminating newline (unless PCRE_DOLLAR_ENDONLY is set). This is the same as
576  Perl.  Perl.
577    .P
578  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
579  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
580  string, respectively, as well as at the very start and end. This is equivalent  subject string, respectively, as well as at the very start and end. This is
581  to Perl's /m option, and it can be changed within a pattern by a (?m) option  equivalent to Perl's /m option, and it can be changed within a pattern by a
582  setting. If there are no "\\n" characters in a subject string, or no  (?m) option setting. If there are no newlines in a subject string, or no
583  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
584    .sp
585      PCRE_NEWLINE_CR
586      PCRE_NEWLINE_LF
587      PCRE_NEWLINE_CRLF
588      PCRE_NEWLINE_ANYCRLF
589      PCRE_NEWLINE_ANY
590    .sp
591    These options override the default newline definition that was chosen when PCRE
592    was built. Setting the first or the second specifies that a newline is
593    indicated by a single character (CR or LF, respectively). Setting
594    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
595    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
596    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
597    that any Unicode newline sequence should be recognized. The Unicode newline
598    sequences are the three just mentioned, plus the single characters VT (vertical
599    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
600    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
601    recognized only in UTF-8 mode.
602    .P
603    The newline setting in the options word uses three bits that are treated
604    as a number, giving eight possibilities. Currently only six are used (default
605    plus the five values above). This means that if you set more than one newline
606    option, the combination may or may not be sensible. For example,
607    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
608    other combinations may yield unused numbers and cause an error.
609    .P
610    The only time that a line break is specially recognized when compiling a
611    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
612    class is encountered. This indicates a comment that lasts until after the next
613    line break sequence. In other circumstances, line break sequences are treated
614    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
615    as whitespace characters and are therefore ignored.
616    .P
617    The newline option that is set at compile time becomes the default that is used
618    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
619    .sp
620    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
621    .sp
622  If this option is set, it disables the use of numbered capturing parentheses in  If this option is set, it disables the use of numbered capturing parentheses in
623  the pattern. Any opening parenthesis that is not followed by ? behaves as if it  the pattern. Any opening parenthesis that is not followed by ? behaves as if it
624  were followed by ?: but named parentheses can still be used for capturing (and  were followed by ?: but named parentheses can still be used for capturing (and
625  they acquire numbers in the usual way). There is no equivalent of this option  they acquire numbers in the usual way). There is no equivalent of this option
626  in Perl.  in Perl.
627    .sp
628    PCRE_UNGREEDY    PCRE_UNGREEDY
629    .sp
630  This option inverts the "greediness" of the quantifiers so that they are not  This option inverts the "greediness" of the quantifiers so that they are not
631  greedy by default, but become greedy if followed by "?". It is not compatible  greedy by default, but become greedy if followed by "?". It is not compatible
632  with Perl. It can also be set by a (?U) option setting within the pattern.  with Perl. It can also be set by a (?U) option setting within the pattern.
633    .sp
634    PCRE_UTF8    PCRE_UTF8
635    .sp
636  This option causes PCRE to regard both the pattern and the subject as strings  This option causes PCRE to regard both the pattern and the subject as strings
637  of UTF-8 characters instead of single-byte character strings. However, it is  of UTF-8 characters instead of single-byte character strings. However, it is
638  available only if PCRE has been built to include UTF-8 support. If not, the use  available only when PCRE is built to include UTF-8 support. If not, the use
639  of this option provokes an error. Details of how this option changes the  of this option provokes an error. Details of how this option changes the
640  behaviour of PCRE are given in the  behaviour of PCRE are given in the
641  .\" HTML <a href="pcre.html#utf8support">  .\" HTML <a href="pcre.html#utf8support">
# Line 367  section on UTF-8 support Line 644  section on UTF-8 support
644  .\"  .\"
645  in the main  in the main
646  .\" HREF  .\" HREF
647  \fBpcre\fR  \fBpcre\fP
648  .\"  .\"
649  page.  page.
650    .sp
651  .SH STUDYING A PATTERN    PCRE_NO_UTF8_CHECK
652    .sp
653    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
654    automatically checked. There is a discussion about the
655    .\" HTML <a href="pcre.html#utf8strings">
656    .\" </a>
657    validity of UTF-8 strings
658    .\"
659    in the main
660    .\" HREF
661    \fBpcre\fP
662    .\"
663    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
664    returns an error. If you already know that your pattern is valid, and you want
665    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
666    option. When it is set, the effect of passing an invalid UTF-8 string as a
667    pattern is undefined. It may cause your program to crash. Note that this option
668    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
669    the UTF-8 validity checking of subject strings.
670    .
671    .
672    .SH "COMPILATION ERROR CODES"
673    .rs
674    .sp
675    The following table lists the error codes than may be returned by
676    \fBpcre_compile2()\fP, along with the error messages that may be returned by
677    both compiling functions. As PCRE has developed, some error codes have fallen
678    out of use. To avoid confusion, they have not been re-used.
679    .sp
680       0  no error
681       1  \e at end of pattern
682       2  \ec at end of pattern
683       3  unrecognized character follows \e
684       4  numbers out of order in {} quantifier
685       5  number too big in {} quantifier
686       6  missing terminating ] for character class
687       7  invalid escape sequence in character class
688       8  range out of order in character class
689       9  nothing to repeat
690      10  [this code is not in use]
691      11  internal error: unexpected repeat
692      12  unrecognized character after (? or (?-
693      13  POSIX named classes are supported only within a class
694      14  missing )
695      15  reference to non-existent subpattern
696      16  erroffset passed as NULL
697      17  unknown option bit(s) set
698      18  missing ) after comment
699      19  [this code is not in use]
700      20  regular expression is too large
701      21  failed to get memory
702      22  unmatched parentheses
703      23  internal error: code overflow
704      24  unrecognized character after (?<
705      25  lookbehind assertion is not fixed length
706      26  malformed number or name after (?(
707      27  conditional group contains more than two branches
708      28  assertion expected after (?(
709      29  (?R or (?[+-]digits must be followed by )
710      30  unknown POSIX class name
711      31  POSIX collating elements are not supported
712      32  this version of PCRE is not compiled with PCRE_UTF8 support
713      33  [this code is not in use]
714      34  character value in \ex{...} sequence is too large
715      35  invalid condition (?(0)
716      36  \eC not allowed in lookbehind assertion
717      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
718      38  number after (?C is > 255
719      39  closing ) for (?C expected
720      40  recursive call could loop indefinitely
721      41  unrecognized character after (?P
722      42  syntax error in subpattern name (missing terminator)
723      43  two named subpatterns have the same name
724      44  invalid UTF-8 string
725      45  support for \eP, \ep, and \eX has not been compiled
726      46  malformed \eP or \ep sequence
727      47  unknown property name after \eP or \ep
728      48  subpattern name is too long (maximum 32 characters)
729      49  too many named subpatterns (maximum 10000)
730      50  [this code is not in use]
731      51  octal value is greater than \e377 (not in UTF-8 mode)
732      52  internal error: overran compiling workspace
733      53  internal error: previously-checked referenced subpattern not found
734      54  DEFINE group contains more than one branch
735      55  repeating a DEFINE group is not allowed
736      56  inconsistent NEWLINE options
737      57  \eg is not followed by a braced, angle-bracketed, or quoted
738            name/number or by a plain number
739      58  a numbered reference must not be zero
740      59  (*VERB) with an argument is not supported
741      60  (*VERB) not recognized
742      61  number is too big
743      62  subpattern name expected
744      63  digit expected after (?+
745      64  ] is an invalid data character in JavaScript compatibility mode
746    .sp
747    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
748    be used if the limits were changed when PCRE was built.
749    .
750    .
751    .SH "STUDYING A PATTERN"
752  .rs  .rs
753  .sp  .sp
754  .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
755  .ti +5n  .ti +5n
756  .B const char **\fIerrptr\fR);  .B const char **\fIerrptr\fP);
757  .PP  .PP
758  When a pattern is going to be used several times, it is worth spending more  If a compiled pattern is going to be used several times, it is worth spending
759  time analyzing it in order to speed up the time taken for matching. The  more time analyzing it in order to speed up the time taken for matching. The
760  function \fBpcre_study()\fR takes a pointer to a compiled pattern as its first  function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
761  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
762  speed up matching, \fBpcre_study()\fR returns a pointer to a \fBpcre_extra\fR  help speed up matching, \fBpcre_study()\fP returns a pointer to a
763  block, in which the \fIstudy_data\fR field points to the results of the study.  \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
764    results of the study.
765  The returned value from a \fBpcre_study()\fR can be passed directly to  .P
766  \fBpcre_exec()\fR. However, the \fBpcre_extra\fR block also contains other  The returned value from \fBpcre_study()\fP can be passed directly to
767    \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
768  fields that can be set by the caller before the block is passed; these are  fields that can be set by the caller before the block is passed; these are
769  described below. If studying the pattern does not produce any additional  described
770  information, \fBpcre_study()\fR returns NULL. In that circumstance, if the  .\" HTML <a href="#extradata">
771  calling program wants to pass some of the other fields to \fBpcre_exec()\fR, it  .\" </a>
772  must set up its own \fBpcre_extra\fR block.  below
773    .\"
774  The second argument contains option bits. At present, no options are defined  in the section on matching a pattern.
775  for \fBpcre_study()\fR, and this argument should always be zero.  .P
776    If studying the pattern does not produce any additional information
777  The third argument for \fBpcre_study()\fR is a pointer for an error message. If  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
778    wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
779    own \fBpcre_extra\fP block.
780    .P
781    The second argument of \fBpcre_study()\fP contains option bits. At present, no
782    options are defined, and this argument should always be zero.
783    .P
784    The third argument for \fBpcre_study()\fP is a pointer for an error message. If
785  studying succeeds (even if no data is returned), the variable it points to is  studying succeeds (even if no data is returned), the variable it points to is
786  set to NULL. Otherwise it points to a textual error message. You should  set to NULL. Otherwise it is set to point to a textual error message. This is a
787  therefore test the error pointer for NULL after calling \fBpcre_study()\fR, to  static string that is part of the library. You must not try to free it. You
788  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
789    sure that it has run successfully.
790  This is a typical call to \fBpcre_study\fR():  .P
791    This is a typical call to \fBpcre_study\fP():
792    .sp
793    pcre_extra *pe;    pcre_extra *pe;
794    pe = pcre_study(    pe = pcre_study(
795      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
796      0,              /* no options exist */      0,              /* no options exist */
797      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
798    .sp
799  At present, studying a pattern is useful only for non-anchored patterns that do  At present, studying a pattern is useful only for non-anchored patterns that do
800  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
801  characters is created.  bytes is created.
802    .
803    .
804  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
805  .SH LOCALE SUPPORT  .SH "LOCALE SUPPORT"
806  .rs  .rs
807  .sp  .sp
808  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
809  digits, or whatever, by reference to a set of tables. When running in UTF-8  digits, or whatever, by reference to a set of tables, indexed by character
810  mode, this applies only to characters with codes less than 256. The library  value. When running in UTF-8 mode, this applies only to characters with codes
811  contains a default set of tables that is created in the default C locale when  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
812  PCRE is compiled. This is used when the final argument of \fBpcre_compile()\fR  can be tested with \ep if PCRE is built with Unicode character property
813  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
814    characters with codes greater than 128, you should either use UTF-8 and
815  An alternative set of tables can, however, be supplied. Such tables are built  Unicode, or use locales, but not try to mix the two.
816  by calling the \fBpcre_maketables()\fR function, which has no arguments, in the  .P
817  relevant locale. The result can then be passed to \fBpcre_compile()\fR as often  PCRE contains an internal set of tables that are used when the final argument
818  as necessary. For example, to build and use tables that are appropriate for the  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
819  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
820  treated as letters), the following code could be used:  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
821    default "C" locale of the local system, which may cause them to be different.
822    setlocale(LC_CTYPE, "fr");  .P
823    The internal tables can always be overridden by tables supplied by the
824    application that calls PCRE. These may be created in a different locale from
825    the default. As more and more applications change to using Unicode, the need
826    for this locale support is expected to die away.
827    .P
828    External tables are built by calling the \fBpcre_maketables()\fP function,
829    which has no arguments, in the relevant locale. The result can then be passed
830    to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
831    example, to build and use tables that are appropriate for the French locale
832    (where accented characters with values greater than 128 are treated as letters),
833    the following code could be used:
834    .sp
835      setlocale(LC_CTYPE, "fr_FR");
836    tables = pcre_maketables();    tables = pcre_maketables();
837    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
838    .sp
839  The tables are built in memory that is obtained via \fBpcre_malloc\fR. The  The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
840  pointer that is passed to \fBpcre_compile\fR is saved with the compiled  are using Windows, the name for the French locale is "french".
841  pattern, and the same tables are used via this pointer by \fBpcre_study()\fR  .P
842  and \fBpcre_exec()\fR. Thus, for any single pattern, compilation, studying and  When \fBpcre_maketables()\fP runs, the tables are built in memory that is
843  matching all happen in the same locale, but different patterns can be compiled  obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
844  in different locales. It is the caller's responsibility to ensure that the  that the memory containing the tables remains available for as long as it is
845  memory containing the tables remains available for as long as it is needed.  needed.
846    .P
847  .SH INFORMATION ABOUT A PATTERN  The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
848    pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
849    and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
850    pattern, compilation, studying and matching all happen in the same locale, but
851    different patterns can be compiled in different locales.
852    .P
853    It is possible to pass a table pointer or NULL (indicating the use of the
854    internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
855    this facility could be used to match a pattern in a different locale from the
856    one in which it was compiled. Passing table pointers at run time is discussed
857    below in the section on matching a pattern.
858    .
859    .
860    .SH "INFORMATION ABOUT A PATTERN"
861  .rs  .rs
862  .sp  .sp
863  .B int pcre_fullinfo(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
864  .ti +5n  .ti +5n
865  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
866  .PP  .PP
867  The \fBpcre_fullinfo()\fR function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
868  pattern. It replaces the obsolete \fBpcre_info()\fR function, which is  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
869  nevertheless retained for backwards compability (and is documented below).  nevertheless retained for backwards compability (and is documented below).
870    .P
871  The first argument for \fBpcre_fullinfo()\fR is a pointer to the compiled  The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
872  pattern. The second argument is the result of \fBpcre_study()\fR, or NULL if  pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
873  the pattern was not studied. The third argument specifies which piece of  the pattern was not studied. The third argument specifies which piece of
874  information is required, and the fourth argument is a pointer to a variable  information is required, and the fourth argument is a pointer to a variable
875  to receive the data. The yield of the function is zero for success, or one of  to receive the data. The yield of the function is zero for success, or one of
876  the following negative numbers:  the following negative numbers:
877    .sp
878    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
879                          the argument \fIwhere\fR was NULL                          the argument \fIwhere\fP was NULL
880    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
881    PCRE_ERROR_BADOPTION  the value of \fIwhat\fR was invalid    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid
882    .sp
883  Here is a typical call of \fBpcre_fullinfo()\fR, to obtain the length of the  The "magic number" is placed at the start of each compiled pattern as an simple
884  compiled pattern:  check against passing an arbitrary memory pointer. Here is a typical call of
885    \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
886    .sp
887    int rc;    int rc;
888    unsigned long int length;    size_t length;
889    rc = pcre_fullinfo(    rc = pcre_fullinfo(
890      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
891      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
892      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
893      &length);         /* where to put the data */      &length);         /* where to put the data */
894    .sp
895  The possible values for the third argument are defined in \fBpcre.h\fR, and are  The possible values for the third argument are defined in \fBpcre.h\fP, and are
896  as follows:  as follows:
897    .sp
898    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
899    .sp
900  Return the number of the highest back reference in the pattern. The fourth  Return the number of the highest back reference in the pattern. The fourth
901  argument should point to an \fBint\fR variable. Zero is returned if there are  argument should point to an \fBint\fP variable. Zero is returned if there are
902  no back references.  no back references.
903    .sp
904    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
905    .sp
906  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
907  should point to an \fbint\fR variable.  should point to an \fBint\fP variable.
908    .sp
909      PCRE_INFO_DEFAULT_TABLES
910    .sp
911    Return a pointer to the internal default character tables within PCRE. The
912    fourth argument should point to an \fBunsigned char *\fP variable. This
913    information call is provided for internal use by the \fBpcre_study()\fP
914    function. External callers can cause PCRE to use its internal tables by passing
915    a NULL table pointer.
916    .sp
917    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
918    .sp
919  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
920  non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the  non-anchored pattern. The fourth argument should point to an \fBint\fP
921  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
922    still recognized for backwards compatibility.)
923  If there is a fixed first byte, e.g. from a pattern such as (cat|cow|coyote),  .P
924  it is returned in the integer pointed to by \fIwhere\fR. Otherwise, if either  If there is a fixed first byte, for example, from a pattern such as
925    (cat|cow|coyote), its value is returned. Otherwise, if either
926    .sp
927  (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch  (a) the pattern was compiled with the PCRE_MULTILINE option, and every branch
928  starts with "^", or  starts with "^", or
929    .sp
930  (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set  (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not set
931  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
932    .sp
933  -1 is returned, indicating that the pattern matches only at the start of a  -1 is returned, indicating that the pattern matches only at the start of a
934  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
935  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
936    .sp
937    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
938    .sp
939  If the pattern was studied, and this resulted in the construction of a 256-bit  If the pattern was studied, and this resulted in the construction of a 256-bit
940  table indicating a fixed set of bytes for the first byte in any matching  table indicating a fixed set of bytes for the first byte in any matching
941  string, a pointer to the table is returned. Otherwise NULL is returned. The  string, a pointer to the table is returned. Otherwise NULL is returned. The
942  fourth argument should point to an \fBunsigned char *\fR variable.  fourth argument should point to an \fBunsigned char *\fP variable.
943    .sp
944      PCRE_INFO_HASCRORLF
945    .sp
946    Return 1 if the pattern contains any explicit matches for CR or LF characters,
947    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
948    explicit match is either a literal CR or LF character, or \er or \en.
949    .sp
950      PCRE_INFO_JCHANGED
951    .sp
952    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
953    0. The fourth argument should point to an \fBint\fP variable. (?J) and
954    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
955    .sp
956    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
957    .sp
958  Return the value of the rightmost literal byte that must exist in any matched  Return the value of the rightmost literal byte that must exist in any matched
959  string, other than at its start, if such a byte has been recorded. The fourth  string, other than at its start, if such a byte has been recorded. The fourth
960  argument should point to an \fBint\fR variable. If there is no such byte, -1 is  argument should point to an \fBint\fP variable. If there is no such byte, -1 is
961  returned. For anchored patterns, a last literal byte is recorded only if it  returned. For anchored patterns, a last literal byte is recorded only if it
962  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
963  /^a\\d+z\\d+/ the returned value is "z", but for /^a\\dz\\d/ the returned value  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
964  is -1.  is -1.
965    .sp
966    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
967    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
968    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
969    .sp
970  PCRE supports the use of named as well as numbered capturing parentheses. The  PCRE supports the use of named as well as numbered capturing parentheses. The
971  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
972  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
973  must convert the name to a number in order to access the correct pointers in  \fBpcre_get_named_substring()\fP are provided for extracting captured
974  the output vector (described with \fBpcre_exec()\fR below). In order to do  substrings by name. It is also possible to extract the data directly, by first
975  this, it must first use these three values to obtain the name-to-number mapping  converting the name to a number in order to access the correct pointers in the
976  table for the pattern.  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
977    you need to use the name-to-number map, which is described by these three
978    values.
979    .P
980  The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives  The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT gives
981  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
982  entry; both of these return an \fBint\fR value. The entry size depends on the  entry; both of these return an \fBint\fP value. The entry size depends on the
983  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
984  entry of the table (a pointer to \fBchar\fR). The first two bytes of each entry  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
985  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
986  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated. The names are in
987  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
988    their parentheses numbers. For example, consider the following pattern (assume
989  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
990    .sp
991    (?P<date> (?P<year>(\\d\\d)?\\d\\d) -  .\" JOIN
992    (?P<month>\\d\\d) - (?P<day>\\d\\d) )    (?<date> (?<year>(\ed\ed)?\ed\ed) -
993      (?<month>\ed\ed) - (?<day>\ed\ed) )
994    .sp
995  There are four named subpatterns, so the table has four entries, and each entry  There are four named subpatterns, so the table has four entries, and each entry
996  in the table is eight bytes long. The table is as follows, with non-printing  in the table is eight bytes long. The table is as follows, with non-printing
997  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
998    .sp
999    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
1000    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
1001    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
1002    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
1003    .sp
1004  When writing code to extract data from named subpatterns, remember that the  When writing code to extract data from named subpatterns using the
1005  length of each entry may be different for each compiled pattern.  name-to-number map, remember that the length of the entries is likely to be
1006    different for each compiled pattern.
1007    .sp
1008      PCRE_INFO_OKPARTIAL
1009    .sp
1010    Return 1 if the pattern can be used for partial matching, otherwise 0. The
1011    fourth argument should point to an \fBint\fP variable. From release 8.00, this
1012    always returns 1, because the restrictions that previously applied to partial
1013    matching have been lifted. The
1014    .\" HREF
1015    \fBpcrepartial\fP
1016    .\"
1017    documentation gives details of partial matching.
1018    .sp
1019    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1020    .sp
1021  Return a copy of the options with which the pattern was compiled. The fourth  Return a copy of the options with which the pattern was compiled. The fourth
1022  argument should point to an \fBunsigned long int\fR variable. These option bits  argument should point to an \fBunsigned long int\fP variable. These option bits
1023  are those specified in the call to \fBpcre_compile()\fR, modified by any  are those specified in the call to \fBpcre_compile()\fP, modified by any
1024  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1025    they are the options that will be in force when matching starts. For example,
1026    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1027    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1028    .P
1029  A pattern is automatically anchored by PCRE if all of its top-level  A pattern is automatically anchored by PCRE if all of its top-level
1030  alternatives begin with one of the following:  alternatives begin with one of the following:
1031    .sp
1032    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
1033    \\A    always    \eA    always
1034    \\G    always    \eG    always
1035    .\" JOIN
1036    .*    if PCRE_DOTALL is set and there are no back    .*    if PCRE_DOTALL is set and there are no back
1037            references to the subpattern in which .* appears            references to the subpattern in which .* appears
1038    .sp
1039  For such patterns, the PCRE_ANCHORED bit is set in the options returned by  For such patterns, the PCRE_ANCHORED bit is set in the options returned by
1040  \fBpcre_fullinfo()\fR.  \fBpcre_fullinfo()\fP.
1041    .sp
1042    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1043    .sp
1044  Return the size of the compiled pattern, that is, the value that was passed as  Return the size of the compiled pattern, that is, the value that was passed as
1045  the argument to \fBpcre_malloc()\fR when PCRE was getting memory in which to  the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to
1046  place the compiled data. The fourth argument should point to a \fBsize_t\fR  place the compiled data. The fourth argument should point to a \fBsize_t\fP
1047  variable.  variable.
1048    .sp
1049    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1050    .sp
1051  Returns the size of the data block pointed to by the \fIstudy_data\fR field in  Return the size of the data block pointed to by the \fIstudy_data\fP field in
1052  a \fBpcre_extra\fR block. That is, it is the value that was passed to  a \fBpcre_extra\fP block. That is, it is the value that was passed to
1053  \fBpcre_malloc()\fR when PCRE was getting memory into which to place the data  \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1054  created by \fBpcre_study()\fR. The fourth argument should point to a  created by \fBpcre_study()\fP. The fourth argument should point to a
1055  \fBsize_t\fR variable.  \fBsize_t\fP variable.
1056    .
1057  .SH OBSOLETE INFO FUNCTION  .
1058    .SH "OBSOLETE INFO FUNCTION"
1059  .rs  .rs
1060  .sp  .sp
1061  .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1062  .B *\fIfirstcharptr\fR);  .B *\fIfirstcharptr\fP);
1063  .PP  .PP
1064  The \fBpcre_info()\fR function is now obsolete because its interface is too  The \fBpcre_info()\fP function is now obsolete because its interface is too
1065  restrictive to return all the available data about a compiled pattern. New  restrictive to return all the available data about a compiled pattern. New
1066  programs should use \fBpcre_fullinfo()\fR instead. The yield of  programs should use \fBpcre_fullinfo()\fP instead. The yield of
1067  \fBpcre_info()\fR is the number of capturing subpatterns, or one of the  \fBpcre_info()\fP is the number of capturing subpatterns, or one of the
1068  following negative numbers:  following negative numbers:
1069    .sp
1070    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
1071    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1072    .sp
1073  If the \fIoptptr\fR argument is not NULL, a copy of the options with which the  If the \fIoptptr\fP argument is not NULL, a copy of the options with which the
1074  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1075  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
1076    .P
1077  If the pattern is not anchored and the \fIfirstcharptr\fR argument is not NULL,  If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,
1078  it is used to pass back information about the first character of any matched  it is used to pass back information about the first character of any matched
1079  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1080    .
1081  .SH MATCHING A PATTERN  .
1082    .SH "REFERENCE COUNTS"
1083  .rs  .rs
1084  .sp  .sp
1085  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1086    .PP
1087    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1088    data block that contains a compiled pattern. It is provided for the benefit of
1089    applications that operate in an object-oriented manner, where different parts
1090    of the application may be using the same compiled pattern, but you want to free
1091    the block when they are all done.
1092    .P
1093    When a pattern is compiled, the reference count field is initialized to zero.
1094    It is changed only by calling this function, whose action is to add the
1095    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1096    function is the new value. However, the value of the count is constrained to
1097    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1098    it is forced to the appropriate limit value.
1099    .P
1100    Except when it is zero, the reference count is not correctly preserved if a
1101    pattern is compiled on one host and then transferred to a host whose byte-order
1102    is different. (This seems a highly unlikely scenario.)
1103    .
1104    .
1105    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1106    .rs
1107    .sp
1108    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1109  .ti +5n  .ti +5n
1110  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1111  .ti +5n  .ti +5n
1112  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1113  .PP  .P
1114  The function \fBpcre_exec()\fR is called to match a subject string against a  The function \fBpcre_exec()\fP is called to match a subject string against a
1115  pre-compiled pattern, which is passed in the \fIcode\fR argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1116  pattern has been studied, the result of the study should be passed in the  pattern has been studied, the result of the study should be passed in the
1117  \fIextra\fR argument.  \fIextra\fP argument. This function is the main matching facility of the
1118    library, and it operates in a Perl-like manner. For specialist use there is
1119  Here is an example of a simple call to \fBpcre_exec()\fR:  also an alternative matching function, which is described
1120    .\" HTML <a href="#dfamatch">
1121    .\" </a>
1122    below
1123    .\"
1124    in the section about the \fBpcre_dfa_exec()\fP function.
1125    .P
1126    In most applications, the pattern will have been compiled (and optionally
1127    studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1128    possible to save compiled patterns and study data, and then use them later
1129    in different processes, possibly even on different hosts. For a discussion
1130    about this, see the
1131    .\" HREF
1132    \fBpcreprecompile\fP
1133    .\"
1134    documentation.
1135    .P
1136    Here is an example of a simple call to \fBpcre_exec()\fP:
1137    .sp
1138    int rc;    int rc;
1139    int ovector[30];    int ovector[30];
1140    rc = pcre_exec(    rc = pcre_exec(
# Line 647  Here is an example of a simple call to \ Line 1144  Here is an example of a simple call to \
1144      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1145      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1146      0,              /* default options */      0,              /* default options */
1147      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1148      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1149    .
1150  If the \fIextra\fR argument is not NULL, it must point to a \fBpcre_extra\fR  .\" HTML <a name="extradata"></a>
1151  data block. The \fBpcre_study()\fR function returns such a block (when it  .SS "Extra data for \fBpcre_exec()\fR"
1152    .rs
1153    .sp
1154    If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1155    data block. The \fBpcre_study()\fP function returns such a block (when it
1156  doesn't return NULL), but you can also create one for yourself, and pass  doesn't return NULL), but you can also create one for yourself, and pass
1157  additional information in it. The fields in the block are as follows:  additional information in it. The \fBpcre_extra\fP block contains the following
1158    fields (not necessarily in this order):
1159    unsigned long int \fIflags\fR;  .sp
1160    void *\fIstudy_data\fR;    unsigned long int \fIflags\fP;
1161    unsigned long int \fImatch_limit\fR;    void *\fIstudy_data\fP;
1162    void *\fIcallout_data\fR;    unsigned long int \fImatch_limit\fP;
1163      unsigned long int \fImatch_limit_recursion\fP;
1164  The \fIflags\fR field is a bitmap that specifies which of the other fields    void *\fIcallout_data\fP;
1165      const unsigned char *\fItables\fP;
1166    .sp
1167    The \fIflags\fP field is a bitmap that specifies which of the other fields
1168  are set. The flag bits are:  are set. The flag bits are:
1169    .sp
1170    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1171    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1172      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1173    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1174      PCRE_EXTRA_TABLES
1175  Other flag bits should be set to zero. The \fIstudy_data\fR field is set in the  .sp
1176  \fBpcre_extra\fR block that is returned by \fBpcre_study()\fR, together with  Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1177  the appropriate flag bit. You should not set this yourself, but you can add to  \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
1178  the block by setting the other fields.  the appropriate flag bit. You should not set this yourself, but you may add to
1179    the block by setting the other fields and their corresponding flag bits.
1180  The \fImatch_limit\fR field provides a means of preventing PCRE from using up a  .P
1181    The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1182  vast amount of resources when running patterns that are not going to match,  vast amount of resources when running patterns that are not going to match,
1183  but which have a very large number of possibilities in their search trees. The  but which have a very large number of possibilities in their search trees. The
1184  classic example is the use of nested unlimited repeats. Internally, PCRE uses a  classic example is the use of nested unlimited repeats.
1185  function called \fBmatch()\fR which it calls repeatedly (sometimes  .P
1186  recursively). The limit is imposed on the number of times this function is  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1187  called during a match, which has the effect of limiting the amount of recursion  (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1188  and backtracking that can take place. For patterns that are not anchored, the  number of times this function is called during a match, which has the effect of
1189  count starts from zero for each position in the subject string.  limiting the amount of backtracking that can take place. For patterns that are
1190    not anchored, the count restarts from zero for each position in the subject
1191  The default limit for the library can be set when PCRE is built; the default  string.
1192    .P
1193    The default value for the limit can be set when PCRE is built; the default
1194  default is 10 million, which handles all but the most extreme cases. You can  default is 10 million, which handles all but the most extreme cases. You can
1195  reduce the default by suppling \fBpcre_exec()\fR with a \fRpcre_extra\fR block  override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1196  in which \fImatch_limit\fR is set to a smaller value, and  block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1197  PCRE_EXTRA_MATCH_LIMIT is set in the \fIflags\fR field. If the limit is  the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1198  exceeded, \fBpcre_exec()\fR returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1199    .P
1200  The \fIpcre_callout\fR field is used in conjunction with the "callout" feature,  The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1201  which is described in the \fBpcrecallout\fR documentation.  instead of limiting the total number of times that \fBmatch()\fP is called, it
1202    limits the depth of recursion. The recursion depth is a smaller number than the
1203  The PCRE_ANCHORED option can be passed in the \fIoptions\fR argument, whose  total number of calls, because not all calls to \fBmatch()\fP are recursive.
1204  unused bits must be zero. This limits \fBpcre_exec()\fR to matching at the  This limit is of use only if it is set smaller than \fImatch_limit\fP.
1205  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  .P
1206  or turned out to be anchored by virtue of its contents, it cannot be made  Limiting the recursion depth limits the amount of stack that can be used, or,
1207  unachored at matching time.  when PCRE has been compiled to use memory on the heap instead of the stack, the
1208    amount of heap memory that can be used.
1209  There are also three further options that can be set only at matching time:  .P
1210    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1211    built; the default default is the same value as the default for
1212    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1213    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1214    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1215    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1216    .P
1217    The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1218    which is described in the
1219    .\" HREF
1220    \fBpcrecallout\fP
1221    .\"
1222    documentation.
1223    .P
1224    The \fItables\fP field is used to pass a character tables pointer to
1225    \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1226    pattern. A non-NULL value is stored with the compiled pattern only if custom
1227    tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1228    If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1229    internal tables to be used. This facility is helpful when re-using patterns
1230    that have been saved after compiling with an external set of tables, because
1231    the external tables might be at a different address when \fBpcre_exec()\fP is
1232    called. See the
1233    .\" HREF
1234    \fBpcreprecompile\fP
1235    .\"
1236    documentation for a discussion of saving compiled patterns for later use.
1237    .
1238    .\" HTML <a name="execoptions"></a>
1239    .SS "Option bits for \fBpcre_exec()\fP"
1240    .rs
1241    .sp
1242    The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1243    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1244    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_START_OPTIMIZE,
1245    PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1246    .sp
1247      PCRE_ANCHORED
1248    .sp
1249    The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1250    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1251    to be anchored by virtue of its contents, it cannot be made unachored at
1252    matching time.
1253    .sp
1254      PCRE_BSR_ANYCRLF
1255      PCRE_BSR_UNICODE
1256    .sp
1257    These options (which are mutually exclusive) control what the \eR escape
1258    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1259    match any Unicode newline sequence. These options override the choice that was
1260    made or defaulted when the pattern was compiled.
1261    .sp
1262      PCRE_NEWLINE_CR
1263      PCRE_NEWLINE_LF
1264      PCRE_NEWLINE_CRLF
1265      PCRE_NEWLINE_ANYCRLF
1266      PCRE_NEWLINE_ANY
1267    .sp
1268    These options override the newline definition that was chosen or defaulted when
1269    the pattern was compiled. For details, see the description of
1270    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1271    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1272    the way the match position is advanced after a match failure for an unanchored
1273    pattern.
1274    .P
1275    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1276    match attempt for an unanchored pattern fails when the current position is at a
1277    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1278    characters, the match position is advanced by two characters instead of one, in
1279    other words, to after the CRLF.
1280    .P
1281    The above rule is a compromise that makes the most common cases work as
1282    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1283    set), it does not match the string "\er\enA" because, after failing at the
1284    start, it skips both the CR and the LF before retrying. However, the pattern
1285    [\er\en]A does match that string, because it contains an explicit CR or LF
1286    reference, and so advances only by one character after the first failure.
1287    .P
1288    An explicit match for CR of LF is either a literal appearance of one of those
1289    characters, or one of the \er or \en escape sequences. Implicit matches such as
1290    [^X] do not count, nor does \es (which includes CR and LF in the characters
1291    that it matches).
1292    .P
1293    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1294    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1295    .sp
1296    PCRE_NOTBOL    PCRE_NOTBOL
1297    .sp
1298  The first character of the string is not the beginning of a line, so the  This option specifies that first character of the subject string is not the
1299  circumflex metacharacter should not match before it. Setting this without  beginning of a line, so the circumflex metacharacter should not match before
1300  PCRE_MULTILINE (at compile time) causes circumflex never to match.  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1301    never to match. This option affects only the behaviour of the circumflex
1302    metacharacter. It does not affect \eA.
1303    .sp
1304    PCRE_NOTEOL    PCRE_NOTEOL
1305    .sp
1306  The end of the string is not the end of a line, so the dollar metacharacter  This option specifies that the end of the subject string is not the end of a
1307  should not match it nor (except in multiline mode) a newline immediately before  line, so the dollar metacharacter should not match it nor (except in multiline
1308  it. Setting this without PCRE_MULTILINE (at compile time) causes dollar never  mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1309  to match.  compile time) causes dollar never to match. This option affects only the
1310    behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1311    .sp
1312    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1313    .sp
1314  An empty string is not considered to be a valid match if this option is set. If  An empty string is not considered to be a valid match if this option is set. If
1315  there are alternatives in the pattern, they are tried. If all the alternatives  there are alternatives in the pattern, they are tried. If all the alternatives
1316  match the empty string, the entire match fails. For example, if the pattern  match the empty string, the entire match fails. For example, if the pattern
1317    .sp
1318    a?b?    a?b?
1319    .sp
1320  is applied to a string not beginning with "a" or "b", it matches the empty  is applied to a string not beginning with "a" or "b", it matches the empty
1321  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1322  valid, so PCRE searches further into the string for occurrences of "a" or "b".  valid, so PCRE searches further into the string for occurrences of "a" or "b".
1323    .P
1324  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1325  of a pattern match of the empty string within its \fBsplit()\fR function, and  of a pattern match of the empty string within its \fBsplit()\fP function, and
1326  when using the /g modifier. It is possible to emulate Perl's behaviour after  when using the /g modifier. It is possible to emulate Perl's behaviour after
1327  matching a null string by first trying the match again at the same offset with  matching a null string by first trying the match again at the same offset with
1328  PCRE_NOTEMPTY set, and then if that fails by advancing the starting offset (see  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1329  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1330    code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1331  The subject string is passed to \fBpcre_exec()\fR as a pointer in  .sp
1332  \fIsubject\fR, a length in \fIlength\fR, and a starting offset in    PCRE_NO_START_OPTIMIZE
1333  \fIstartoffset\fR. Unlike the pattern string, the subject may contain binary  .sp
1334    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1335    a match, in order to speed up the process. For example, if it is known that a
1336    match must start with a specific character, it searches the subject for that
1337    character, and fails immediately if it cannot find it, without actually running
1338    the main matching function. When callouts are in use, these optimizations can
1339    cause them to be skipped. This option disables the "start-up" optimizations,
1340    causing performance to suffer, but ensuring that the callouts do occur.
1341    .sp
1342      PCRE_NO_UTF8_CHECK
1343    .sp
1344    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1345    string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1346    The value of \fIstartoffset\fP is also checked to ensure that it points to the
1347    start of a UTF-8 character. There is a discussion about the validity of UTF-8
1348    strings in the
1349    .\" HTML <a href="pcre.html#utf8strings">
1350    .\" </a>
1351    section on UTF-8 support
1352    .\"
1353    in the main
1354    .\" HREF
1355    \fBpcre\fP
1356    .\"
1357    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1358    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1359    PCRE_ERROR_BADUTF8_OFFSET is returned.
1360    .P
1361    If you already know that your subject is valid, and you want to skip these
1362    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1363    calling \fBpcre_exec()\fP. You might want to do this for the second and
1364    subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1365    all the matches in a single subject string. However, you should be sure that
1366    the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1367    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1368    subject, or a value of \fIstartoffset\fP that does not point to the start of a
1369    UTF-8 character, is undefined. Your program may crash.
1370    .sp
1371      PCRE_PARTIAL
1372    .sp
1373    This option turns on the partial matching feature. If the subject string fails
1374    to match the pattern, but at some point during the matching process the end of
1375    the subject was reached (that is, the subject partially matches the pattern and
1376    the failure to match occurred only because there were not enough subject
1377    characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1378    PCRE_ERROR_NOMATCH. The portion of the string that provided the longest partial
1379    match is set as the first matching string. There is further discussion in the
1380    .\" HREF
1381    \fBpcrepartial\fP
1382    .\"
1383    documentation.
1384    .
1385    .SS "The string to be matched by \fBpcre_exec()\fP"
1386    .rs
1387    .sp
1388    The subject string is passed to \fBpcre_exec()\fP as a pointer in
1389    \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1390    in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of
1391    a UTF-8 character. Unlike the pattern string, the subject may contain binary
1392  zero bytes. When the starting offset is zero, the search for a match starts at  zero bytes. When the starting offset is zero, the search for a match starts at
1393  the beginning of the subject, and this is by far the most common case.  the beginning of the subject, and this is by far the most common case.
1394    .P
 If the pattern was compiled with the PCRE_UTF8 option, the subject must be a  
 sequence of bytes that is a valid UTF-8 string. If an invalid UTF-8 string is  
 passed, PCRE's behaviour is not defined.  
   
1395  A non-zero starting offset is useful when searching for another match in the  A non-zero starting offset is useful when searching for another match in the
1396  same subject by calling \fBpcre_exec()\fR again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
1397  Setting \fIstartoffset\fR differs from just passing over a shortened string and  Setting \fIstartoffset\fP differs from just passing over a shortened string and
1398  setting PCRE_NOTBOL in the case of a pattern that begins with any kind of  setting PCRE_NOTBOL in the case of a pattern that begins with any kind of
1399  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
1400    .sp
1401    \\Biss\\B    \eBiss\eB
1402    .sp
1403  which finds occurrences of "iss" in the middle of words. (\\B matches only if  which finds occurrences of "iss" in the middle of words. (\eB matches only if
1404  the current position in the subject is not a word boundary.) When applied to  the current position in the subject is not a word boundary.) When applied to
1405  the string "Mississipi" the first call to \fBpcre_exec()\fR finds the first  the string "Mississipi" the first call to \fBpcre_exec()\fP finds the first
1406  occurrence. If \fBpcre_exec()\fR is called again with just the remainder of the  occurrence. If \fBpcre_exec()\fP is called again with just the remainder of the
1407  subject, namely "issipi", it does not match, because \\B is always false at the  subject, namely "issipi", it does not match, because \eB is always false at the
1408  start of the subject, which is deemed to be a word boundary. However, if  start of the subject, which is deemed to be a word boundary. However, if
1409  \fBpcre_exec()\fR is passed the entire string again, but with \fIstartoffset\fR  \fBpcre_exec()\fP is passed the entire string again, but with \fIstartoffset\fP
1410  set to 4, it finds the second occurrence of "iss" because it is able to look  set to 4, it finds the second occurrence of "iss" because it is able to look
1411  behind the starting point to discover that it is preceded by a letter.  behind the starting point to discover that it is preceded by a letter.
1412    .P
1413  If a non-zero starting offset is passed when the pattern is anchored, one  If a non-zero starting offset is passed when the pattern is anchored, one
1414  attempt to match at the given offset is tried. This can only succeed if the  attempt to match at the given offset is made. This can only succeed if the
1415  pattern does not require the match to be at the start of the subject.  pattern does not require the match to be at the start of the subject.
1416    .
1417    .SS "How \fBpcre_exec()\fP returns captured substrings"
1418    .rs
1419    .sp
1420  In general, a pattern matches a certain portion of the subject, and in  In general, a pattern matches a certain portion of the subject, and in
1421  addition, further substrings from the subject may be picked out by parts of the  addition, further substrings from the subject may be picked out by parts of the
1422  pattern. Following the usage in Jeffrey Friedl's book, this is called  pattern. Following the usage in Jeffrey Friedl's book, this is called
1423  "capturing" in what follows, and the phrase "capturing subpattern" is used for  "capturing" in what follows, and the phrase "capturing subpattern" is used for
1424  a fragment of a pattern that picks out a substring. PCRE supports several other  a fragment of a pattern that picks out a substring. PCRE supports several other
1425  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1426    .P
1427  Captured substrings are returned to the caller via a vector of integer offsets  Captured substrings are returned to the caller via a vector of integers whose
1428  whose address is passed in \fIovector\fR. The number of elements in the vector  address is passed in \fIovector\fP. The number of elements in the vector is
1429  is passed in \fIovecsize\fR. The first two-thirds of the vector is used to pass  passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1430  back captured substrings, each substring using a pair of integers. The  argument is NOT the size of \fIovector\fP in bytes.
1431  remaining third of the vector is used as workspace by \fBpcre_exec()\fR while  .P
1432  matching capturing subpatterns, and is not available for passing back  The first two-thirds of the vector is used to pass back captured substrings,
1433  information. The length passed in \fIovecsize\fR should always be a multiple of  each substring using a pair of integers. The remaining third of the vector is
1434  three. If it is not, it is rounded down.  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1435    and is not available for passing back information. The number passed in
1436  When a match has been successful, information about captured substrings is  \fIovecsize\fP should always be a multiple of three. If it is not, it is
1437  returned in pairs of integers, starting at the beginning of \fIovector\fR, and  rounded down.
1438  continuing up to two-thirds of its length at the most. The first element of a  .P
1439  pair is set to the offset of the first character in a substring, and the second  When a match is successful, information about captured substrings is returned
1440  is set to the offset of the first character after the end of a substring. The  in pairs of integers, starting at the beginning of \fIovector\fP, and
1441  first pair, \fIovector[0]\fR and \fIovector[1]\fR, identify the portion of the  continuing up to two-thirds of its length at the most. The first element of
1442  subject string matched by the entire pattern. The next pair is used for the  each pair is set to the byte offset of the first character in a substring, and
1443  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fR  the second is set to the byte offset of the first character after the end of a
1444  is the number of pairs that have been set. If there are no capturing  substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1445  subpatterns, the return value from a successful match is 1, indicating that  mode. They are not character counts.
1446  just the first pair of offsets has been set.  .P
1447    The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1448  Some convenience functions are provided for extracting the captured substrings  portion of the subject string matched by the entire pattern. The next pair is
1449  as separate strings. These are described in the following section.  used for the first capturing subpattern, and so on. The value returned by
1450    \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1451  It is possible for an capturing subpattern number \fIn+1\fR to match some  For example, if two substrings have been captured, the returned value is 3. If
1452  part of the subject when subpattern \fIn\fR has not been used at all. For  there are no capturing subpatterns, the return value from a successful match is
1453  example, if the string "abc" is matched against the pattern (a|(z))(bc)  1, indicating that just the first pair of offsets has been set.
1454  subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset  .P
 values corresponding to the unused subpattern are set to -1.  
   
1455  If a capturing subpattern is matched repeatedly, it is the last portion of the  If a capturing subpattern is matched repeatedly, it is the last portion of the
1456  string that it matched that gets returned.  string that it matched that is returned.
1457    .P
1458  If the vector is too small to hold all the captured substrings, it is used as  If the vector is too small to hold all the captured substring offsets, it is
1459  far as possible (up to two-thirds of its length), and the function returns a  used as far as possible (up to two-thirds of its length), and the function
1460  value of zero. In particular, if the substring offsets are not of interest,  returns a value of zero. If the substring offsets are not of interest,
1461  \fBpcre_exec()\fR may be called with \fIovector\fR passed as NULL and  \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1462  \fIovecsize\fR as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1463  the \fIovector\fR isn't big enough to remember the related substrings, PCRE has  the \fIovector\fP is not big enough to remember the related substrings, PCRE
1464  to get additional memory for use during matching. Thus it is usually advisable  has to get additional memory for use during matching. Thus it is usually
1465  to supply an \fIovector\fR.  advisable to supply an \fIovector\fP.
1466    .P
1467  Note that \fBpcre_info()\fR can be used to find out how many capturing  The \fBpcre_info()\fP function can be used to find out how many capturing
1468  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1469  \fIovector\fR that will allow for \fIn\fR captured substrings, in addition to  \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1470  the offsets of the substring matched by the whole pattern, is (\fIn\fR+1)*3.  the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1471    .P
1472  If \fBpcre_exec()\fR fails, it returns a negative number. The following are  It is possible for capturing subpattern number \fIn+1\fP to match some part of
1473    the subject when subpattern \fIn\fP has not been used at all. For example, if
1474    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1475    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1476    happens, both values in the offset pairs corresponding to unused subpatterns
1477    are set to -1.
1478    .P
1479    Offset values that correspond to unused subpatterns at the end of the
1480    expression are also set to -1. For example, if the string "abc" is matched
1481    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1482    return from the function is 2, because the highest used capturing subpattern
1483    number is 1. However, you can refer to the offsets for the second and third
1484    capturing subpatterns if you wish (assuming the vector is large enough, of
1485    course).
1486    .P
1487    Some convenience functions are provided for extracting the captured substrings
1488    as separate strings. These are described below.
1489    .
1490    .\" HTML <a name="errorlist"></a>
1491    .SS "Error return values from \fBpcre_exec()\fP"
1492    .rs
1493    .sp
1494    If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1495  defined in the header file:  defined in the header file:
1496    .sp
1497    PCRE_ERROR_NOMATCH        (-1)    PCRE_ERROR_NOMATCH        (-1)
1498    .sp
1499  The subject string did not match the pattern.  The subject string did not match the pattern.
1500    .sp
1501    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1502    .sp
1503  Either \fIcode\fR or \fIsubject\fR was passed as NULL, or \fIovector\fR was  Either \fIcode\fP or \fIsubject\fP was passed as NULL, or \fIovector\fP was
1504  NULL and \fIovecsize\fR was not zero.  NULL and \fIovecsize\fP was not zero.
1505    .sp
1506    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1507    .sp
1508  An unrecognized bit was set in the \fIoptions\fR argument.  An unrecognized bit was set in the \fIoptions\fP argument.
1509    .sp
1510    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1511    .sp
1512  PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch  PCRE stores a 4-byte "magic number" at the start of the compiled code, to catch
1513  the case when it is passed a junk pointer. This is the error it gives when the  the case when it is passed a junk pointer and to detect when a pattern that was
1514  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1515    other endianness. This is the error that PCRE gives when the magic number is
1516    PCRE_ERROR_UNKNOWN_NODE   (-5)  not present.
1517    .sp
1518      PCRE_ERROR_UNKNOWN_OPCODE (-5)
1519    .sp
1520  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1521  compiled pattern. This error could be caused by a bug in PCRE or by overwriting  compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1522  of the compiled pattern.  of the compiled pattern.
1523    .sp
1524    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1525    .sp
1526  If a pattern contains back references, but the \fIovector\fR that is passed to  If a pattern contains back references, but the \fIovector\fP that is passed to
1527  \fBpcre_exec()\fR is not big enough to remember the referenced substrings, PCRE  \fBpcre_exec()\fP is not big enough to remember the referenced substrings, PCRE
1528  gets a block of memory at the start of matching to use for this purpose. If the  gets a block of memory at the start of matching to use for this purpose. If the
1529  call via \fBpcre_malloc()\fR fails, this error is given. The memory is freed at  call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1530  the end of matching.  automatically freed at the end of matching.
1531    .sp
1532    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1533    .sp
1534  This error is used by the \fBpcre_copy_substring()\fR,  This error is used by the \fBpcre_copy_substring()\fP,
1535  \fBpcre_get_substring()\fR, and \fBpcre_get_substring_list()\fR functions (see  \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1536  below). It is never returned by \fBpcre_exec()\fR.  below). It is never returned by \fBpcre_exec()\fP.
1537    .sp
1538    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1539    .sp
1540  The recursion and backtracking limit, as specified by the \fImatch_limit\fR  The backtracking limit, as specified by the \fImatch_limit\fP field in a
1541  field in a \fBpcre_extra\fR structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1542  description above.  above.
1543    .sp
1544    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1545    .sp
1546  This error is never generated by \fBpcre_exec()\fR itself. It is provided for  This error is never generated by \fBpcre_exec()\fP itself. It is provided for
1547  use by callout functions that want to yield a distinctive error code. See the  use by callout functions that want to yield a distinctive error code. See the
1548  \fBpcrecallout\fR documentation for details.  .\" HREF
1549    \fBpcrecallout\fP
1550  .SH EXTRACTING CAPTURED SUBSTRINGS BY NUMBER  .\"
1551    documentation for details.
1552    .sp
1553      PCRE_ERROR_BADUTF8        (-10)
1554    .sp
1555    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1556    .sp
1557      PCRE_ERROR_BADUTF8_OFFSET (-11)
1558    .sp
1559    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1560    of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1561    .sp
1562      PCRE_ERROR_PARTIAL        (-12)
1563    .sp
1564    The subject string did not match, but it did match partially. See the
1565    .\" HREF
1566    \fBpcrepartial\fP
1567    .\"
1568    documentation for details of partial matching.
1569    .sp
1570      PCRE_ERROR_BADPARTIAL     (-13)
1571    .sp
1572    This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1573    option was used with a compiled pattern containing items that were not
1574    supported for partial matching. From release 8.00 onwards, there are no
1575    restrictions on partial matching.
1576    .sp
1577      PCRE_ERROR_INTERNAL       (-14)
1578    .sp
1579    An unexpected internal error has occurred. This error could be caused by a bug
1580    in PCRE or by overwriting of the compiled pattern.
1581    .sp
1582      PCRE_ERROR_BADCOUNT       (-15)
1583    .sp
1584    This error is given if the value of the \fIovecsize\fP argument is negative.
1585    .sp
1586      PCRE_ERROR_RECURSIONLIMIT (-21)
1587    .sp
1588    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1589    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1590    description above.
1591    .sp
1592      PCRE_ERROR_BADNEWLINE     (-23)
1593    .sp
1594    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1595    .P
1596    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1597    .
1598    .
1599    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1600  .rs  .rs
1601  .sp  .sp
1602  .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1603  .ti +5n  .ti +5n
1604  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1605  .ti +5n  .ti +5n
1606  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
1607  .PP  .PP
1608  .br  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
 .B int pcre_get_substring(const char *\fIsubject\fR, int *\fIovector\fR,  
1609  .ti +5n  .ti +5n
1610  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1611  .ti +5n  .ti +5n
1612  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1613  .PP  .PP
1614  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
1615  .ti +5n  .ti +5n
1616  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1617  .PP  .PP
1618  Captured substrings can be accessed directly by using the offsets returned by  Captured substrings can be accessed directly by using the offsets returned by
1619  \fBpcre_exec()\fR in \fIovector\fR. For convenience, the functions  \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1620  \fBpcre_copy_substring()\fR, \fBpcre_get_substring()\fR, and  \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1621  \fBpcre_get_substring_list()\fR are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1622  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1623  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1624  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1625  has a further zero added on the end, but the result is not, of course,  .P
1626  a C string.  A substring that contains a binary zero is correctly extracted and has a
1627    further zero added on the end, but the result is not, of course, a C string.
1628    However, you can process such a string by referring to the length that is
1629    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1630    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1631    for handling strings containing binary zeros, because the end of the final
1632    string is not independently indicated.
1633    .P
1634  The first three arguments are the same for all three of these functions:  The first three arguments are the same for all three of these functions:
1635  \fIsubject\fR is the subject string which has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
1636  \fIovector\fR is a pointer to the vector of integer offsets that was passed to  \fIovector\fP is a pointer to the vector of integer offsets that was passed to
1637  \fBpcre_exec()\fR, and \fIstringcount\fR is the number of substrings that were  \fBpcre_exec()\fP, and \fIstringcount\fP is the number of substrings that were
1638  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1639  expression. This is the value returned by \fBpcre_exec\fR if it is greater than  expression. This is the value returned by \fBpcre_exec()\fP if it is greater
1640  zero. If \fBpcre_exec()\fR returned zero, indicating that it ran out of space  than zero. If \fBpcre_exec()\fP returned zero, indicating that it ran out of
1641  in \fIovector\fR, the value passed as \fIstringcount\fR should be the size of  space in \fIovector\fP, the value passed as \fIstringcount\fP should be the
1642  the vector divided by three.  number of elements in the vector divided by three.
1643    .P
1644  The functions \fBpcre_copy_substring()\fR and \fBpcre_get_substring()\fR  The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1645  extract a single substring, whose number is given as \fIstringnumber\fR. A  extract a single substring, whose number is given as \fIstringnumber\fP. A
1646  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1647  higher values extract the captured substrings. For \fBpcre_copy_substring()\fR,  higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1648  the string is placed in \fIbuffer\fR, whose length is given by  the string is placed in \fIbuffer\fP, whose length is given by
1649  \fIbuffersize\fR, while for \fBpcre_get_substring()\fR a new block of memory is  \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
1650  obtained via \fBpcre_malloc\fR, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1651  \fIstringptr\fR. The yield of the function is the length of the string, not  \fIstringptr\fP. The yield of the function is the length of the string, not
1652  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1653    .sp
1654    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1655    .sp
1656  The buffer was too small for \fBpcre_copy_substring()\fR, or the attempt to get  The buffer was too small for \fBpcre_copy_substring()\fP, or the attempt to get
1657  memory failed for \fBpcre_get_substring()\fR.  memory failed for \fBpcre_get_substring()\fP.
1658    .sp
1659    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1660    .sp
1661  There is no substring whose number is \fIstringnumber\fR.  There is no substring whose number is \fIstringnumber\fP.
1662    .P
1663  The \fBpcre_get_substring_list()\fR function extracts all available substrings  The \fBpcre_get_substring_list()\fP function extracts all available substrings
1664  and builds a list of pointers to them. All this is done in a single block of  and builds a list of pointers to them. All this is done in a single block of
1665  memory which is obtained via \fBpcre_malloc\fR. The address of the memory block  memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
1666  is returned via \fIlistptr\fR, which is also the start of the list of string  is returned via \fIlistptr\fP, which is also the start of the list of string
1667  pointers. The end of the list is marked by a NULL pointer. The yield of the  pointers. The end of the list is marked by a NULL pointer. The yield of the
1668  function is zero if all went well, or  function is zero if all went well, or the error code
1669    .sp
1670    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1671    .sp
1672  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1673    .P
1674  When any of these functions encounter a substring that is unset, which can  When any of these functions encounter a substring that is unset, which can
1675  happen when capturing subpattern number \fIn+1\fR matches some part of the  happen when capturing subpattern number \fIn+1\fP matches some part of the
1676  subject, but subpattern \fIn\fR has not been used at all, they return an empty  subject, but subpattern \fIn\fP has not been used at all, they return an empty
1677  string. This can be distinguished from a genuine zero-length substring by  string. This can be distinguished from a genuine zero-length substring by
1678  inspecting the appropriate offset in \fIovector\fR, which is negative for unset  inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1679  substrings.  substrings.
1680    .P
1681  The two convenience functions \fBpcre_free_substring()\fR and  The two convenience functions \fBpcre_free_substring()\fP and
1682  \fBpcre_free_substring_list()\fR can be used to free the memory returned by  \fBpcre_free_substring_list()\fP can be used to free the memory returned by
1683  a previous call of \fBpcre_get_substring()\fR or  a previous call of \fBpcre_get_substring()\fP or
1684  \fBpcre_get_substring_list()\fR, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1685  the function pointed to by \fBpcre_free\fR, which of course could be called  the function pointed to by \fBpcre_free\fP, which of course could be called
1686  directly from a C program. However, PCRE is used in some situations where it is  directly from a C program. However, PCRE is used in some situations where it is
1687  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1688  \fBpcre_free\fR directly; it is for these cases that the functions are  \fBpcre_free\fP directly; it is for these cases that the functions are
1689  provided.  provided.
1690    .
1691  .SH EXTRACTING CAPTURED SUBSTRINGS BY NAME  .
1692    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1693  .rs  .rs
1694  .sp  .sp
1695  .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1696  .ti +5n  .ti +5n
1697  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIname\fP);
1698    .PP
1699    .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1700  .ti +5n  .ti +5n
1701  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1702  .ti +5n  .ti +5n
1703  .B char *\fIbuffer\fR, int \fIbuffersize\fR);  .B int \fIstringcount\fP, const char *\fIstringname\fP,
 .PP  
 .br  
 .B int pcre_get_stringnumber(const pcre *\fIcode\fR,  
1704  .ti +5n  .ti +5n
1705  .B const char *\fIname\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1706  .PP  .PP
1707  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
 .B int pcre_get_named_substring(const pcre *\fIcode\fR,  
1708  .ti +5n  .ti +5n
1709  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1710  .ti +5n  .ti +5n
1711  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
1712  .ti +5n  .ti +5n
1713  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1714  .PP  .PP
1715  To extract a substring by name, you first have to find associated number. This  To extract a substring by name, you first have to find associated number.
1716  can be done by calling \fBpcre_get_stringnumber()\fR. The first argument is the  For example, for this pattern
1717  compiled pattern, and the second is the name. For example, for this pattern  .sp
1718      (a+)b(?<xxx>\ed+)...
1719    ab(?<xxx>\\d+)...  .sp
1720    the number of the subpattern called "xxx" is 2. If the name is known to be
1721  the number of the subpattern called "xxx" is 1. Given the number, you can then  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
1722  extract the substring directly, or use one of the functions described in the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1723  previous section. For convenience, there are also two functions that do the  pattern, and the second is the name. The yield of the function is the
1724  whole job.  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1725    that name.
1726  Most of the arguments of \fIpcre_copy_named_substring()\fR and  .P
1727  \fIpcre_get_named_substring()\fR are the same as those for the functions that  Given the number, you can extract the substring directly, or use one of the
1728  extract by number, and so are not re-described here. There are just two  functions described in the previous section. For convenience, there are also
1729  differences.  two functions that do the whole job.
1730    .P
1731    Most of the arguments of \fBpcre_copy_named_substring()\fP and
1732    \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1733    functions that extract by number. As these are described in the previous
1734    section, they are not re-described here. There are just two differences:
1735    .P
1736  First, instead of a substring number, a substring name is given. Second, there  First, instead of a substring number, a substring name is given. Second, there
1737  is an extra argument, given at the start, which is a pointer to the compiled  is an extra argument, given at the start, which is a pointer to the compiled
1738  pattern. This is needed in order to gain access to the name-to-number  pattern. This is needed in order to gain access to the name-to-number
1739  translation table.  translation table.
1740    .P
1741  These functions call \fBpcre_get_stringnumber()\fR, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1742  then call \fIpcre_copy_substring()\fR or \fIpcre_get_substring()\fR, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1743  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1744    the behaviour may not be what you want (see the next section).
1745  .in 0  .P
1746  Last updated: 03 February 2003  \fBWarning:\fP If the pattern uses the "(?|" feature to set up multiple
1747  .br  subpatterns with the same number, you cannot use names to distinguish them,
1748  Copyright (c) 1997-2003 University of Cambridge.  because names are not included in the compiled code. The matching process uses
1749    only numbers.
1750    .
1751    .SH "DUPLICATE SUBPATTERN NAMES"
1752    .rs
1753    .sp
1754    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1755    .ti +5n
1756    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1757    .PP
1758    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1759    are not required to be unique. Normally, patterns with duplicate names are such
1760    that in any one match, only one of the named subpatterns participates. An
1761    example is shown in the
1762    .\" HREF
1763    \fBpcrepattern\fP
1764    .\"
1765    documentation.
1766    .P
1767    When duplicates are present, \fBpcre_copy_named_substring()\fP and
1768    \fBpcre_get_named_substring()\fP return the first substring corresponding to
1769    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1770    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1771    returns one of the numbers that are associated with the name, but it is not
1772    defined which it is.
1773    .P
1774    If you want to get full details of all captured substrings for a given name,
1775    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1776    argument is the compiled pattern, and the second is the name. The third and
1777    fourth are pointers to variables which are updated by the function. After it
1778    has run, they point to the first and last entries in the name-to-number table
1779    for the given name. The function itself returns the length of each entry, or
1780    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1781    described above in the section entitled \fIInformation about a pattern\fP.
1782    Given all the relevant entries for the name, you can extract each of their
1783    numbers, and hence the captured data, if any.
1784    .
1785    .
1786    .SH "FINDING ALL POSSIBLE MATCHES"
1787    .rs
1788    .sp
1789    The traditional matching function uses a similar algorithm to Perl, which stops
1790    when it finds the first match, starting at a given point in the subject. If you
1791    want to find all possible matches, or the longest possible match, consider
1792    using the alternative matching function (see below) instead. If you cannot use
1793    the alternative function, but still need to find all possible matches, you
1794    can kludge it up by making use of the callout facility, which is described in
1795    the
1796    .\" HREF
1797    \fBpcrecallout\fP
1798    .\"
1799    documentation.
1800    .P
1801    What you have to do is to insert a callout right at the end of the pattern.
1802    When your callout function is called, extract and save the current matched
1803    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1804    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1805    will yield PCRE_ERROR_NOMATCH.
1806    .
1807    .
1808    .\" HTML <a name="dfamatch"></a>
1809    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1810    .rs
1811    .sp
1812    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1813    .ti +5n
1814    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1815    .ti +5n
1816    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1817    .ti +5n
1818    .B int *\fIworkspace\fP, int \fIwscount\fP);
1819    .P
1820    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1821    a compiled pattern, using a matching algorithm that scans the subject string
1822    just once, and does not backtrack. This has different characteristics to the
1823    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1824    patterns are not supported. Nevertheless, there are times when this kind of
1825    matching can be useful. For a discussion of the two matching algorithms, see
1826    the
1827    .\" HREF
1828    \fBpcrematching\fP
1829    .\"
1830    documentation.
1831    .P
1832    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1833    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1834    different way, and this is described below. The other common arguments are used
1835    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1836    here.
1837    .P
1838    The two additional arguments provide workspace for the function. The workspace
1839    vector should contain at least 20 elements. It is used for keeping track of
1840    multiple paths through the pattern tree. More workspace will be needed for
1841    patterns and subjects where there are a lot of potential matches.
1842    .P
1843    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1844    .sp
1845      int rc;
1846      int ovector[10];
1847      int wspace[20];
1848      rc = pcre_dfa_exec(
1849        re,             /* result of pcre_compile() */
1850        NULL,           /* we didn't study the pattern */
1851        "some string",  /* the subject string */
1852        11,             /* the length of the subject string */
1853        0,              /* start at offset 0 in the subject */
1854        0,              /* default options */
1855        ovector,        /* vector of integers for substring information */
1856        10,             /* number of elements (NOT size in bytes) */
1857        wspace,         /* working space vector */
1858        20);            /* number of elements (NOT size in bytes) */
1859    .
1860    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1861    .rs
1862    .sp
1863    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1864    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1865    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1866    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1867    exactly the same as for \fBpcre_exec()\fP, so their description is not repeated
1868    here.
1869    .sp
1870      PCRE_PARTIAL
1871    .sp
1872    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1873    details are slightly different. When PCRE_PARTIAL is set for
1874    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1875    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1876    complete matches, but there is still at least one matching possibility. The
1877    portion of the string that provided the longest partial match is set as the
1878    first matching string.
1879    .sp
1880      PCRE_DFA_SHORTEST
1881    .sp
1882    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1883    soon as it has found one match. Because of the way the alternative algorithm
1884    works, this is necessarily the shortest possible match at the first possible
1885    matching point in the subject string.
1886    .sp
1887      PCRE_DFA_RESTART
1888    .sp
1889    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1890    a partial match, it is possible to call it again, with additional subject
1891    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1892    option requests this action; when it is set, the \fIworkspace\fP and
1893    \fIwscount\fP options must reference the same vector as before because data
1894    about the match so far is left in them after a partial match. There is more
1895    discussion of this facility in the
1896    .\" HREF
1897    \fBpcrepartial\fP
1898    .\"
1899    documentation.
1900    .
1901    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1902    .rs
1903    .sp
1904    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1905    substring in the subject. Note, however, that all the matches from one run of
1906    the function start at the same point in the subject. The shorter matches are
1907    all initial substrings of the longer matches. For example, if the pattern
1908    .sp
1909      <.*>
1910    .sp
1911    is matched against the string
1912    .sp
1913      This is <something> <something else> <something further> no more
1914    .sp
1915    the three matched strings are
1916    .sp
1917      <something>
1918      <something> <something else>
1919      <something> <something else> <something further>
1920    .sp
1921    On success, the yield of the function is a number greater than zero, which is
1922    the number of matched substrings. The substrings themselves are returned in
1923    \fIovector\fP. Each string uses two elements; the first is the offset to the
1924    start, and the second is the offset to the end. In fact, all the strings have
1925    the same start offset. (Space could have been saved by giving this only once,
1926    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1927    returns data, even though the meaning of the strings is different.)
1928    .P
1929    The strings are returned in reverse order of length; that is, the longest
1930    matching string is given first. If there were too many matches to fit into
1931    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1932    the longest matches.
1933    .
1934    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1935    .rs
1936    .sp
1937    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1938    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1939    described
1940    .\" HTML <a href="#errorlist">
1941    .\" </a>
1942    above.
1943    .\"
1944    There are in addition the following errors that are specific to
1945    \fBpcre_dfa_exec()\fP:
1946    .sp
1947      PCRE_ERROR_DFA_UITEM      (-16)
1948    .sp
1949    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1950    that it does not support, for instance, the use of \eC or a back reference.
1951    .sp
1952      PCRE_ERROR_DFA_UCOND      (-17)
1953    .sp
1954    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1955    uses a back reference for the condition, or a test for recursion in a specific
1956    group. These are not supported.
1957    .sp
1958      PCRE_ERROR_DFA_UMLIMIT    (-18)
1959    .sp
1960    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1961    block that contains a setting of the \fImatch_limit\fP field. This is not
1962    supported (it is meaningless).
1963    .sp
1964      PCRE_ERROR_DFA_WSSIZE     (-19)
1965    .sp
1966    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1967    \fIworkspace\fP vector.
1968    .sp
1969      PCRE_ERROR_DFA_RECURSE    (-20)
1970    .sp
1971    When a recursive subpattern is processed, the matching function calls itself
1972    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1973    error is given if the output vector is not large enough. This should be
1974    extremely rare, as a vector of size 1000 is used.
1975    .
1976    .
1977    .SH "SEE ALSO"
1978    .rs
1979    .sp
1980    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1981    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1982    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1983    .
1984    .
1985    .SH AUTHOR
1986    .rs
1987    .sp
1988    .nf
1989    Philip Hazel
1990    University Computing Service
1991    Cambridge CB2 3QH, England.
1992    .fi
1993    .
1994    .
1995    .SH REVISION
1996    .rs
1997    .sp
1998    .nf
1999    Last updated: 26 August 2009
2000    Copyright (c) 1997-2009 University of Cambridge.
2001    .fi

Legend:
Removed from v.65  
changed lines
  Added in v.426

  ViewVC Help
Powered by ViewVC 1.1.5