/[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 454 by ph10, Tue Sep 22 09:42:11 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 PCRE
139    source distribution. A listing of this program is given in the
140  \fBpcre_free_substring()\fR and \fBpcre_free_substring_list()\fR are also  .\" HREF
141    \fBpcredemo\fP
142    .\"
143    documentation, and the
144    .\" HREF
145    \fBpcresample\fP
146    .\"
147    documentation describes how to compile and run it.
148    .P
149    A second matching function, \fBpcre_dfa_exec()\fP, which is not
150    Perl-compatible, is also provided. This uses a different algorithm for the
151    matching. The alternative algorithm finds all possible matches (at a given
152    point in the subject), and scans the subject just once (unless there are
153    lookbehind assertions). However, this algorithm does not return captured
154    substrings. A description of the two matching algorithms and their advantages
155    and disadvantages is given in the
156    .\" HREF
157    \fBpcrematching\fP
158    .\"
159    documentation.
160    .P
161    In addition to the main compiling and matching functions, there are convenience
162    functions for extracting captured substrings from a subject string that is
163    matched by \fBpcre_exec()\fP. They are:
164    .sp
165      \fBpcre_copy_substring()\fP
166      \fBpcre_copy_named_substring()\fP
167      \fBpcre_get_substring()\fP
168      \fBpcre_get_named_substring()\fP
169      \fBpcre_get_substring_list()\fP
170      \fBpcre_get_stringnumber()\fP
171      \fBpcre_get_stringtable_entries()\fP
172    .sp
173    \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
174  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
175    .P
176  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
177  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,
178    or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
179  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
180  compiled pattern; \fBpcre_info()\fR is an obsolete version which returns only  internal tables that are generated when PCRE is built are used.
181    .P
182    The function \fBpcre_fullinfo()\fP is used to find out information about a
183    compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
184  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
185  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
186  version of PCRE and its date of release.  version of PCRE and its date of release.
187    .P
188  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
189  the entry points of the standard \fBmalloc()\fR and \fBfree()\fR functions  containing a compiled pattern. This is provided for the benefit of
190    object-oriented applications.
191    .P
192    The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
193    the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
194  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
195  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
196  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
197    .P
198  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
199    indirections to memory management functions. These special functions are used
200    only when PCRE is compiled to use the heap for remembering data, instead of
201    recursive function calls, when running the \fBpcre_exec()\fP function. See the
202    .\" HREF
203    \fBpcrebuild\fP
204    .\"
205    documentation for details of how to do this. It is a non-standard way of
206    building PCRE, for use in environments that have limited stacks. Because of the
207    greater use of memory management, it runs more slowly. Separate functions are
208    provided so that special-purpose external code can be used for this case. When
209    used, these functions are always called in a stack-like manner (last obtained,
210    first freed), and always for memory blocks of the same size. There is a
211    discussion about PCRE's stack usage in the
212    .\" HREF
213    \fBpcrestack\fP
214    .\"
215    documentation.
216    .P
217    The global variable \fBpcre_callout\fP initially contains NULL. It can be set
218  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
219  points during a matching operation. Details are given in the \fBpcrecallout\fR  points during a matching operation. Details are given in the
220    .\" HREF
221    \fBpcrecallout\fP
222    .\"
223  documentation.  documentation.
224    .
225    .
226    .\" HTML <a name="newlines"></a>
227    .SH NEWLINES
228    .rs
229    .sp
230    PCRE supports five different conventions for indicating line breaks in
231    strings: a single CR (carriage return) character, a single LF (linefeed)
232    character, the two-character sequence CRLF, any of the three preceding, or any
233    Unicode newline sequence. The Unicode newline sequences are the three just
234    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
235    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
236    (paragraph separator, U+2029).
237    .P
238    Each of the first three conventions is used by at least one operating system as
239    its standard newline sequence. When PCRE is built, a default can be specified.
240    The default default is LF, which is the Unix standard. When PCRE is run, the
241    default can be overridden, either when a pattern is compiled, or when it is
242    matched.
243    .P
244    At compile time, the newline convention can be specified by the \fIoptions\fP
245    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
246    start of the pattern itself; this overrides any other settings. See the
247    .\" HREF
248    \fBpcrepattern\fP
249    .\"
250    page for details of the special character sequences.
251    .P
252    In the PCRE documentation the word "newline" is used to mean "the character or
253    pair of characters that indicate a line break". The choice of newline
254    convention affects the handling of the dot, circumflex, and dollar
255    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
256    recognized line ending sequence, the match position advancement for a
257    non-anchored pattern. There is more detail about this in the
258    .\" HTML <a href="#execoptions">
259    .\" </a>
260    section on \fBpcre_exec()\fP options
261    .\"
262    below.
263    .P
264    The choice of newline convention does not affect the interpretation of
265    the \en or \er escape sequences, nor does it affect what \eR matches, which is
266    controlled in a similar way, but by separate options.
267    .
268    .
269  .SH MULTITHREADING  .SH MULTITHREADING
270  .rs  .rs
271  .sp  .sp
272  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
273  proviso that the memory management functions pointed to by \fBpcre_malloc\fR  proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
274  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
275  are shared by all threads.  callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
276    .P
277  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
278  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.
279    .
280  .SH CHECKING BUILD-TIME OPTIONS  .
281    .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
282  .rs  .rs
283  .sp  .sp
284  .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
285    time, possibly by a different program, and even on a host other than the one on
286    which it was compiled. Details are given in the
287    .\" HREF
288    \fBpcreprecompile\fP
289    .\"
290    documentation. However, compiling a regular expression with one version of PCRE
291    for use with a different version is not guaranteed to work and may cause
292    crashes.
293    .
294    .
295    .SH "CHECKING BUILD-TIME OPTIONS"
296    .rs
297    .sp
298    .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
299  .PP  .PP
300  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
301  discover which optional features have been compiled into the PCRE library. The  discover which optional features have been compiled into the PCRE library. The
302  .\" HREF  .\" HREF
303  \fBpcrebuild\fR  \fBpcrebuild\fP
304  .\"  .\"
305  documentation has more details about these optional features.  documentation has more details about these optional features.
306    .P
307  The first argument for \fBpcre_config()\fR is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
308  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
309  which the information is placed. The following information is available:  which the information is placed. The following information is available:
310    .sp
311    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
312    .sp
313  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;
314  otherwise it is set to zero.  otherwise it is set to zero.
315    .sp
316      PCRE_CONFIG_UNICODE_PROPERTIES
317    .sp
318    The output is an integer that is set to one if support for Unicode character
319    properties is available; otherwise it is set to zero.
320    .sp
321    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
322    .sp
323  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
324  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
325  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.
326    Though they are derived from ASCII, the same values are returned in EBCDIC
327    environments. The default should normally correspond to the standard sequence
328    for your operating system.
329    .sp
330      PCRE_CONFIG_BSR
331    .sp
332    The output is an integer whose value indicates what character sequences the \eR
333    escape sequence matches by default. A value of 0 means that \eR matches any
334    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
335    or CRLF. The default can be overridden when a pattern is compiled or matched.
336    .sp
337    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
338    .sp
339  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
340  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
341  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
342  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
343  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.
344    .sp
345    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
346    .sp
347  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
348  interface uses \fBmalloc()\fR for output vectors. Further details are given in  interface uses \fBmalloc()\fP for output vectors. Further details are given in
349  the \fBpcreposix\fR documentation.  the
350    .\" HREF
351    \fBpcreposix\fP
352    .\"
353    documentation.
354    .sp
355    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
356    .sp
357  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
358  internal matching function calls in a \fBpcre_exec()\fR execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
359  details are given with \fBpcre_exec()\fR below.  details are given with \fBpcre_exec()\fP below.
360    .sp
361  .SH COMPILING A PATTERN    PCRE_CONFIG_MATCH_LIMIT_RECURSION
362  .rs  .sp
363  .sp  The output is a long integer that gives the default limit for the depth of
364  .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  recursion when calling the internal matching function in a \fBpcre_exec()\fP
365  .ti +5n  execution. Further details are given with \fBpcre_exec()\fP below.
366  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .sp
367  .ti +5n    PCRE_CONFIG_STACKRECURSE
368  .B const unsigned char *\fItableptr\fR);  .sp
369  .PP  The output is an integer that is set to one if internal recursion when running
370    \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
371  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
372  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
373  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
374  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
375  code and related data. The \fBpcre\fR type is defined for the returned block;  avoiding the use of the stack.
376  this is a typedef for a structure whose contents are not externally defined. It  .
377  is up to the caller to free the memory when it is no longer required.  .
378    .SH "COMPILING A PATTERN"
379    .rs
380    .sp
381    .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
382    .ti +5n
383    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
384    .ti +5n
385    .B const unsigned char *\fItableptr\fP);
386    .sp
387    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
388    .ti +5n
389    .B int *\fIerrorcodeptr\fP,
390    .ti +5n
391    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
392    .ti +5n
393    .B const unsigned char *\fItableptr\fP);
394    .P
395    Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
396    called to compile a pattern into an internal form. The only difference between
397    the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
398    \fIerrorcodeptr\fP, via which a numerical error code can be returned.
399    .P
400    The pattern is a C string terminated by a binary zero, and is passed in the
401    \fIpattern\fP argument. A pointer to a single block of memory that is obtained
402    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
403    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
404    for a structure whose contents are not externally defined. It is up to the
405    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
406    .P
407  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
408  depend on memory location, the complete \fBpcre\fR data block is not  depend on memory location, the complete \fBpcre\fP data block is not
409  fully relocatable, because it contains a copy of the \fItableptr\fR argument,  fully relocatable, because it may contain a copy of the \fItableptr\fP
410  which is an address (see below).  argument, which is an address (see below).
411    .P
412  The \fIoptions\fR argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
413  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
414  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
415  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
416  in the \fBpcrepattern\fR documentation). For these options, the contents of the  within the pattern (see the detailed description in the
417  \fIoptions\fR argument specifies their initial settings at the start of  .\" HREF
418  compilation and execution. The PCRE_ANCHORED option can be set at the time of  \fBpcrepattern\fP
419  matching as well as at compile time.  .\"
420    documentation). For those options that can be different in different parts of
421  If \fIerrptr\fR is NULL, \fBpcre_compile()\fR returns NULL immediately.  the pattern, the contents of the \fIoptions\fP argument specifies their initial
422  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fR returns  settings at the start of compilation and execution. The PCRE_ANCHORED and
423  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
424  error message. The offset from the start of the pattern to the character where  compile time.
425  the error was discovered is placed in the variable pointed to by  .P
426  \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.
427    Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
428  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
429  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
430  locale. Otherwise, \fItableptr\fR must be the result of a call to  not try to free it. The byte offset from the start of the pattern to the
431  \fBpcre_maketables()\fR. See the section on locale support below.  character that was being processes when the error was discovered is placed in
432    the variable pointed to by \fIerroffset\fP, which must not be NULL. If it is,
433  This code fragment shows a typical straightforward call to \fBpcre_compile()\fR:  an immediate error is given. Some errors are not detected until checks are
434    carried out when the whole pattern has been scanned; in this case the offset is
435    set to the end of the pattern.
436    .P
437    If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
438    \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
439    returned via this argument in the event of an error. This is in addition to the
440    textual error message. Error codes and messages are listed below.
441    .P
442    If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
443    character tables that are built when PCRE is compiled, using the default C
444    locale. Otherwise, \fItableptr\fP must be an address that is the result of a
445    call to \fBpcre_maketables()\fP. This value is stored with the compiled
446    pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
447    passed to it. For more discussion, see the section on locale support below.
448    .P
449    This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
450    .sp
451    pcre *re;    pcre *re;
452    const char *error;    const char *error;
453    int erroffset;    int erroffset;
# Line 265  This code fragment shows a typical strai Line 457  This code fragment shows a typical strai
457      &error,           /* for error message */      &error,           /* for error message */
458      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
459      NULL);            /* use default character tables */      NULL);            /* use default character tables */
460    .sp
461  The following option bits are defined:  The following names for option bits are defined in the \fBpcre.h\fP header
462    file:
463    .sp
464    PCRE_ANCHORED    PCRE_ANCHORED
465    .sp
466  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
467  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
468  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
469  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
470  Perl.  Perl.
471    .sp
472      PCRE_AUTO_CALLOUT
473    .sp
474    If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
475    all with number 255, before each pattern item. For discussion of the callout
476    facility, see the
477    .\" HREF
478    \fBpcrecallout\fP
479    .\"
480    documentation.
481    .sp
482      PCRE_BSR_ANYCRLF
483      PCRE_BSR_UNICODE
484    .sp
485    These options (which are mutually exclusive) control what the \eR escape
486    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
487    match any Unicode newline sequence. The default is specified when PCRE is
488    built. It can be overridden from within the pattern, or by setting an option
489    when a compiled pattern is matched.
490    .sp
491    PCRE_CASELESS    PCRE_CASELESS
492    .sp
493  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
494  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
495  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
496    concept of case for characters whose values are less than 128, so caseless
497    matching is always possible. For characters with higher values, the concept of
498    case is supported if PCRE is compiled with Unicode property support, but not
499    otherwise. If you want to use caseless matching for characters 128 and above,
500    you must ensure that PCRE is compiled with Unicode property support as well as
501    with UTF-8 support.
502    .sp
503    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
504    .sp
505  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
506  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
507  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
508  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.
509  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
510  a pattern.  pattern.
511    .sp
512    PCRE_DOTALL    PCRE_DOTALL
513    .sp
514  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,
515  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
516  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
517  (?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
518  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
519    the setting of this option.
520    .sp
521      PCRE_DUPNAMES
522    .sp
523    If this bit is set, names used to identify capturing subpatterns need not be
524    unique. This can be helpful for certain types of pattern when it is known that
525    only one instance of the named subpattern can ever be matched. There are more
526    details of named subpatterns below; see also the
527    .\" HREF
528    \fBpcrepattern\fP
529    .\"
530    documentation.
531    .sp
532    PCRE_EXTENDED    PCRE_EXTENDED
533    .sp
534  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
535  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
536  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
537  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
538  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
539  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
540    .P
541  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
542  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
543  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
544  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
545    .sp
546    PCRE_EXTRA    PCRE_EXTRA
547    .sp
548  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
549  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
550  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
551  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
552  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
553  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
554  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
555  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
556    .sp
557      PCRE_FIRSTLINE
558    .sp
559    If this option is set, an unanchored pattern is required to match before or at
560    the first newline in the subject string, though the matched text may continue
561    over the newline.
562    .sp
563      PCRE_JAVASCRIPT_COMPAT
564    .sp
565    If this option is set, PCRE's behaviour is changed in some ways so that it is
566    compatible with JavaScript rather than Perl. The changes are as follows:
567    .P
568    (1) A lone closing square bracket in a pattern causes a compile-time error,
569    because this is illegal in JavaScript (by default it is treated as a data
570    character). Thus, the pattern AB]CD becomes illegal when this option is set.
571    .P
572    (2) At run time, a back reference to an unset subpattern group matches an empty
573    string (by default this causes the current matching alternative to fail). A
574    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
575    an "a" in the subject), whereas it fails by default, for Perl compatibility.
576    .sp
577    PCRE_MULTILINE    PCRE_MULTILINE
578    .sp
579  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
580  characters (even if it actually contains several newlines). The "start of line"  characters (even if it actually contains newlines). The "start of line"
581  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
582  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
583  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
584  Perl.  Perl.
585    .P
586  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
587  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
588  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
589  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
590  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
591  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
592    .sp
593      PCRE_NEWLINE_CR
594      PCRE_NEWLINE_LF
595      PCRE_NEWLINE_CRLF
596      PCRE_NEWLINE_ANYCRLF
597      PCRE_NEWLINE_ANY
598    .sp
599    These options override the default newline definition that was chosen when PCRE
600    was built. Setting the first or the second specifies that a newline is
601    indicated by a single character (CR or LF, respectively). Setting
602    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
603    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
604    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
605    that any Unicode newline sequence should be recognized. The Unicode newline
606    sequences are the three just mentioned, plus the single characters VT (vertical
607    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
608    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
609    recognized only in UTF-8 mode.
610    .P
611    The newline setting in the options word uses three bits that are treated
612    as a number, giving eight possibilities. Currently only six are used (default
613    plus the five values above). This means that if you set more than one newline
614    option, the combination may or may not be sensible. For example,
615    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
616    other combinations may yield unused numbers and cause an error.
617    .P
618    The only time that a line break is specially recognized when compiling a
619    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
620    class is encountered. This indicates a comment that lasts until after the next
621    line break sequence. In other circumstances, line break sequences are treated
622    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
623    as whitespace characters and are therefore ignored.
624    .P
625    The newline option that is set at compile time becomes the default that is used
626    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
627    .sp
628    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
629    .sp
630  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
631  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
632  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
633  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
634  in Perl.  in Perl.
635    .sp
636    PCRE_UNGREEDY    PCRE_UNGREEDY
637    .sp
638  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
639  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
640  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.
641    .sp
642    PCRE_UTF8    PCRE_UTF8
643    .sp
644  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
645  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
646  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
647  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
648  behaviour of PCRE are given in the  behaviour of PCRE are given in the
649  .\" HTML <a href="pcre.html#utf8support">  .\" HTML <a href="pcre.html#utf8support">
# Line 367  section on UTF-8 support Line 652  section on UTF-8 support
652  .\"  .\"
653  in the main  in the main
654  .\" HREF  .\" HREF
655  \fBpcre\fR  \fBpcre\fP
656  .\"  .\"
657  page.  page.
658    .sp
659  .SH STUDYING A PATTERN    PCRE_NO_UTF8_CHECK
660    .sp
661    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
662    automatically checked. There is a discussion about the
663    .\" HTML <a href="pcre.html#utf8strings">
664    .\" </a>
665    validity of UTF-8 strings
666    .\"
667    in the main
668    .\" HREF
669    \fBpcre\fP
670    .\"
671    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
672    returns an error. If you already know that your pattern is valid, and you want
673    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
674    option. When it is set, the effect of passing an invalid UTF-8 string as a
675    pattern is undefined. It may cause your program to crash. Note that this option
676    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
677    the UTF-8 validity checking of subject strings.
678    .
679    .
680    .SH "COMPILATION ERROR CODES"
681  .rs  .rs
682  .sp  .sp
683  .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  The following table lists the error codes than may be returned by
684    \fBpcre_compile2()\fP, along with the error messages that may be returned by
685    both compiling functions. As PCRE has developed, some error codes have fallen
686    out of use. To avoid confusion, they have not been re-used.
687    .sp
688       0  no error
689       1  \e at end of pattern
690       2  \ec at end of pattern
691       3  unrecognized character follows \e
692       4  numbers out of order in {} quantifier
693       5  number too big in {} quantifier
694       6  missing terminating ] for character class
695       7  invalid escape sequence in character class
696       8  range out of order in character class
697       9  nothing to repeat
698      10  [this code is not in use]
699      11  internal error: unexpected repeat
700      12  unrecognized character after (? or (?-
701      13  POSIX named classes are supported only within a class
702      14  missing )
703      15  reference to non-existent subpattern
704      16  erroffset passed as NULL
705      17  unknown option bit(s) set
706      18  missing ) after comment
707      19  [this code is not in use]
708      20  regular expression is too large
709      21  failed to get memory
710      22  unmatched parentheses
711      23  internal error: code overflow
712      24  unrecognized character after (?<
713      25  lookbehind assertion is not fixed length
714      26  malformed number or name after (?(
715      27  conditional group contains more than two branches
716      28  assertion expected after (?(
717      29  (?R or (?[+-]digits must be followed by )
718      30  unknown POSIX class name
719      31  POSIX collating elements are not supported
720      32  this version of PCRE is not compiled with PCRE_UTF8 support
721      33  [this code is not in use]
722      34  character value in \ex{...} sequence is too large
723      35  invalid condition (?(0)
724      36  \eC not allowed in lookbehind assertion
725      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
726      38  number after (?C is > 255
727      39  closing ) for (?C expected
728      40  recursive call could loop indefinitely
729      41  unrecognized character after (?P
730      42  syntax error in subpattern name (missing terminator)
731      43  two named subpatterns have the same name
732      44  invalid UTF-8 string
733      45  support for \eP, \ep, and \eX has not been compiled
734      46  malformed \eP or \ep sequence
735      47  unknown property name after \eP or \ep
736      48  subpattern name is too long (maximum 32 characters)
737      49  too many named subpatterns (maximum 10000)
738      50  [this code is not in use]
739      51  octal value is greater than \e377 (not in UTF-8 mode)
740      52  internal error: overran compiling workspace
741      53  internal error: previously-checked referenced subpattern not found
742      54  DEFINE group contains more than one branch
743      55  repeating a DEFINE group is not allowed
744      56  inconsistent NEWLINE options
745      57  \eg is not followed by a braced, angle-bracketed, or quoted
746            name/number or by a plain number
747      58  a numbered reference must not be zero
748      59  (*VERB) with an argument is not supported
749      60  (*VERB) not recognized
750      61  number is too big
751      62  subpattern name expected
752      63  digit expected after (?+
753      64  ] is an invalid data character in JavaScript compatibility mode
754    .sp
755    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
756    be used if the limits were changed when PCRE was built.
757    .
758    .
759    .SH "STUDYING A PATTERN"
760    .rs
761    .sp
762    .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
763  .ti +5n  .ti +5n
764  .B const char **\fIerrptr\fR);  .B const char **\fIerrptr\fP);
765  .PP  .PP
766  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
767  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
768  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
769  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
770  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
771  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
772    results of the study.
773  The returned value from a \fBpcre_study()\fR can be passed directly to  .P
774  \fBpcre_exec()\fR. However, the \fBpcre_extra\fR block also contains other  The returned value from \fBpcre_study()\fP can be passed directly to
775    \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
776  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
777  described below. If studying the pattern does not produce any additional  described
778  information, \fBpcre_study()\fR returns NULL. In that circumstance, if the  .\" HTML <a href="#extradata">
779  calling program wants to pass some of the other fields to \fBpcre_exec()\fR, it  .\" </a>
780  must set up its own \fBpcre_extra\fR block.  below
781    .\"
782  The second argument contains option bits. At present, no options are defined  in the section on matching a pattern.
783  for \fBpcre_study()\fR, and this argument should always be zero.  .P
784    If studying the pattern does not produce any additional information
785  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
786    wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
787    own \fBpcre_extra\fP block.
788    .P
789    The second argument of \fBpcre_study()\fP contains option bits. At present, no
790    options are defined, and this argument should always be zero.
791    .P
792    The third argument for \fBpcre_study()\fP is a pointer for an error message. If
793  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
794  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
795  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
796  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
797    sure that it has run successfully.
798  This is a typical call to \fBpcre_study\fR():  .P
799    This is a typical call to \fBpcre_study\fP():
800    .sp
801    pcre_extra *pe;    pcre_extra *pe;
802    pe = pcre_study(    pe = pcre_study(
803      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
804      0,              /* no options exist */      0,              /* no options exist */
805      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
806    .sp
807  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
808  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
809  characters is created.  bytes is created.
810    .
811    .
812  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
813  .SH LOCALE SUPPORT  .SH "LOCALE SUPPORT"
814  .rs  .rs
815  .sp  .sp
816  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
817  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
818  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
819  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
820  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
821  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
822    characters with codes greater than 128, you should either use UTF-8 and
823  An alternative set of tables can, however, be supplied. Such tables are built  Unicode, or use locales, but not try to mix the two.
824  by calling the \fBpcre_maketables()\fR function, which has no arguments, in the  .P
825  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
826  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.
827  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
828  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
829    default "C" locale of the local system, which may cause them to be different.
830    setlocale(LC_CTYPE, "fr");  .P
831    The internal tables can always be overridden by tables supplied by the
832    application that calls PCRE. These may be created in a different locale from
833    the default. As more and more applications change to using Unicode, the need
834    for this locale support is expected to die away.
835    .P
836    External tables are built by calling the \fBpcre_maketables()\fP function,
837    which has no arguments, in the relevant locale. The result can then be passed
838    to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
839    example, to build and use tables that are appropriate for the French locale
840    (where accented characters with values greater than 128 are treated as letters),
841    the following code could be used:
842    .sp
843      setlocale(LC_CTYPE, "fr_FR");
844    tables = pcre_maketables();    tables = pcre_maketables();
845    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
846    .sp
847  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
848  pointer that is passed to \fBpcre_compile\fR is saved with the compiled  are using Windows, the name for the French locale is "french".
849  pattern, and the same tables are used via this pointer by \fBpcre_study()\fR  .P
850  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
851  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
852  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
853  memory containing the tables remains available for as long as it is needed.  needed.
854    .P
855  .SH INFORMATION ABOUT A PATTERN  The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
856    pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
857    and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
858    pattern, compilation, studying and matching all happen in the same locale, but
859    different patterns can be compiled in different locales.
860    .P
861    It is possible to pass a table pointer or NULL (indicating the use of the
862    internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
863    this facility could be used to match a pattern in a different locale from the
864    one in which it was compiled. Passing table pointers at run time is discussed
865    below in the section on matching a pattern.
866    .
867    .
868    .SH "INFORMATION ABOUT A PATTERN"
869  .rs  .rs
870  .sp  .sp
871  .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,"
872  .ti +5n  .ti +5n
873  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
874  .PP  .PP
875  The \fBpcre_fullinfo()\fR function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
876  pattern. It replaces the obsolete \fBpcre_info()\fR function, which is  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
877  nevertheless retained for backwards compability (and is documented below).  nevertheless retained for backwards compability (and is documented below).
878    .P
879  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
880  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
881  the pattern was not studied. The third argument specifies which piece of  the pattern was not studied. The third argument specifies which piece of
882  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
883  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
884  the following negative numbers:  the following negative numbers:
885    .sp
886    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
887                          the argument \fIwhere\fR was NULL                          the argument \fIwhere\fP was NULL
888    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
889    PCRE_ERROR_BADOPTION  the value of \fIwhat\fR was invalid    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid
890    .sp
891  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
892  compiled pattern:  check against passing an arbitrary memory pointer. Here is a typical call of
893    \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
894    .sp
895    int rc;    int rc;
896    unsigned long int length;    size_t length;
897    rc = pcre_fullinfo(    rc = pcre_fullinfo(
898      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
899      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
900      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
901      &length);         /* where to put the data */      &length);         /* where to put the data */
902    .sp
903  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
904  as follows:  as follows:
905    .sp
906    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
907    .sp
908  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
909  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
910  no back references.  no back references.
911    .sp
912    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
913    .sp
914  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
915  should point to an \fbint\fR variable.  should point to an \fBint\fP variable.
916    .sp
917      PCRE_INFO_DEFAULT_TABLES
918    .sp
919    Return a pointer to the internal default character tables within PCRE. The
920    fourth argument should point to an \fBunsigned char *\fP variable. This
921    information call is provided for internal use by the \fBpcre_study()\fP
922    function. External callers can cause PCRE to use its internal tables by passing
923    a NULL table pointer.
924    .sp
925    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
926    .sp
927  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
928  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
929  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
930    still recognized for backwards compatibility.)
931  If there is a fixed first byte, e.g. from a pattern such as (cat|cow|coyote),  .P
932  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
933    (cat|cow|coyote), its value is returned. Otherwise, if either
934    .sp
935  (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
936  starts with "^", or  starts with "^", or
937    .sp
938  (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
939  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
940    .sp
941  -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
942  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
943  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
944    .sp
945    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
946    .sp
947  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
948  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
949  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
950  fourth argument should point to an \fBunsigned char *\fR variable.  fourth argument should point to an \fBunsigned char *\fP variable.
951    .sp
952      PCRE_INFO_HASCRORLF
953    .sp
954    Return 1 if the pattern contains any explicit matches for CR or LF characters,
955    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
956    explicit match is either a literal CR or LF character, or \er or \en.
957    .sp
958      PCRE_INFO_JCHANGED
959    .sp
960    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
961    0. The fourth argument should point to an \fBint\fP variable. (?J) and
962    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
963    .sp
964    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
965    .sp
966  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
967  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
968  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
969  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
970  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
971  /^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
972  is -1.  is -1.
973    .sp
974    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
975    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
976    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
977    .sp
978  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
979  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
980  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
981  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
982  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
983  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
984  table for the pattern.  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
985    you need to use the name-to-number map, which is described by these three
986    values.
987    .P
988  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
989  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
990  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
991  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
992  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
993  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
994  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
995  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
996    their parentheses numbers. For example, consider the following pattern (assume
997  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
998    .sp
999    (?P<date> (?P<year>(\\d\\d)?\\d\\d) -  .\" JOIN
1000    (?P<month>\\d\\d) - (?P<day>\\d\\d) )    (?<date> (?<year>(\ed\ed)?\ed\ed) -
1001      (?<month>\ed\ed) - (?<day>\ed\ed) )
1002    .sp
1003  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
1004  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
1005  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
1006    .sp
1007    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
1008    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
1009    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
1010    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
1011    .sp
1012  When writing code to extract data from named subpatterns, remember that the  When writing code to extract data from named subpatterns using the
1013  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
1014    different for each compiled pattern.
1015    .sp
1016      PCRE_INFO_OKPARTIAL
1017    .sp
1018    Return 1 if the pattern can be used for partial matching with
1019    \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1020    \fBint\fP variable. From release 8.00, this always returns 1, because the
1021    restrictions that previously applied to partial matching have been lifted. The
1022    .\" HREF
1023    \fBpcrepartial\fP
1024    .\"
1025    documentation gives details of partial matching.
1026    .sp
1027    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1028    .sp
1029  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
1030  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
1031  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
1032  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1033    they are the options that will be in force when matching starts. For example,
1034    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1035    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1036    .P
1037  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
1038  alternatives begin with one of the following:  alternatives begin with one of the following:
1039    .sp
1040    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
1041    \\A    always    \eA    always
1042    \\G    always    \eG    always
1043    .\" JOIN
1044    .*    if PCRE_DOTALL is set and there are no back    .*    if PCRE_DOTALL is set and there are no back
1045            references to the subpattern in which .* appears            references to the subpattern in which .* appears
1046    .sp
1047  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
1048  \fBpcre_fullinfo()\fR.  \fBpcre_fullinfo()\fP.
1049    .sp
1050    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1051    .sp
1052  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
1053  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
1054  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
1055  variable.  variable.
1056    .sp
1057    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1058    .sp
1059  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
1060  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
1061  \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
1062  created by \fBpcre_study()\fR. The fourth argument should point to a  created by \fBpcre_study()\fP. The fourth argument should point to a
1063  \fBsize_t\fR variable.  \fBsize_t\fP variable.
1064    .
1065  .SH OBSOLETE INFO FUNCTION  .
1066    .SH "OBSOLETE INFO FUNCTION"
1067  .rs  .rs
1068  .sp  .sp
1069  .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1070  .B *\fIfirstcharptr\fR);  .B *\fIfirstcharptr\fP);
1071  .PP  .PP
1072  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
1073  restrictive to return all the available data about a compiled pattern. New  restrictive to return all the available data about a compiled pattern. New
1074  programs should use \fBpcre_fullinfo()\fR instead. The yield of  programs should use \fBpcre_fullinfo()\fP instead. The yield of
1075  \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
1076  following negative numbers:  following negative numbers:
1077    .sp
1078    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
1079    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1080    .sp
1081  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
1082  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1083  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
1084    .P
1085  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,
1086  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
1087  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1088    .
1089  .SH MATCHING A PATTERN  .
1090    .SH "REFERENCE COUNTS"
1091  .rs  .rs
1092  .sp  .sp
1093  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1094    .PP
1095    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1096    data block that contains a compiled pattern. It is provided for the benefit of
1097    applications that operate in an object-oriented manner, where different parts
1098    of the application may be using the same compiled pattern, but you want to free
1099    the block when they are all done.
1100    .P
1101    When a pattern is compiled, the reference count field is initialized to zero.
1102    It is changed only by calling this function, whose action is to add the
1103    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1104    function is the new value. However, the value of the count is constrained to
1105    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1106    it is forced to the appropriate limit value.
1107    .P
1108    Except when it is zero, the reference count is not correctly preserved if a
1109    pattern is compiled on one host and then transferred to a host whose byte-order
1110    is different. (This seems a highly unlikely scenario.)
1111    .
1112    .
1113    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1114    .rs
1115    .sp
1116    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1117  .ti +5n  .ti +5n
1118  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1119  .ti +5n  .ti +5n
1120  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1121  .PP  .P
1122  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
1123  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
1124  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
1125  \fIextra\fR argument.  \fIextra\fP argument. This function is the main matching facility of the
1126    library, and it operates in a Perl-like manner. For specialist use there is
1127  Here is an example of a simple call to \fBpcre_exec()\fR:  also an alternative matching function, which is described
1128    .\" HTML <a href="#dfamatch">
1129    .\" </a>
1130    below
1131    .\"
1132    in the section about the \fBpcre_dfa_exec()\fP function.
1133    .P
1134    In most applications, the pattern will have been compiled (and optionally
1135    studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1136    possible to save compiled patterns and study data, and then use them later
1137    in different processes, possibly even on different hosts. For a discussion
1138    about this, see the
1139    .\" HREF
1140    \fBpcreprecompile\fP
1141    .\"
1142    documentation.
1143    .P
1144    Here is an example of a simple call to \fBpcre_exec()\fP:
1145    .sp
1146    int rc;    int rc;
1147    int ovector[30];    int ovector[30];
1148    rc = pcre_exec(    rc = pcre_exec(
# Line 647  Here is an example of a simple call to \ Line 1152  Here is an example of a simple call to \
1152      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1153      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1154      0,              /* default options */      0,              /* default options */
1155      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1156      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1157    .
1158  If the \fIextra\fR argument is not NULL, it must point to a \fBpcre_extra\fR  .\" HTML <a name="extradata"></a>
1159  data block. The \fBpcre_study()\fR function returns such a block (when it  .SS "Extra data for \fBpcre_exec()\fR"
1160    .rs
1161    .sp
1162    If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1163    data block. The \fBpcre_study()\fP function returns such a block (when it
1164  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
1165  additional information in it. The fields in the block are as follows:  additional information in it. The \fBpcre_extra\fP block contains the following
1166    fields (not necessarily in this order):
1167    unsigned long int \fIflags\fR;  .sp
1168    void *\fIstudy_data\fR;    unsigned long int \fIflags\fP;
1169    unsigned long int \fImatch_limit\fR;    void *\fIstudy_data\fP;
1170    void *\fIcallout_data\fR;    unsigned long int \fImatch_limit\fP;
1171      unsigned long int \fImatch_limit_recursion\fP;
1172  The \fIflags\fR field is a bitmap that specifies which of the other fields    void *\fIcallout_data\fP;
1173      const unsigned char *\fItables\fP;
1174    .sp
1175    The \fIflags\fP field is a bitmap that specifies which of the other fields
1176  are set. The flag bits are:  are set. The flag bits are:
1177    .sp
1178    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1179    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1180      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1181    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1182      PCRE_EXTRA_TABLES
1183  Other flag bits should be set to zero. The \fIstudy_data\fR field is set in the  .sp
1184  \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
1185  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
1186  the block by setting the other fields.  the appropriate flag bit. You should not set this yourself, but you may add to
1187    the block by setting the other fields and their corresponding flag bits.
1188  The \fImatch_limit\fR field provides a means of preventing PCRE from using up a  .P
1189    The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1190  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,
1191  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
1192  classic example is the use of nested unlimited repeats. Internally, PCRE uses a  classic example is the use of nested unlimited repeats.
1193  function called \fBmatch()\fR which it calls repeatedly (sometimes  .P
1194  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
1195  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
1196  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
1197  count starts from zero for each position in the subject string.  limiting the amount of backtracking that can take place. For patterns that are
1198    not anchored, the count restarts from zero for each position in the subject
1199  The default limit for the library can be set when PCRE is built; the default  string.
1200    .P
1201    The default value for the limit can be set when PCRE is built; the default
1202  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
1203  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
1204  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
1205  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
1206  exceeded, \fBpcre_exec()\fR returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1207    .P
1208  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
1209  which is described in the \fBpcrecallout\fR documentation.  instead of limiting the total number of times that \fBmatch()\fP is called, it
1210    limits the depth of recursion. The recursion depth is a smaller number than the
1211  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.
1212  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.
1213  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  .P
1214  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,
1215  unachored at matching time.  when PCRE has been compiled to use memory on the heap instead of the stack, the
1216    amount of heap memory that can be used.
1217  There are also three further options that can be set only at matching time:  .P
1218    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1219    built; the default default is the same value as the default for
1220    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1221    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1222    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1223    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1224    .P
1225    The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1226    and is described in the
1227    .\" HREF
1228    \fBpcrecallout\fP
1229    .\"
1230    documentation.
1231    .P
1232    The \fItables\fP field is used to pass a character tables pointer to
1233    \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1234    pattern. A non-NULL value is stored with the compiled pattern only if custom
1235    tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1236    If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1237    internal tables to be used. This facility is helpful when re-using patterns
1238    that have been saved after compiling with an external set of tables, because
1239    the external tables might be at a different address when \fBpcre_exec()\fP is
1240    called. See the
1241    .\" HREF
1242    \fBpcreprecompile\fP
1243    .\"
1244    documentation for a discussion of saving compiled patterns for later use.
1245    .
1246    .\" HTML <a name="execoptions"></a>
1247    .SS "Option bits for \fBpcre_exec()\fP"
1248    .rs
1249    .sp
1250    The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1251    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1252    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1253    PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1254    PCRE_PARTIAL_HARD.
1255    .sp
1256      PCRE_ANCHORED
1257    .sp
1258    The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1259    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1260    to be anchored by virtue of its contents, it cannot be made unachored at
1261    matching time.
1262    .sp
1263      PCRE_BSR_ANYCRLF
1264      PCRE_BSR_UNICODE
1265    .sp
1266    These options (which are mutually exclusive) control what the \eR escape
1267    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1268    match any Unicode newline sequence. These options override the choice that was
1269    made or defaulted when the pattern was compiled.
1270    .sp
1271      PCRE_NEWLINE_CR
1272      PCRE_NEWLINE_LF
1273      PCRE_NEWLINE_CRLF
1274      PCRE_NEWLINE_ANYCRLF
1275      PCRE_NEWLINE_ANY
1276    .sp
1277    These options override the newline definition that was chosen or defaulted when
1278    the pattern was compiled. For details, see the description of
1279    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1280    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1281    the way the match position is advanced after a match failure for an unanchored
1282    pattern.
1283    .P
1284    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1285    match attempt for an unanchored pattern fails when the current position is at a
1286    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1287    characters, the match position is advanced by two characters instead of one, in
1288    other words, to after the CRLF.
1289    .P
1290    The above rule is a compromise that makes the most common cases work as
1291    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1292    set), it does not match the string "\er\enA" because, after failing at the
1293    start, it skips both the CR and the LF before retrying. However, the pattern
1294    [\er\en]A does match that string, because it contains an explicit CR or LF
1295    reference, and so advances only by one character after the first failure.
1296    .P
1297    An explicit match for CR of LF is either a literal appearance of one of those
1298    characters, or one of the \er or \en escape sequences. Implicit matches such as
1299    [^X] do not count, nor does \es (which includes CR and LF in the characters
1300    that it matches).
1301    .P
1302    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1303    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1304    .sp
1305    PCRE_NOTBOL    PCRE_NOTBOL
1306    .sp
1307  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
1308  circumflex metacharacter should not match before it. Setting this without  beginning of a line, so the circumflex metacharacter should not match before
1309  PCRE_MULTILINE (at compile time) causes circumflex never to match.  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1310    never to match. This option affects only the behaviour of the circumflex
1311    metacharacter. It does not affect \eA.
1312    .sp
1313    PCRE_NOTEOL    PCRE_NOTEOL
1314    .sp
1315  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
1316  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
1317  it. Setting this without PCRE_MULTILINE (at compile time) causes dollar never  mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1318  to match.  compile time) causes dollar never to match. This option affects only the
1319    behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1320    .sp
1321    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1322    .sp
1323  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
1324  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
1325  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
1326    .sp
1327    a?b?    a?b?
1328    .sp
1329  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 an empty
1330  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
1331  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".
1332    .sp
1333  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case    PCRE_NOTEMPTY_ATSTART
1334  of a pattern match of the empty string within its \fBsplit()\fR function, and  .sp
1335  when using the /g modifier. It is possible to emulate Perl's behaviour after  This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1336  matching a null string by first trying the match again at the same offset with  the start of the subject is permitted. If the pattern is anchored, such a match
1337  PCRE_NOTEMPTY set, and then if that fails by advancing the starting offset (see  can occur only if the pattern contains \eK.
1338  below) and trying an ordinary match again.  .P
1339    Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1340  The subject string is passed to \fBpcre_exec()\fR as a pointer in  does make a special case of a pattern match of the empty string within its
1341  \fIsubject\fR, a length in \fIlength\fR, and a starting offset in  \fBsplit()\fP function, and when using the /g modifier. It is possible to
1342  \fIstartoffset\fR. Unlike the pattern string, the subject may contain binary  emulate Perl's behaviour after matching a null string by first trying the match
1343    again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1344    if that fails, by advancing the starting offset (see below) and trying an
1345    ordinary match again. There is some code that demonstrates how to do this in
1346    the
1347    .\" HREF
1348    \fBpcredemo\fP
1349    .\"
1350    sample program.
1351    .sp
1352      PCRE_NO_START_OPTIMIZE
1353    .sp
1354    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1355    a match, in order to speed up the process. For example, if it is known that a
1356    match must start with a specific character, it searches the subject for that
1357    character, and fails immediately if it cannot find it, without actually running
1358    the main matching function. When callouts are in use, these optimizations can
1359    cause them to be skipped. This option disables the "start-up" optimizations,
1360    causing performance to suffer, but ensuring that the callouts do occur.
1361    .sp
1362      PCRE_NO_UTF8_CHECK
1363    .sp
1364    When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1365    string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1366    The value of \fIstartoffset\fP is also checked to ensure that it points to the
1367    start of a UTF-8 character. There is a discussion about the validity of UTF-8
1368    strings in the
1369    .\" HTML <a href="pcre.html#utf8strings">
1370    .\" </a>
1371    section on UTF-8 support
1372    .\"
1373    in the main
1374    .\" HREF
1375    \fBpcre\fP
1376    .\"
1377    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1378    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1379    PCRE_ERROR_BADUTF8_OFFSET is returned.
1380    .P
1381    If you already know that your subject is valid, and you want to skip these
1382    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1383    calling \fBpcre_exec()\fP. You might want to do this for the second and
1384    subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1385    all the matches in a single subject string. However, you should be sure that
1386    the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1387    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1388    subject, or a value of \fIstartoffset\fP that does not point to the start of a
1389    UTF-8 character, is undefined. Your program may crash.
1390    .sp
1391      PCRE_PARTIAL_HARD
1392      PCRE_PARTIAL_SOFT
1393    .sp
1394    These options turn on the partial matching feature. For backwards
1395    compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1396    occurs if the end of the subject string is reached successfully, but there are
1397    not enough subject characters to complete the match. If this happens when
1398    PCRE_PARTIAL_HARD is set, \fBpcre_exec()\fP immediately returns
1399    PCRE_ERROR_PARTIAL. Otherwise, if PCRE_PARTIAL_SOFT is set, matching continues
1400    by testing any other alternatives. Only if they all fail is PCRE_ERROR_PARTIAL
1401    returned (instead of PCRE_ERROR_NOMATCH). The portion of the string that
1402    was inspected when the partial match was found is set as the first matching
1403    string. There is a more detailed discussion in the
1404    .\" HREF
1405    \fBpcrepartial\fP
1406    .\"
1407    documentation.
1408    .
1409    .SS "The string to be matched by \fBpcre_exec()\fP"
1410    .rs
1411    .sp
1412    The subject string is passed to \fBpcre_exec()\fP as a pointer in
1413    \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1414    in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of
1415    a UTF-8 character. Unlike the pattern string, the subject may contain binary
1416  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
1417  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.
1418    .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.  
   
1419  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
1420  same subject by calling \fBpcre_exec()\fR again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
1421  Setting \fIstartoffset\fR differs from just passing over a shortened string and  Setting \fIstartoffset\fP differs from just passing over a shortened string and
1422  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
1423  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
1424    .sp
1425    \\Biss\\B    \eBiss\eB
1426    .sp
1427  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
1428  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
1429  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
1430  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
1431  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
1432  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
1433  \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
1434  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
1435  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.
1436    .P
1437  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
1438  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
1439  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.
1440    .
1441    .SS "How \fBpcre_exec()\fP returns captured substrings"
1442    .rs
1443    .sp
1444  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
1445  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
1446  pattern. Following the usage in Jeffrey Friedl's book, this is called  pattern. Following the usage in Jeffrey Friedl's book, this is called
1447  "capturing" in what follows, and the phrase "capturing subpattern" is used for  "capturing" in what follows, and the phrase "capturing subpattern" is used for
1448  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
1449  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1450    .P
1451  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
1452  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
1453  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
1454  back captured substrings, each substring using a pair of integers. The  argument is NOT the size of \fIovector\fP in bytes.
1455  remaining third of the vector is used as workspace by \fBpcre_exec()\fR while  .P
1456  matching capturing subpatterns, and is not available for passing back  The first two-thirds of the vector is used to pass back captured substrings,
1457  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
1458  three. If it is not, it is rounded down.  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1459    and is not available for passing back information. The number passed in
1460  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
1461  returned in pairs of integers, starting at the beginning of \fIovector\fR, and  rounded down.
1462  continuing up to two-thirds of its length at the most. The first element of a  .P
1463  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
1464  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
1465  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
1466  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
1467  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
1468  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
1469  subpatterns, the return value from a successful match is 1, indicating that  mode. They are not character counts.
1470  just the first pair of offsets has been set.  .P
1471    The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1472  Some convenience functions are provided for extracting the captured substrings  portion of the subject string matched by the entire pattern. The next pair is
1473  as separate strings. These are described in the following section.  used for the first capturing subpattern, and so on. The value returned by
1474    \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1475  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
1476  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
1477  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.
1478  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.  
   
1479  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
1480  string that it matched that gets returned.  string that it matched that is returned.
1481    .P
1482  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
1483  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
1484  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,
1485  \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
1486  \fIovecsize\fR as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1487  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
1488  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
1489  to supply an \fIovector\fR.  advisable to supply an \fIovector\fP.
1490    .P
1491  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
1492  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1493  \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
1494  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.
1495    .P
1496  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
1497    the subject when subpattern \fIn\fP has not been used at all. For example, if
1498    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1499    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1500    happens, both values in the offset pairs corresponding to unused subpatterns
1501    are set to -1.
1502    .P
1503    Offset values that correspond to unused subpatterns at the end of the
1504    expression are also set to -1. For example, if the string "abc" is matched
1505    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1506    return from the function is 2, because the highest used capturing subpattern
1507    number is 1. However, you can refer to the offsets for the second and third
1508    capturing subpatterns if you wish (assuming the vector is large enough, of
1509    course).
1510    .P
1511    Some convenience functions are provided for extracting the captured substrings
1512    as separate strings. These are described below.
1513    .
1514    .\" HTML <a name="errorlist"></a>
1515    .SS "Error return values from \fBpcre_exec()\fP"
1516    .rs
1517    .sp
1518    If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1519  defined in the header file:  defined in the header file:
1520    .sp
1521    PCRE_ERROR_NOMATCH        (-1)    PCRE_ERROR_NOMATCH        (-1)
1522    .sp
1523  The subject string did not match the pattern.  The subject string did not match the pattern.
1524    .sp
1525    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1526    .sp
1527  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
1528  NULL and \fIovecsize\fR was not zero.  NULL and \fIovecsize\fP was not zero.
1529    .sp
1530    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1531    .sp
1532  An unrecognized bit was set in the \fIoptions\fR argument.  An unrecognized bit was set in the \fIoptions\fP argument.
1533    .sp
1534    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1535    .sp
1536  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
1537  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
1538  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1539    other endianness. This is the error that PCRE gives when the magic number is
1540    PCRE_ERROR_UNKNOWN_NODE   (-5)  not present.
1541    .sp
1542      PCRE_ERROR_UNKNOWN_OPCODE (-5)
1543    .sp
1544  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1545  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
1546  of the compiled pattern.  of the compiled pattern.
1547    .sp
1548    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1549    .sp
1550  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
1551  \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
1552  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
1553  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
1554  the end of matching.  automatically freed at the end of matching.
1555    .sp
1556    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1557    .sp
1558  This error is used by the \fBpcre_copy_substring()\fR,  This error is used by the \fBpcre_copy_substring()\fP,
1559  \fBpcre_get_substring()\fR, and \fBpcre_get_substring_list()\fR functions (see  \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1560  below). It is never returned by \fBpcre_exec()\fR.  below). It is never returned by \fBpcre_exec()\fP.
1561    .sp
1562    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1563    .sp
1564  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
1565  field in a \fBpcre_extra\fR structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1566  description above.  above.
1567    .sp
1568    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1569    .sp
1570  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
1571  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
1572  \fBpcrecallout\fR documentation for details.  .\" HREF
1573    \fBpcrecallout\fP
1574  .SH EXTRACTING CAPTURED SUBSTRINGS BY NUMBER  .\"
1575    documentation for details.
1576    .sp
1577      PCRE_ERROR_BADUTF8        (-10)
1578    .sp
1579    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1580    .sp
1581      PCRE_ERROR_BADUTF8_OFFSET (-11)
1582    .sp
1583    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1584    of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1585    .sp
1586      PCRE_ERROR_PARTIAL        (-12)
1587    .sp
1588    The subject string did not match, but it did match partially. See the
1589    .\" HREF
1590    \fBpcrepartial\fP
1591    .\"
1592    documentation for details of partial matching.
1593    .sp
1594      PCRE_ERROR_BADPARTIAL     (-13)
1595    .sp
1596    This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1597    option was used with a compiled pattern containing items that were not
1598    supported for partial matching. From release 8.00 onwards, there are no
1599    restrictions on partial matching.
1600    .sp
1601      PCRE_ERROR_INTERNAL       (-14)
1602    .sp
1603    An unexpected internal error has occurred. This error could be caused by a bug
1604    in PCRE or by overwriting of the compiled pattern.
1605    .sp
1606      PCRE_ERROR_BADCOUNT       (-15)
1607    .sp
1608    This error is given if the value of the \fIovecsize\fP argument is negative.
1609    .sp
1610      PCRE_ERROR_RECURSIONLIMIT (-21)
1611    .sp
1612    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1613    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1614    description above.
1615    .sp
1616      PCRE_ERROR_BADNEWLINE     (-23)
1617    .sp
1618    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1619    .P
1620    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1621    .
1622    .
1623    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1624  .rs  .rs
1625  .sp  .sp
1626  .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1627  .ti +5n  .ti +5n
1628  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1629  .ti +5n  .ti +5n
1630  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
1631  .PP  .PP
1632  .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,  
1633  .ti +5n  .ti +5n
1634  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1635  .ti +5n  .ti +5n
1636  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1637  .PP  .PP
1638  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
1639  .ti +5n  .ti +5n
1640  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1641  .PP  .PP
1642  Captured substrings can be accessed directly by using the offsets returned by  Captured substrings can be accessed directly by using the offsets returned by
1643  \fBpcre_exec()\fR in \fIovector\fR. For convenience, the functions  \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1644  \fBpcre_copy_substring()\fR, \fBpcre_get_substring()\fR, and  \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1645  \fBpcre_get_substring_list()\fR are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1646  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1647  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1648  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1649  has a further zero added on the end, but the result is not, of course,  .P
1650  a C string.  A substring that contains a binary zero is correctly extracted and has a
1651    further zero added on the end, but the result is not, of course, a C string.
1652    However, you can process such a string by referring to the length that is
1653    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1654    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1655    for handling strings containing binary zeros, because the end of the final
1656    string is not independently indicated.
1657    .P
1658  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:
1659  \fIsubject\fR is the subject string which has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
1660  \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
1661  \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
1662  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1663  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
1664  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
1665  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
1666  the vector divided by three.  number of elements in the vector divided by three.
1667    .P
1668  The functions \fBpcre_copy_substring()\fR and \fBpcre_get_substring()\fR  The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1669  extract a single substring, whose number is given as \fIstringnumber\fR. A  extract a single substring, whose number is given as \fIstringnumber\fP. A
1670  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1671  higher values extract the captured substrings. For \fBpcre_copy_substring()\fR,  higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1672  the string is placed in \fIbuffer\fR, whose length is given by  the string is placed in \fIbuffer\fP, whose length is given by
1673  \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
1674  obtained via \fBpcre_malloc\fR, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1675  \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
1676  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1677    .sp
1678    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1679    .sp
1680  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
1681  memory failed for \fBpcre_get_substring()\fR.  memory failed for \fBpcre_get_substring()\fP.
1682    .sp
1683    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1684    .sp
1685  There is no substring whose number is \fIstringnumber\fR.  There is no substring whose number is \fIstringnumber\fP.
1686    .P
1687  The \fBpcre_get_substring_list()\fR function extracts all available substrings  The \fBpcre_get_substring_list()\fP function extracts all available substrings
1688  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
1689  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
1690  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
1691  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
1692  function is zero if all went well, or  function is zero if all went well, or the error code
1693    .sp
1694    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1695    .sp
1696  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1697    .P
1698  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
1699  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
1700  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
1701  string. This can be distinguished from a genuine zero-length substring by  string. This can be distinguished from a genuine zero-length substring by
1702  inspecting the appropriate offset in \fIovector\fR, which is negative for unset  inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1703  substrings.  substrings.
1704    .P
1705  The two convenience functions \fBpcre_free_substring()\fR and  The two convenience functions \fBpcre_free_substring()\fP and
1706  \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
1707  a previous call of \fBpcre_get_substring()\fR or  a previous call of \fBpcre_get_substring()\fP or
1708  \fBpcre_get_substring_list()\fR, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1709  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
1710  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
1711  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1712  \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
1713  provided.  provided.
1714    .
1715  .SH EXTRACTING CAPTURED SUBSTRINGS BY NAME  .
1716    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1717  .rs  .rs
1718  .sp  .sp
1719  .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1720  .ti +5n  .ti +5n
1721  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIname\fP);
1722    .PP
1723    .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1724  .ti +5n  .ti +5n
1725  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1726  .ti +5n  .ti +5n
1727  .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,  
1728  .ti +5n  .ti +5n
1729  .B const char *\fIname\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1730  .PP  .PP
1731  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
 .B int pcre_get_named_substring(const pcre *\fIcode\fR,  
1732  .ti +5n  .ti +5n
1733  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1734  .ti +5n  .ti +5n
1735  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
1736  .ti +5n  .ti +5n
1737  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1738  .PP  .PP
1739  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.
1740  can be done by calling \fBpcre_get_stringnumber()\fR. The first argument is the  For example, for this pattern
1741  compiled pattern, and the second is the name. For example, for this pattern  .sp
1742      (a+)b(?<xxx>\ed+)...
1743    ab(?<xxx>\\d+)...  .sp
1744    the number of the subpattern called "xxx" is 2. If the name is known to be
1745  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
1746  extract the substring directly, or use one of the functions described in the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1747  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
1748  whole job.  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1749    that name.
1750  Most of the arguments of \fIpcre_copy_named_substring()\fR and  .P
1751  \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
1752  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
1753  differences.  two functions that do the whole job.
1754    .P
1755    Most of the arguments of \fBpcre_copy_named_substring()\fP and
1756    \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1757    functions that extract by number. As these are described in the previous
1758    section, they are not re-described here. There are just two differences:
1759    .P
1760  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
1761  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
1762  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
1763  translation table.  translation table.
1764    .P
1765  These functions call \fBpcre_get_stringnumber()\fR, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1766  then call \fIpcre_copy_substring()\fR or \fIpcre_get_substring()\fR, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1767  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1768    the behaviour may not be what you want (see the next section).
1769  .in 0  .P
1770  Last updated: 03 February 2003  \fBWarning:\fP If the pattern uses the "(?|" feature to set up multiple
1771  .br  subpatterns with the same number, you cannot use names to distinguish them,
1772  Copyright (c) 1997-2003 University of Cambridge.  because names are not included in the compiled code. The matching process uses
1773    only numbers.
1774    .
1775    .SH "DUPLICATE SUBPATTERN NAMES"
1776    .rs
1777    .sp
1778    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1779    .ti +5n
1780    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1781    .PP
1782    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1783    are not required to be unique. Normally, patterns with duplicate names are such
1784    that in any one match, only one of the named subpatterns participates. An
1785    example is shown in the
1786    .\" HREF
1787    \fBpcrepattern\fP
1788    .\"
1789    documentation.
1790    .P
1791    When duplicates are present, \fBpcre_copy_named_substring()\fP and
1792    \fBpcre_get_named_substring()\fP return the first substring corresponding to
1793    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1794    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1795    returns one of the numbers that are associated with the name, but it is not
1796    defined which it is.
1797    .P
1798    If you want to get full details of all captured substrings for a given name,
1799    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1800    argument is the compiled pattern, and the second is the name. The third and
1801    fourth are pointers to variables which are updated by the function. After it
1802    has run, they point to the first and last entries in the name-to-number table
1803    for the given name. The function itself returns the length of each entry, or
1804    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1805    described above in the section entitled \fIInformation about a pattern\fP.
1806    Given all the relevant entries for the name, you can extract each of their
1807    numbers, and hence the captured data, if any.
1808    .
1809    .
1810    .SH "FINDING ALL POSSIBLE MATCHES"
1811    .rs
1812    .sp
1813    The traditional matching function uses a similar algorithm to Perl, which stops
1814    when it finds the first match, starting at a given point in the subject. If you
1815    want to find all possible matches, or the longest possible match, consider
1816    using the alternative matching function (see below) instead. If you cannot use
1817    the alternative function, but still need to find all possible matches, you
1818    can kludge it up by making use of the callout facility, which is described in
1819    the
1820    .\" HREF
1821    \fBpcrecallout\fP
1822    .\"
1823    documentation.
1824    .P
1825    What you have to do is to insert a callout right at the end of the pattern.
1826    When your callout function is called, extract and save the current matched
1827    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1828    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1829    will yield PCRE_ERROR_NOMATCH.
1830    .
1831    .
1832    .\" HTML <a name="dfamatch"></a>
1833    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1834    .rs
1835    .sp
1836    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1837    .ti +5n
1838    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1839    .ti +5n
1840    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1841    .ti +5n
1842    .B int *\fIworkspace\fP, int \fIwscount\fP);
1843    .P
1844    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1845    a compiled pattern, using a matching algorithm that scans the subject string
1846    just once, and does not backtrack. This has different characteristics to the
1847    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1848    patterns are not supported. Nevertheless, there are times when this kind of
1849    matching can be useful. For a discussion of the two matching algorithms, and a
1850    list of features that \fBpcre_dfa_exec()\fP does not support, see the
1851    .\" HREF
1852    \fBpcrematching\fP
1853    .\"
1854    documentation.
1855    .P
1856    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1857    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1858    different way, and this is described below. The other common arguments are used
1859    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1860    here.
1861    .P
1862    The two additional arguments provide workspace for the function. The workspace
1863    vector should contain at least 20 elements. It is used for keeping track of
1864    multiple paths through the pattern tree. More workspace will be needed for
1865    patterns and subjects where there are a lot of potential matches.
1866    .P
1867    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1868    .sp
1869      int rc;
1870      int ovector[10];
1871      int wspace[20];
1872      rc = pcre_dfa_exec(
1873        re,             /* result of pcre_compile() */
1874        NULL,           /* we didn't study the pattern */
1875        "some string",  /* the subject string */
1876        11,             /* the length of the subject string */
1877        0,              /* start at offset 0 in the subject */
1878        0,              /* default options */
1879        ovector,        /* vector of integers for substring information */
1880        10,             /* number of elements (NOT size in bytes) */
1881        wspace,         /* working space vector */
1882        20);            /* number of elements (NOT size in bytes) */
1883    .
1884    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1885    .rs
1886    .sp
1887    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1888    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1889    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1890    PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST,
1891    and PCRE_DFA_RESTART. All but the last four of these are exactly the same as
1892    for \fBpcre_exec()\fP, so their description is not repeated here.
1893    .sp
1894      PCRE_PARTIAL_HARD
1895      PCRE_PARTIAL_SOFT
1896    .sp
1897    These have the same general effect as they do for \fBpcre_exec()\fP, but the
1898    details are slightly different. When PCRE_PARTIAL_HARD is set for
1899    \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
1900    is reached and there is still at least one matching possibility that requires
1901    additional characters. This happens even if some complete matches have also
1902    been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
1903    is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
1904    there have been no complete matches, but there is still at least one matching
1905    possibility. The portion of the string that was inspected when the longest
1906    partial match was found is set as the first matching string in both cases.
1907    .sp
1908      PCRE_DFA_SHORTEST
1909    .sp
1910    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1911    soon as it has found one match. Because of the way the alternative algorithm
1912    works, this is necessarily the shortest possible match at the first possible
1913    matching point in the subject string.
1914    .sp
1915      PCRE_DFA_RESTART
1916    .sp
1917    When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
1918    again, with additional subject characters, and have it continue with the same
1919    match. The PCRE_DFA_RESTART option requests this action; when it is set, the
1920    \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
1921    before because data about the match so far is left in them after a partial
1922    match. There is more discussion of this facility in the
1923    .\" HREF
1924    \fBpcrepartial\fP
1925    .\"
1926    documentation.
1927    .
1928    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1929    .rs
1930    .sp
1931    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1932    substring in the subject. Note, however, that all the matches from one run of
1933    the function start at the same point in the subject. The shorter matches are
1934    all initial substrings of the longer matches. For example, if the pattern
1935    .sp
1936      <.*>
1937    .sp
1938    is matched against the string
1939    .sp
1940      This is <something> <something else> <something further> no more
1941    .sp
1942    the three matched strings are
1943    .sp
1944      <something>
1945      <something> <something else>
1946      <something> <something else> <something further>
1947    .sp
1948    On success, the yield of the function is a number greater than zero, which is
1949    the number of matched substrings. The substrings themselves are returned in
1950    \fIovector\fP. Each string uses two elements; the first is the offset to the
1951    start, and the second is the offset to the end. In fact, all the strings have
1952    the same start offset. (Space could have been saved by giving this only once,
1953    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1954    returns data, even though the meaning of the strings is different.)
1955    .P
1956    The strings are returned in reverse order of length; that is, the longest
1957    matching string is given first. If there were too many matches to fit into
1958    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1959    the longest matches.
1960    .
1961    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1962    .rs
1963    .sp
1964    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1965    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1966    described
1967    .\" HTML <a href="#errorlist">
1968    .\" </a>
1969    above.
1970    .\"
1971    There are in addition the following errors that are specific to
1972    \fBpcre_dfa_exec()\fP:
1973    .sp
1974      PCRE_ERROR_DFA_UITEM      (-16)
1975    .sp
1976    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1977    that it does not support, for instance, the use of \eC or a back reference.
1978    .sp
1979      PCRE_ERROR_DFA_UCOND      (-17)
1980    .sp
1981    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1982    uses a back reference for the condition, or a test for recursion in a specific
1983    group. These are not supported.
1984    .sp
1985      PCRE_ERROR_DFA_UMLIMIT    (-18)
1986    .sp
1987    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1988    block that contains a setting of the \fImatch_limit\fP field. This is not
1989    supported (it is meaningless).
1990    .sp
1991      PCRE_ERROR_DFA_WSSIZE     (-19)
1992    .sp
1993    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1994    \fIworkspace\fP vector.
1995    .sp
1996      PCRE_ERROR_DFA_RECURSE    (-20)
1997    .sp
1998    When a recursive subpattern is processed, the matching function calls itself
1999    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2000    error is given if the output vector is not large enough. This should be
2001    extremely rare, as a vector of size 1000 is used.
2002    .
2003    .
2004    .SH "SEE ALSO"
2005    .rs
2006    .sp
2007    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2008    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2009    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2010    .
2011    .
2012    .SH AUTHOR
2013    .rs
2014    .sp
2015    .nf
2016    Philip Hazel
2017    University Computing Service
2018    Cambridge CB2 3QH, England.
2019    .fi
2020    .
2021    .
2022    .SH REVISION
2023    .rs
2024    .sp
2025    .nf
2026    Last updated: 22 September 2009
2027    Copyright (c) 1997-2009 University of Cambridge.
2028    .fi

Legend:
Removed from v.65  
changed lines
  Added in v.454

  ViewVC Help
Powered by ViewVC 1.1.5