/[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 63 by nigel, Sat Feb 24 21:40:03 2007 UTC revision 312 by ph10, Wed Jan 23 18:02:23 2008 UTC
# Line 1  Line 1 
1  .TH PCRE 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH SYNOPSIS OF PCRE API  .SH "PCRE NATIVE API"
5  .rs  .rs
6  .sp  .sp
7  .B #include <pcre.h>  .B #include <pcre.h>
8  .PP  .PP
9  .SM  .SM
10  .br  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
 .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  
11  .ti +5n  .ti +5n
12  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13  .ti +5n  .ti +5n
14  .B const unsigned char *\fItableptr\fR);  .B const unsigned char *\fItableptr\fP);
15  .PP  .PP
16  .br  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
 .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  
17  .ti +5n  .ti +5n
18  .B const char **\fIerrptr\fR);  .B int *\fIerrorcodeptr\fP,
19    .ti +5n
20    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21    .ti +5n
22    .B const unsigned char *\fItableptr\fP);
23    .PP
24    .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25    .ti +5n
26    .B const char **\fIerrptr\fP);
27    .PP
28    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
29    .ti +5n
30    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31    .ti +5n
32    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33  .PP  .PP
34  .br  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .ti +5n
36    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37  .ti +5n  .ti +5n
38  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39  .ti +5n  .ti +5n
40  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int *\fIworkspace\fP, int \fIwscount\fP);
41  .PP  .PP
42  .br  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
 .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  
43  .ti +5n  .ti +5n
44  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
45  .ti +5n  .ti +5n
46  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
47  .ti +5n  .ti +5n
48  .B char *\fIbuffer\fR, int \fIbuffersize\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49  .PP  .PP
50  .br  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
 .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  
51  .ti +5n  .ti +5n
52  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53  .ti +5n  .ti +5n
54  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
55  .PP  .PP
56  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
57  .B int pcre_get_named_substring(const pcre *\fIcode\fR,  .ti +5n
58    .B const char *\fIsubject\fP, int *\fIovector\fP,
59  .ti +5n  .ti +5n
60  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
61  .ti +5n  .ti +5n
62  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char **\fIstringptr\fP);
63    .PP
64    .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65  .ti +5n  .ti +5n
66  .B const char **\fIstringptr\fR);  .B const char *\fIname\fP);
67  .PP  .PP
68  .br  .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
 .B int pcre_get_stringnumber(const pcre *\fIcode\fR,  
69  .ti +5n  .ti +5n
70  .B const char *\fIname\fR);  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71  .PP  .PP
72  .br  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
 .B int pcre_get_substring(const char *\fIsubject\fR, int *\fIovector\fR,  
73  .ti +5n  .ti +5n
74  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
75  .ti +5n  .ti +5n
76  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
77  .PP  .PP
78  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
79  .ti +5n  .ti +5n
80  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81  .PP  .PP
82  .br  .B void pcre_free_substring(const char *\fIstringptr\fP);
 .B void pcre_free_substring(const char *\fIstringptr\fR);  
83  .PP  .PP
84  .br  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
 .B void pcre_free_substring_list(const char **\fIstringptr\fR);  
85  .PP  .PP
 .br  
86  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
87  .PP  .PP
88  .br  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
 .B int pcre_fullinfo(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  
89  .ti +5n  .ti +5n
90  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
91    .PP
92    .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93    .B *\fIfirstcharptr\fP);
94  .PP  .PP
95  .br  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
 .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  
 .B *\fIfirstcharptr\fR);  
96  .PP  .PP
97  .br  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
 .B int pcre_config(int \fIwhat\fR, void *\fIwhere\fR);  
98  .PP  .PP
 .br  
99  .B char *pcre_version(void);  .B char *pcre_version(void);
100  .PP  .PP
 .br  
101  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
102  .PP  .PP
 .br  
103  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
104  .PP  .PP
105  .br  .B void *(*pcre_stack_malloc)(size_t);
106    .PP
107    .B void (*pcre_stack_free)(void *);
108    .PP
109  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
110    .
111  .SH PCRE API  .
112    .SH "PCRE API OVERVIEW"
113  .rs  .rs
114  .sp  .sp
115  PCRE has its own native API, which is described in this document. There is also  PCRE has its own native API, which is described in this document. There are
116  a set of wrapper functions that correspond to the POSIX regular expression API.  also some wrapper functions that correspond to the POSIX regular expression
117  These are described in the \fBpcreposix\fR documentation.  API. These are described in the
118    .\" HREF
119  The native API function prototypes are defined in the header file \fBpcre.h\fR,  \fBpcreposix\fP
120  and on Unix systems the library itself is called \fBlibpcre.a\fR, so can be  .\"
121  accessed by adding \fB-lpcre\fR to the command for linking an application which  documentation. Both of these APIs define a set of C function calls. A C++
122  calls it. The header file defines the macros PCRE_MAJOR and PCRE_MINOR to  wrapper is distributed with PCRE. It is documented in the
123  contain the major and minor release numbers for the library. Applications can  .\" HREF
124  use these to include support for different releases.  \fBpcrecpp\fP
125    .\"
126  The functions \fBpcre_compile()\fR, \fBpcre_study()\fR, and \fBpcre_exec()\fR  page.
127  are used for compiling and matching regular expressions. A sample program that  .P
128  demonstrates the simplest way of using them is given in the file  The native API C function prototypes are defined in the header file
129  \fIpcredemo.c\fR. The \fBpcresample\fR documentation describes how to run it.  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130    It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131  There are convenience functions for extracting captured substrings from a  an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132  matched subject string. They are:  and PCRE_MINOR to contain the major and minor release numbers for the library.
133    Applications can use these to include support for different releases of PCRE.
134    \fBpcre_copy_substring()\fR  .P
135    \fBpcre_copy_named_substring()\fR  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136    \fBpcre_get_substring()\fR  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137    \fBpcre_get_named_substring()\fR  in a Perl-compatible manner. A sample program that demonstrates the simplest
138    \fBpcre_get_substring_list()\fR  way of using them is provided in the file called \fIpcredemo.c\fP in the source
139    distribution. The
140  \fBpcre_free_substring()\fR and \fBpcre_free_substring_list()\fR are also  .\" HREF
141    \fBpcresample\fP
142    .\"
143    documentation describes how to compile and run it.
144    .P
145    A second matching function, \fBpcre_dfa_exec()\fP, which is not
146    Perl-compatible, is also provided. This uses a different algorithm for the
147    matching. The alternative algorithm finds all possible matches (at a given
148    point in the subject), and scans the subject just once. However, this algorithm
149    does not return captured substrings. A description of the two matching
150    algorithms and their advantages and disadvantages is given in the
151    .\" HREF
152    \fBpcrematching\fP
153    .\"
154    documentation.
155    .P
156    In addition to the main compiling and matching functions, there are convenience
157    functions for extracting captured substrings from a subject string that is
158    matched by \fBpcre_exec()\fP. They are:
159    .sp
160      \fBpcre_copy_substring()\fP
161      \fBpcre_copy_named_substring()\fP
162      \fBpcre_get_substring()\fP
163      \fBpcre_get_named_substring()\fP
164      \fBpcre_get_substring_list()\fP
165      \fBpcre_get_stringnumber()\fP
166      \fBpcre_get_stringtable_entries()\fP
167    .sp
168    \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
169  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
170    .P
171  The function \fBpcre_maketables()\fR is used (optionally) to build a set of  The function \fBpcre_maketables()\fP is used to build a set of character tables
172  character tables in the current locale for passing to \fBpcre_compile()\fR.  in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
173    or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
174  The function \fBpcre_fullinfo()\fR is used to find out information about a  specialist use. Most commonly, no special tables are passed, in which case
175  compiled pattern; \fBpcre_info()\fR is an obsolete version which returns only  internal tables that are generated when PCRE is built are used.
176    .P
177    The function \fBpcre_fullinfo()\fP is used to find out information about a
178    compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
179  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
180  The function \fBpcre_version()\fR returns a pointer to a string containing the  The function \fBpcre_version()\fP returns a pointer to a string containing the
181  version of PCRE and its date of release.  version of PCRE and its date of release.
182    .P
183  The global variables \fBpcre_malloc\fR and \fBpcre_free\fR initially contain  The function \fBpcre_refcount()\fP maintains a reference count in a data block
184  the entry points of the standard \fBmalloc()\fR and \fBfree()\fR functions  containing a compiled pattern. This is provided for the benefit of
185    object-oriented applications.
186    .P
187    The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
188    the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
189  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
190  so a calling program can replace them if it wishes to intercept the calls. This  so a calling program can replace them if it wishes to intercept the calls. This
191  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
192    .P
193  The global variable \fBpcre_callout\fR initially contains NULL. It can be set  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
194    indirections to memory management functions. These special functions are used
195    only when PCRE is compiled to use the heap for remembering data, instead of
196    recursive function calls, when running the \fBpcre_exec()\fP function. See the
197    .\" HREF
198    \fBpcrebuild\fP
199    .\"
200    documentation for details of how to do this. It is a non-standard way of
201    building PCRE, for use in environments that have limited stacks. Because of the
202    greater use of memory management, it runs more slowly. Separate functions are
203    provided so that special-purpose external code can be used for this case. When
204    used, these functions are always called in a stack-like manner (last obtained,
205    first freed), and always for memory blocks of the same size. There is a
206    discussion about PCRE's stack usage in the
207    .\" HREF
208    \fBpcrestack\fP
209    .\"
210    documentation.
211    .P
212    The global variable \fBpcre_callout\fP initially contains NULL. It can be set
213  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
214  points during a matching operation. Details are given in the \fBpcrecallout\fR  points during a matching operation. Details are given in the
215    .\" HREF
216    \fBpcrecallout\fP
217    .\"
218  documentation.  documentation.
219    .
220    .
221    .\" HTML <a name="newlines"></a>
222    .SH NEWLINES
223    .rs
224    .sp
225    PCRE supports five different conventions for indicating line breaks in
226    strings: a single CR (carriage return) character, a single LF (linefeed)
227    character, the two-character sequence CRLF, any of the three preceding, or any
228    Unicode newline sequence. The Unicode newline sequences are the three just
229    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
230    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
231    (paragraph separator, U+2029).
232    .P
233    Each of the first three conventions is used by at least one operating system as
234    its standard newline sequence. When PCRE is built, a default can be specified.
235    The default default is LF, which is the Unix standard. When PCRE is run, the
236    default can be overridden, either when a pattern is compiled, or when it is
237    matched.
238    .P
239    At compile time, the newline convention can be specified by the \fIoptions\fP
240    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
241    start of the pattern itself; this overrides any other settings. See the
242    .\" HREF
243    \fBpcrepattern\fP
244    .\"
245    page for details of the special character sequences.
246    .P
247    In the PCRE documentation the word "newline" is used to mean "the character or
248    pair of characters that indicate a line break". The choice of newline
249    convention affects the handling of the dot, circumflex, and dollar
250    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
251    recognized line ending sequence, the match position advancement for a
252    non-anchored pattern. There is more detail about this in the
253    .\" HTML <a href="#execoptions">
254    .\" </a>
255    section on \fBpcre_exec()\fP options
256    .\"
257    below.
258    .P
259    The choice of newline convention does not affect the interpretation of
260    the \en or \er escape sequences, nor does it affect what \eR matches, which is
261    controlled in a similar way, but by separate options.
262    .
263    .
264  .SH MULTITHREADING  .SH MULTITHREADING
265  .rs  .rs
266  .sp  .sp
267  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
268  proviso that the memory management functions pointed to by \fBpcre_malloc\fR  proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
269  and \fBpcre_free\fR, and the callout function pointed to by \fBpcre_callout\fR,  \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
270  are shared by all threads.  callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
271    .P
272  The compiled form of a regular expression is not altered during matching, so  The compiled form of a regular expression is not altered during matching, so
273  the same compiled pattern can safely be used by several threads at once.  the same compiled pattern can safely be used by several threads at once.
274    .
275  .SH CHECKING BUILD-TIME OPTIONS  .
276    .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
277  .rs  .rs
278  .sp  .sp
279  .B int pcre_config(int \fIwhat\fR, void *\fIwhere\fR);  The compiled form of a regular expression can be saved and re-used at a later
280    time, possibly by a different program, and even on a host other than the one on
281    which it was compiled. Details are given in the
282    .\" HREF
283    \fBpcreprecompile\fP
284    .\"
285    documentation. However, compiling a regular expression with one version of PCRE
286    for use with a different version is not guaranteed to work and may cause
287    crashes.
288    .
289    .
290    .SH "CHECKING BUILD-TIME OPTIONS"
291    .rs
292    .sp
293    .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
294  .PP  .PP
295  The function \fBpcre_config()\fR makes it possible for a PCRE client to  The function \fBpcre_config()\fP makes it possible for a PCRE client to
296  discover which optional features have been compiled into the PCRE library. The  discover which optional features have been compiled into the PCRE library. The
297  .\" HREF  .\" HREF
298  \fBpcrebuild\fR  \fBpcrebuild\fP
299  .\"  .\"
300  documentation has more details about these optional features.  documentation has more details about these optional features.
301    .P
302  The first argument for \fBpcre_config()\fR is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
303  information is required; the second argument is a pointer to a variable into  information is required; the second argument is a pointer to a variable into
304  which the information is placed. The following information is available:  which the information is placed. The following information is available:
305    .sp
306    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
307    .sp
308  The output is an integer that is set to one if UTF-8 support is available;  The output is an integer that is set to one if UTF-8 support is available;
309  otherwise it is set to zero.  otherwise it is set to zero.
310    .sp
311      PCRE_CONFIG_UNICODE_PROPERTIES
312    .sp
313    The output is an integer that is set to one if support for Unicode character
314    properties is available; otherwise it is set to zero.
315    .sp
316    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
317    .sp
318  The output is an integer that is set to the value of the code that is used for  The output is an integer whose value specifies the default character sequence
319  the newline character. It is either linefeed (10) or carriage return (13), and  that is recognized as meaning "newline". The four values that are supported
320  should normally be the standard character for your operating system.  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
321    default should normally be the standard sequence for your operating system.
322    .sp
323      PCRE_CONFIG_BSR
324    .sp
325    The output is an integer whose value indicates what character sequences the \eR
326    escape sequence matches by default. A value of 0 means that \eR matches any
327    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
328    or CRLF. The default can be overridden when a pattern is compiled or matched.
329    .sp
330    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
331    .sp
332  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
333  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
334  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
335  matching. The default value of 2 is sufficient for all but the most massive  matching. The default value of 2 is sufficient for all but the most massive
336  patterns, since it allows the compiled pattern to be up to 64K in size.  patterns, since it allows the compiled pattern to be up to 64K in size.
337    .sp
338    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
339    .sp
340  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
341  interface uses \fBmalloc()\fR for output vectors. Further details are given in  interface uses \fBmalloc()\fP for output vectors. Further details are given in
342  the \fBpcreposix\fR documentation.  the
343    .\" HREF
344    \fBpcreposix\fP
345    .\"
346    documentation.
347    .sp
348    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
349    .sp
350  The output is an integer that gives the default limit for the number of  The output is an integer that gives the default limit for the number of
351  internal matching function calls in a \fBpcre_exec()\fR execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
352  details are given with \fBpcre_exec()\fR below.  details are given with \fBpcre_exec()\fP below.
353    .sp
354  .SH COMPILING A PATTERN    PCRE_CONFIG_MATCH_LIMIT_RECURSION
355  .rs  .sp
356  .sp  The output is an integer that gives the default limit for the depth of
357  .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  recursion when calling the internal matching function in a \fBpcre_exec()\fP
358  .ti +5n  execution. Further details are given with \fBpcre_exec()\fP below.
359  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .sp
360  .ti +5n    PCRE_CONFIG_STACKRECURSE
361  .B const unsigned char *\fItableptr\fR);  .sp
362  .PP  The output is an integer that is set to one if internal recursion when running
363    \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
364  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
365  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
366  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
367  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
368  code and related data. The \fBpcre\fR type is defined for the returned block;  avoiding the use of the stack.
369  this is a typedef for a structure whose contents are not externally defined. It  .
370  is up to the caller to free the memory when it is no longer required.  .
371    .SH "COMPILING A PATTERN"
372    .rs
373    .sp
374    .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
375    .ti +5n
376    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
377    .ti +5n
378    .B const unsigned char *\fItableptr\fP);
379    .sp
380    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
381    .ti +5n
382    .B int *\fIerrorcodeptr\fP,
383    .ti +5n
384    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
385    .ti +5n
386    .B const unsigned char *\fItableptr\fP);
387    .P
388    Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
389    called to compile a pattern into an internal form. The only difference between
390    the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
391    \fIerrorcodeptr\fP, via which a numerical error code can be returned.
392    .P
393    The pattern is a C string terminated by a binary zero, and is passed in the
394    \fIpattern\fP argument. A pointer to a single block of memory that is obtained
395    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
396    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
397    for a structure whose contents are not externally defined. It is up to the
398    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
399    .P
400  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
401  depend on memory location, the complete \fBpcre\fR data block is not  depend on memory location, the complete \fBpcre\fP data block is not
402  fully relocatable, because it contains a copy of the \fItableptr\fR argument,  fully relocatable, because it may contain a copy of the \fItableptr\fP
403  which is an address (see below).  argument, which is an address (see below).
404    .P
405  The \fIoptions\fR argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
406  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
407  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
408  from within the pattern (see the detailed description of regular expressions  compatible with Perl, can also be set and unset from within the pattern (see
409  in the \fBpcrepattern\fR documentation). For these options, the contents of the  the detailed description in the
410  \fIoptions\fR argument specifies their initial settings at the start of  .\" HREF
411  compilation and execution. The PCRE_ANCHORED option can be set at the time of  \fBpcrepattern\fP
412    .\"
413    documentation). For these options, the contents of the \fIoptions\fP argument
414    specifies their initial settings at the start of compilation and execution. The
415    PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
416  matching as well as at compile time.  matching as well as at compile time.
417    .P
418  If \fIerrptr\fR is NULL, \fBpcre_compile()\fR returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
419  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fR returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
420  NULL, and sets the variable pointed to by \fIerrptr\fR to point to a textual  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
421  error message. The offset from the start of the pattern to the character where  error message. This is a static string that is part of the library. You must
422  the error was discovered is placed in the variable pointed to by  not try to free it. The offset from the start of the pattern to the character
423  \fIerroffset\fR, which must not be NULL. If it is, an immediate error is given.  where the error was discovered is placed in the variable pointed to by
424    \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
425  If the final argument, \fItableptr\fR, is NULL, PCRE uses a default set of  .P
426  character tables which are built when it is compiled, using the default C  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
427  locale. Otherwise, \fItableptr\fR must be the result of a call to  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
428  \fBpcre_maketables()\fR. See the section on locale support below.  returned via this argument in the event of an error. This is in addition to the
429    textual error message. Error codes and messages are listed below.
430  This code fragment shows a typical straightforward call to \fBpcre_compile()\fR:  .P
431    If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
432    character tables that are built when PCRE is compiled, using the default C
433    locale. Otherwise, \fItableptr\fP must be an address that is the result of a
434    call to \fBpcre_maketables()\fP. This value is stored with the compiled
435    pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
436    passed to it. For more discussion, see the section on locale support below.
437    .P
438    This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
439    .sp
440    pcre *re;    pcre *re;
441    const char *error;    const char *error;
442    int erroffset;    int erroffset;
# Line 265  This code fragment shows a typical strai Line 446  This code fragment shows a typical strai
446      &error,           /* for error message */      &error,           /* for error message */
447      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
448      NULL);            /* use default character tables */      NULL);            /* use default character tables */
449    .sp
450  The following option bits are defined:  The following names for option bits are defined in the \fBpcre.h\fP header
451    file:
452    .sp
453    PCRE_ANCHORED    PCRE_ANCHORED
454    .sp
455  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
456  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
457  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
458  appropriate constructs in the pattern itself, which is the only way to do it in  appropriate constructs in the pattern itself, which is the only way to do it in
459  Perl.  Perl.
460    .sp
461      PCRE_AUTO_CALLOUT
462    .sp
463    If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
464    all with number 255, before each pattern item. For discussion of the callout
465    facility, see the
466    .\" HREF
467    \fBpcrecallout\fP
468    .\"
469    documentation.
470    .sp
471      PCRE_BSR_ANYCRLF
472      PCRE_BSR_UNICODE
473    .sp
474    These options (which are mutually exclusive) control what the \eR escape
475    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
476    match any Unicode newline sequence. The default is specified when PCRE is
477    built. It can be overridden from within the pattern, or by setting an option
478    when a compiled pattern is matched.
479    .sp
480    PCRE_CASELESS    PCRE_CASELESS
481    .sp
482  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
483  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
484  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
485    concept of case for characters whose values are less than 128, so caseless
486    matching is always possible. For characters with higher values, the concept of
487    case is supported if PCRE is compiled with Unicode property support, but not
488    otherwise. If you want to use caseless matching for characters 128 and above,
489    you must ensure that PCRE is compiled with Unicode property support as well as
490    with UTF-8 support.
491    .sp
492    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
493    .sp
494  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
495  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
496  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
497  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.
498  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
499  a pattern.  pattern.
500    .sp
501    PCRE_DOTALL    PCRE_DOTALL
502    .sp
503  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,
504  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
505  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
506  (?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
507  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
508    the setting of this option.
509    .sp
510      PCRE_DUPNAMES
511    .sp
512    If this bit is set, names used to identify capturing subpatterns need not be
513    unique. This can be helpful for certain types of pattern when it is known that
514    only one instance of the named subpattern can ever be matched. There are more
515    details of named subpatterns below; see also the
516    .\" HREF
517    \fBpcrepattern\fP
518    .\"
519    documentation.
520    .sp
521    PCRE_EXTENDED    PCRE_EXTENDED
522    .sp
523  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
524  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
525  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
526  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
527  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
528  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
529    .P
530  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
531  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
532  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
533  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
534    .sp
535    PCRE_EXTRA    PCRE_EXTRA
536    .sp
537  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
538  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
539  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
540  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
541  expansion. By default, as in Perl, a backslash followed by a letter with no  expansion. By default, as in Perl, a backslash followed by a letter with no
542  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
543  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
544  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
545    .sp
546      PCRE_FIRSTLINE
547    .sp
548    If this option is set, an unanchored pattern is required to match before or at
549    the first newline in the subject string, though the matched text may continue
550    over the newline.
551    .sp
552    PCRE_MULTILINE    PCRE_MULTILINE
553    .sp
554  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
555  characters (even if it actually contains several newlines). The "start of line"  characters (even if it actually contains newlines). The "start of line"
556  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
557  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
558  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
559  Perl.  Perl.
560    .P
561  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
562  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
563  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
564  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
565  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
566  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
567    .sp
568      PCRE_NEWLINE_CR
569      PCRE_NEWLINE_LF
570      PCRE_NEWLINE_CRLF
571      PCRE_NEWLINE_ANYCRLF
572      PCRE_NEWLINE_ANY
573    .sp
574    These options override the default newline definition that was chosen when PCRE
575    was built. Setting the first or the second specifies that a newline is
576    indicated by a single character (CR or LF, respectively). Setting
577    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
578    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
579    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
580    that any Unicode newline sequence should be recognized. The Unicode newline
581    sequences are the three just mentioned, plus the single characters VT (vertical
582    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
583    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
584    recognized only in UTF-8 mode.
585    .P
586    The newline setting in the options word uses three bits that are treated
587    as a number, giving eight possibilities. Currently only six are used (default
588    plus the five values above). This means that if you set more than one newline
589    option, the combination may or may not be sensible. For example,
590    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
591    other combinations may yield unused numbers and cause an error.
592    .P
593    The only time that a line break is specially recognized when compiling a
594    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
595    class is encountered. This indicates a comment that lasts until after the next
596    line break sequence. In other circumstances, line break sequences are treated
597    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
598    as whitespace characters and are therefore ignored.
599    .P
600    The newline option that is set at compile time becomes the default that is used
601    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
602    .sp
603    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
604    .sp
605  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
606  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
607  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
608  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
609  in Perl.  in Perl.
610    .sp
611    PCRE_UNGREEDY    PCRE_UNGREEDY
612    .sp
613  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
614  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
615  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.
616    .sp
617    PCRE_UTF8    PCRE_UTF8
618    .sp
619  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
620  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
621  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
622  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
623  behaviour of PCRE are given in the  behaviour of PCRE are given in the
624  .\" HTML <a href="pcre.html#utf8support">  .\" HTML <a href="pcre.html#utf8support">
# Line 367  section on UTF-8 support Line 627  section on UTF-8 support
627  .\"  .\"
628  in the main  in the main
629  .\" HREF  .\" HREF
630  \fBpcre\fR  \fBpcre\fP
631  .\"  .\"
632  page.  page.
633    .sp
634  .SH STUDYING A PATTERN    PCRE_NO_UTF8_CHECK
635    .sp
636    When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
637    automatically checked. There is a discussion about the
638    .\" HTML <a href="pcre.html#utf8strings">
639    .\" </a>
640    validity of UTF-8 strings
641    .\"
642    in the main
643    .\" HREF
644    \fBpcre\fP
645    .\"
646    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
647    returns an error. If you already know that your pattern is valid, and you want
648    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
649    option. When it is set, the effect of passing an invalid UTF-8 string as a
650    pattern is undefined. It may cause your program to crash. Note that this option
651    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
652    the UTF-8 validity checking of subject strings.
653    .
654    .
655    .SH "COMPILATION ERROR CODES"
656    .rs
657    .sp
658    The following table lists the error codes than may be returned by
659    \fBpcre_compile2()\fP, along with the error messages that may be returned by
660    both compiling functions. As PCRE has developed, some error codes have fallen
661    out of use. To avoid confusion, they have not been re-used.
662    .sp
663       0  no error
664       1  \e at end of pattern
665       2  \ec at end of pattern
666       3  unrecognized character follows \e
667       4  numbers out of order in {} quantifier
668       5  number too big in {} quantifier
669       6  missing terminating ] for character class
670       7  invalid escape sequence in character class
671       8  range out of order in character class
672       9  nothing to repeat
673      10  [this code is not in use]
674      11  internal error: unexpected repeat
675      12  unrecognized character after (? or (?-
676      13  POSIX named classes are supported only within a class
677      14  missing )
678      15  reference to non-existent subpattern
679      16  erroffset passed as NULL
680      17  unknown option bit(s) set
681      18  missing ) after comment
682      19  [this code is not in use]
683      20  regular expression is too large
684      21  failed to get memory
685      22  unmatched parentheses
686      23  internal error: code overflow
687      24  unrecognized character after (?<
688      25  lookbehind assertion is not fixed length
689      26  malformed number or name after (?(
690      27  conditional group contains more than two branches
691      28  assertion expected after (?(
692      29  (?R or (?[+-]digits must be followed by )
693      30  unknown POSIX class name
694      31  POSIX collating elements are not supported
695      32  this version of PCRE is not compiled with PCRE_UTF8 support
696      33  [this code is not in use]
697      34  character value in \ex{...} sequence is too large
698      35  invalid condition (?(0)
699      36  \eC not allowed in lookbehind assertion
700      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
701      38  number after (?C is > 255
702      39  closing ) for (?C expected
703      40  recursive call could loop indefinitely
704      41  unrecognized character after (?P
705      42  syntax error in subpattern name (missing terminator)
706      43  two named subpatterns have the same name
707      44  invalid UTF-8 string
708      45  support for \eP, \ep, and \eX has not been compiled
709      46  malformed \eP or \ep sequence
710      47  unknown property name after \eP or \ep
711      48  subpattern name is too long (maximum 32 characters)
712      49  too many named subpatterns (maximum 10000)
713      50  [this code is not in use]
714      51  octal value is greater than \e377 (not in UTF-8 mode)
715      52  internal error: overran compiling workspace
716      53  internal error: previously-checked referenced subpattern not found
717      54  DEFINE group contains more than one branch
718      55  repeating a DEFINE group is not allowed
719      56  inconsistent NEWLINE options
720      57  \eg is not followed by a braced name or an optionally braced
721            non-zero number
722      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
723      59  (*VERB) with an argument is not supported
724      60  (*VERB) not recognized
725      61  number is too big
726      62  subpattern name expected
727      63  digit expected after (?+
728    .sp
729    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
730    be used if the limits were changed when PCRE was built.
731    .
732    .
733    .SH "STUDYING A PATTERN"
734  .rs  .rs
735  .sp  .sp
736  .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
737  .ti +5n  .ti +5n
738  .B const char **\fIerrptr\fR);  .B const char **\fIerrptr\fP);
739  .PP  .PP
740  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
741  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
742  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
743  argument. If studing the pattern produces additional information that will help  argument. If studying the pattern produces additional information that will
744  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
745  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
746    results of the study.
747  The returned value from a \fBpcre_study()\fR can be passed directly to  .P
748  \fBpcre_exec()\fR. However, the \fBpcre_extra\fR block also contains other  The returned value from \fBpcre_study()\fP can be passed directly to
749    \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
750  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
751  described below. If studying the pattern does not produce any additional  described
752  information, \fBpcre_study()\fR returns NULL. In that circumstance, if the  .\" HTML <a href="#extradata">
753  calling program wants to pass some of the other fields to \fBpcre_exec()\fR, it  .\" </a>
754  must set up its own \fBpcre_extra\fR block.  below
755    .\"
756  The second argument contains option bits. At present, no options are defined  in the section on matching a pattern.
757  for \fBpcre_study()\fR, and this argument should always be zero.  .P
758    If studying the pattern does not produce any additional information
759  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
760    wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
761    own \fBpcre_extra\fP block.
762    .P
763    The second argument of \fBpcre_study()\fP contains option bits. At present, no
764    options are defined, and this argument should always be zero.
765    .P
766    The third argument for \fBpcre_study()\fP is a pointer for an error message. If
767  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
768  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
769  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
770  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
771    sure that it has run successfully.
772  This is a typical call to \fBpcre_study\fR():  .P
773    This is a typical call to \fBpcre_study\fP():
774    .sp
775    pcre_extra *pe;    pcre_extra *pe;
776    pe = pcre_study(    pe = pcre_study(
777      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
778      0,              /* no options exist */      0,              /* no options exist */
779      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
780    .sp
781  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
782  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
783  characters is created.  bytes is created.
784    .
785    .
786  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
787  .SH LOCALE SUPPORT  .SH "LOCALE SUPPORT"
788  .rs  .rs
789  .sp  .sp
790  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
791  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
792  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
793  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
794  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
795  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
796    characters with codes greater than 128, you should either use UTF-8 and
797  An alternative set of tables can, however, be supplied. Such tables are built  Unicode, or use locales, but not try to mix the two.
798  by calling the \fBpcre_maketables()\fR function, which has no arguments, in the  .P
799  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
800  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.
801  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
802  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
803    default "C" locale of the local system, which may cause them to be different.
804    setlocale(LC_CTYPE, "fr");  .P
805    The internal tables can always be overridden by tables supplied by the
806    application that calls PCRE. These may be created in a different locale from
807    the default. As more and more applications change to using Unicode, the need
808    for this locale support is expected to die away.
809    .P
810    External tables are built by calling the \fBpcre_maketables()\fP function,
811    which has no arguments, in the relevant locale. The result can then be passed
812    to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
813    example, to build and use tables that are appropriate for the French locale
814    (where accented characters with values greater than 128 are treated as letters),
815    the following code could be used:
816    .sp
817      setlocale(LC_CTYPE, "fr_FR");
818    tables = pcre_maketables();    tables = pcre_maketables();
819    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
820    .sp
821  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
822  pointer that is passed to \fBpcre_compile\fR is saved with the compiled  are using Windows, the name for the French locale is "french".
823  pattern, and the same tables are used via this pointer by \fBpcre_study()\fR  .P
824  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
825  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
826  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
827  memory containing the tables remains available for as long as it is needed.  needed.
828    .P
829  .SH INFORMATION ABOUT A PATTERN  The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
830    pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
831    and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
832    pattern, compilation, studying and matching all happen in the same locale, but
833    different patterns can be compiled in different locales.
834    .P
835    It is possible to pass a table pointer or NULL (indicating the use of the
836    internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
837    this facility could be used to match a pattern in a different locale from the
838    one in which it was compiled. Passing table pointers at run time is discussed
839    below in the section on matching a pattern.
840    .
841    .
842    .SH "INFORMATION ABOUT A PATTERN"
843  .rs  .rs
844  .sp  .sp
845  .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,"
846  .ti +5n  .ti +5n
847  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
848  .PP  .PP
849  The \fBpcre_fullinfo()\fR function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
850  pattern. It replaces the obsolete \fBpcre_info()\fR function, which is  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
851  nevertheless retained for backwards compability (and is documented below).  nevertheless retained for backwards compability (and is documented below).
852    .P
853  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
854  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
855  the pattern was not studied. The third argument specifies which piece of  the pattern was not studied. The third argument specifies which piece of
856  information is required, and the fourth argument is a pointer to a variable  information is required, and the fourth argument is a pointer to a variable
857  to receive the data. The yield of the function is zero for success, or one of  to receive the data. The yield of the function is zero for success, or one of
858  the following negative numbers:  the following negative numbers:
859    .sp
860    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
861                          the argument \fIwhere\fR was NULL                          the argument \fIwhere\fP was NULL
862    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
863    PCRE_ERROR_BADOPTION  the value of \fIwhat\fR was invalid    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid
864    .sp
865  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
866  compiled pattern:  check against passing an arbitrary memory pointer. Here is a typical call of
867    \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
868    .sp
869    int rc;    int rc;
870    unsigned long int length;    size_t length;
871    rc = pcre_fullinfo(    rc = pcre_fullinfo(
872      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
873      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
874      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
875      &length);         /* where to put the data */      &length);         /* where to put the data */
876    .sp
877  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
878  as follows:  as follows:
879    .sp
880    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
881    .sp
882  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
883  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
884  no back references.  no back references.
885    .sp
886    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
887    .sp
888  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
889  should point to an \fbint\fR variable.  should point to an \fBint\fP variable.
890    .sp
891      PCRE_INFO_DEFAULT_TABLES
892    .sp
893    Return a pointer to the internal default character tables within PCRE. The
894    fourth argument should point to an \fBunsigned char *\fP variable. This
895    information call is provided for internal use by the \fBpcre_study()\fP
896    function. External callers can cause PCRE to use its internal tables by passing
897    a NULL table pointer.
898    .sp
899    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
900    .sp
901  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
902  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
903  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
904    still recognized for backwards compatibility.)
905  If there is a fixed first byte, e.g. from a pattern such as (cat|cow|coyote),  .P
906  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
907    (cat|cow|coyote), its value is returned. Otherwise, if either
908    .sp
909  (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
910  starts with "^", or  starts with "^", or
911    .sp
912  (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
913  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
914    .sp
915  -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
916  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
917  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
918    .sp
919    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
920    .sp
921  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
922  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
923  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
924  fourth argument should point to an \fBunsigned char *\fR variable.  fourth argument should point to an \fBunsigned char *\fP variable.
925    .sp
926      PCRE_INFO_HASCRORLF
927    .sp
928    Return 1 if the pattern contains any explicit matches for CR or LF characters,
929    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
930    explicit match is either a literal CR or LF character, or \er or \en.
931    .sp
932      PCRE_INFO_JCHANGED
933    .sp
934    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
935    0. The fourth argument should point to an \fBint\fP variable. (?J) and
936    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
937    .sp
938    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
939    .sp
940  For a non-anchored pattern, return the value of the rightmost literal byte  Return the value of the rightmost literal byte that must exist in any matched
941  which must exist in any matched string, other than at its start. The fourth  string, other than at its start, if such a byte has been recorded. The fourth
942  argument should point to an \fBint\fR variable. If there is no such byte, or if  argument should point to an \fBint\fP variable. If there is no such byte, -1 is
943  the pattern is anchored, -1 is returned. For example, for the pattern  returned. For anchored patterns, a last literal byte is recorded only if it
944  /a\\d+z\\d+/ the returned value is 'z'.  follows something of variable length. For example, for the pattern
945    /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
946    is -1.
947    .sp
948    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
949    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
950    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
951    .sp
952  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
953  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
954  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
955  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
956  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
957  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
958  table for the pattern.  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
959    you need to use the name-to-number map, which is described by these three
960    values.
961    .P
962  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
963  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
964  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
965  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
966  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
967  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
968  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated. The names are in
969  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
970    their parentheses numbers. For example, consider the following pattern (assume
971  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
972    .sp
973    (?P<date> (?P<year>(\\d\\d)?\\d\\d) -  .\" JOIN
974    (?P<month>\\d\\d) - (?P<day>\\d\\d) )    (?<date> (?<year>(\ed\ed)?\ed\ed) -
975      (?<month>\ed\ed) - (?<day>\ed\ed) )
976    .sp
977  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
978  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
979  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
980    .sp
981    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
982    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
983    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
984    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
985    .sp
986  When writing code to extract data from named subpatterns, remember that the  When writing code to extract data from named subpatterns using the
987  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
988    different for each compiled pattern.
989    .sp
990      PCRE_INFO_OKPARTIAL
991    .sp
992    Return 1 if the pattern can be used for partial matching, otherwise 0. The
993    fourth argument should point to an \fBint\fP variable. The
994    .\" HREF
995    \fBpcrepartial\fP
996    .\"
997    documentation lists the restrictions that apply to patterns when partial
998    matching is used.
999    .sp
1000    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1001    .sp
1002  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
1003  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
1004  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
1005  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1006    they are the options that will be in force when matching starts. For example,
1007    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1008    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1009    .P
1010  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
1011  alternatives begin with one of the following:  alternatives begin with one of the following:
1012    .sp
1013    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
1014    \\A    always    \eA    always
1015    \\G    always    \eG    always
1016    .\" JOIN
1017    .*    if PCRE_DOTALL is set and there are no back    .*    if PCRE_DOTALL is set and there are no back
1018            references to the subpattern in which .* appears            references to the subpattern in which .* appears
1019    .sp
1020  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
1021  \fBpcre_fullinfo()\fR.  \fBpcre_fullinfo()\fP.
1022    .sp
1023    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1024    .sp
1025  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
1026  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
1027  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
1028  variable.  variable.
1029    .sp
1030    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1031    .sp
1032  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
1033  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
1034  \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
1035  created by \fBpcre_study()\fR. The fourth argument should point to a  created by \fBpcre_study()\fP. The fourth argument should point to a
1036  \fBsize_t\fR variable.  \fBsize_t\fP variable.
1037    .
1038  .SH OBSOLETE INFO FUNCTION  .
1039    .SH "OBSOLETE INFO FUNCTION"
1040  .rs  .rs
1041  .sp  .sp
1042  .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1043  .B *\fIfirstcharptr\fR);  .B *\fIfirstcharptr\fP);
1044  .PP  .PP
1045  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
1046  restrictive to return all the available data about a compiled pattern. New  restrictive to return all the available data about a compiled pattern. New
1047  programs should use \fBpcre_fullinfo()\fR instead. The yield of  programs should use \fBpcre_fullinfo()\fP instead. The yield of
1048  \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
1049  following negative numbers:  following negative numbers:
1050    .sp
1051    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
1052    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1053    .sp
1054  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
1055  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1056  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
1057    .P
1058  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,
1059  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
1060  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1061    .
1062  .SH MATCHING A PATTERN  .
1063    .SH "REFERENCE COUNTS"
1064  .rs  .rs
1065  .sp  .sp
1066  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1067    .PP
1068    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1069    data block that contains a compiled pattern. It is provided for the benefit of
1070    applications that operate in an object-oriented manner, where different parts
1071    of the application may be using the same compiled pattern, but you want to free
1072    the block when they are all done.
1073    .P
1074    When a pattern is compiled, the reference count field is initialized to zero.
1075    It is changed only by calling this function, whose action is to add the
1076    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1077    function is the new value. However, the value of the count is constrained to
1078    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1079    it is forced to the appropriate limit value.
1080    .P
1081    Except when it is zero, the reference count is not correctly preserved if a
1082    pattern is compiled on one host and then transferred to a host whose byte-order
1083    is different. (This seems a highly unlikely scenario.)
1084    .
1085    .
1086    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1087    .rs
1088    .sp
1089    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1090  .ti +5n  .ti +5n
1091  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1092  .ti +5n  .ti +5n
1093  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1094  .PP  .P
1095  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
1096  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
1097  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
1098  \fIextra\fR argument.  \fIextra\fP argument. This function is the main matching facility of the
1099    library, and it operates in a Perl-like manner. For specialist use there is
1100  Here is an example of a simple call to \fBpcre_exec()\fR:  also an alternative matching function, which is described
1101    .\" HTML <a href="#dfamatch">
1102    .\" </a>
1103    below
1104    .\"
1105    in the section about the \fBpcre_dfa_exec()\fP function.
1106    .P
1107    In most applications, the pattern will have been compiled (and optionally
1108    studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1109    possible to save compiled patterns and study data, and then use them later
1110    in different processes, possibly even on different hosts. For a discussion
1111    about this, see the
1112    .\" HREF
1113    \fBpcreprecompile\fP
1114    .\"
1115    documentation.
1116    .P
1117    Here is an example of a simple call to \fBpcre_exec()\fP:
1118    .sp
1119    int rc;    int rc;
1120    int ovector[30];    int ovector[30];
1121    rc = pcre_exec(    rc = pcre_exec(
# Line 645  Here is an example of a simple call to \ Line 1125  Here is an example of a simple call to \
1125      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1126      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1127      0,              /* default options */      0,              /* default options */
1128      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1129      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1130    .
1131  If the \fIextra\fR argument is not NULL, it must point to a \fBpcre_extra\fR  .\" HTML <a name="extradata"></a>
1132  data block. The \fBpcre_study()\fR function returns such a block (when it  .SS "Extra data for \fBpcre_exec()\fR"
1133    .rs
1134    .sp
1135    If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1136    data block. The \fBpcre_study()\fP function returns such a block (when it
1137  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
1138  additional information in it. The fields in the block are as follows:  additional information in it. The \fBpcre_extra\fP block contains the following
1139    fields (not necessarily in this order):
1140    unsigned long int \fIflags\fR;  .sp
1141    void *\fIstudy_data\fR;    unsigned long int \fIflags\fP;
1142    unsigned long int \fImatch_limit\fR;    void *\fIstudy_data\fP;
1143    void *\fIcallout_data\fR;    unsigned long int \fImatch_limit\fP;
1144      unsigned long int \fImatch_limit_recursion\fP;
1145  The \fIflags\fR field is a bitmap that specifies which of the other fields    void *\fIcallout_data\fP;
1146      const unsigned char *\fItables\fP;
1147    .sp
1148    The \fIflags\fP field is a bitmap that specifies which of the other fields
1149  are set. The flag bits are:  are set. The flag bits are:
1150    .sp
1151    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1152    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1153      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1154    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1155      PCRE_EXTRA_TABLES
1156  Other flag bits should be set to zero. The \fIstudy_data\fR field is set in the  .sp
1157  \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
1158  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
1159  the block by setting the other fields.  the appropriate flag bit. You should not set this yourself, but you may add to
1160    the block by setting the other fields and their corresponding flag bits.
1161  The \fImatch_limit\fR field provides a means of preventing PCRE from using up a  .P
1162    The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1163  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,
1164  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
1165  classic example is the use of nested unlimited repeats. Internally, PCRE uses a  classic example is the use of nested unlimited repeats.
1166  function called \fBmatch()\fR which it calls repeatedly (sometimes  .P
1167  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
1168  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
1169  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
1170  count starts from zero for each position in the subject string.  limiting the amount of backtracking that can take place. For patterns that are
1171    not anchored, the count restarts from zero for each position in the subject
1172  The default limit for the library can be set when PCRE is built; the default  string.
1173    .P
1174    The default value for the limit can be set when PCRE is built; the default
1175  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
1176  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
1177  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
1178  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
1179  exceeded, \fBpcre_exec()\fR returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1180    .P
1181  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
1182  which is described in the \fBpcrecallout\fR documentation.  instead of limiting the total number of times that \fBmatch()\fP is called, it
1183    limits the depth of recursion. The recursion depth is a smaller number than the
1184  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.
1185  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.
1186  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  .P
1187  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,
1188  unachored at matching time.  when PCRE has been compiled to use memory on the heap instead of the stack, the
1189    amount of heap memory that can be used.
1190  There are also three further options that can be set only at matching time:  .P
1191    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1192    built; the default default is the same value as the default for
1193    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1194    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1195    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1196    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1197    .P
1198    The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1199    which is described in the
1200    .\" HREF
1201    \fBpcrecallout\fP
1202    .\"
1203    documentation.
1204    .P
1205    The \fItables\fP field is used to pass a character tables pointer to
1206    \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1207    pattern. A non-NULL value is stored with the compiled pattern only if custom
1208    tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1209    If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1210    internal tables to be used. This facility is helpful when re-using patterns
1211    that have been saved after compiling with an external set of tables, because
1212    the external tables might be at a different address when \fBpcre_exec()\fP is
1213    called. See the
1214    .\" HREF
1215    \fBpcreprecompile\fP
1216    .\"
1217    documentation for a discussion of saving compiled patterns for later use.
1218    .
1219    .\" HTML <a name="execoptions"></a>
1220    .SS "Option bits for \fBpcre_exec()\fP"
1221    .rs
1222    .sp
1223    The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1224    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1225    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1226    .sp
1227      PCRE_ANCHORED
1228    .sp
1229    The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1230    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1231    to be anchored by virtue of its contents, it cannot be made unachored at
1232    matching time.
1233    .sp
1234      PCRE_BSR_ANYCRLF
1235      PCRE_BSR_UNICODE
1236    .sp
1237    These options (which are mutually exclusive) control what the \eR escape
1238    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1239    match any Unicode newline sequence. These options override the choice that was
1240    made or defaulted when the pattern was compiled.
1241    .sp
1242      PCRE_NEWLINE_CR
1243      PCRE_NEWLINE_LF
1244      PCRE_NEWLINE_CRLF
1245      PCRE_NEWLINE_ANYCRLF
1246      PCRE_NEWLINE_ANY
1247    .sp
1248    These options override the newline definition that was chosen or defaulted when
1249    the pattern was compiled. For details, see the description of
1250    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1251    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1252    the way the match position is advanced after a match failure for an unanchored
1253    pattern.
1254    .P
1255    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1256    match attempt for an unanchored pattern fails when the current position is at a
1257    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1258    characters, the match position is advanced by two characters instead of one, in
1259    other words, to after the CRLF.
1260    .P
1261    The above rule is a compromise that makes the most common cases work as
1262    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1263    set), it does not match the string "\er\enA" because, after failing at the
1264    start, it skips both the CR and the LF before retrying. However, the pattern
1265    [\er\en]A does match that string, because it contains an explicit CR or LF
1266    reference, and so advances only by one character after the first failure.
1267    .P
1268    An explicit match for CR of LF is either a literal appearance of one of those
1269    characters, or one of the \er or \en escape sequences. Implicit matches such as
1270    [^X] do not count, nor does \es (which includes CR and LF in the characters
1271    that it matches).
1272    .P
1273    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1274    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1275    .sp
1276    PCRE_NOTBOL    PCRE_NOTBOL
1277    .sp
1278  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
1279  circumflex metacharacter should not match before it. Setting this without  beginning of a line, so the circumflex metacharacter should not match before
1280  PCRE_MULTILINE (at compile time) causes circumflex never to match.  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1281    never to match. This option affects only the behaviour of the circumflex
1282    metacharacter. It does not affect \eA.
1283    .sp
1284    PCRE_NOTEOL    PCRE_NOTEOL
1285    .sp
1286  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
1287  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
1288  it. Setting this without PCRE_MULTILINE (at compile time) causes dollar never  mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1289  to match.  compile time) causes dollar never to match. This option affects only the
1290    behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1291    .sp
1292    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1293    .sp
1294  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
1295  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
1296  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
1297    .sp
1298    a?b?    a?b?
1299    .sp
1300  is applied to a string not beginning with "a" or "b", it matches the empty  is applied to a string not beginning with "a" or "b", it matches the empty
1301  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
1302  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".
1303    .P
1304  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case
1305  of a pattern match of the empty string within its \fBsplit()\fR function, and  of a pattern match of the empty string within its \fBsplit()\fP function, and
1306  when using the /g modifier. It is possible to emulate Perl's behaviour after  when using the /g modifier. It is possible to emulate Perl's behaviour after
1307  matching a null string by first trying the match again at the same offset with  matching a null string by first trying the match again at the same offset with
1308  PCRE_NOTEMPTY set, and then if that fails by advancing the starting offset (see  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the
1309  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1310    code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1311  The subject string is passed to \fBpcre_exec()\fR as a pointer in  .sp
1312  \fIsubject\fR, a length in \fIlength\fR, and a starting offset in    PCRE_NO_UTF8_CHECK
1313  \fIstartoffset\fR. Unlike the pattern string, the subject may contain binary  .sp
1314  zero bytes. When the starting offset is zero, the search for a match starts at  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1315  the beginning of the subject, and this is by far the most common case.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1316    The value of \fIstartoffset\fP is also checked to ensure that it points to the
1317  If the pattern was compiled with the PCRE_UTF8 option, the subject must be a  start of a UTF-8 character. There is a discussion about the validity of UTF-8
1318  sequence of bytes that is a valid UTF-8 string. If an invalid UTF-8 string is  strings in the
1319  passed, PCRE's behaviour is not defined.  .\" HTML <a href="pcre.html#utf8strings">
1320    .\" </a>
1321    section on UTF-8 support
1322    .\"
1323    in the main
1324    .\" HREF
1325    \fBpcre\fP
1326    .\"
1327    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1328    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1329    PCRE_ERROR_BADUTF8_OFFSET is returned.
1330    .P
1331    If you already know that your subject is valid, and you want to skip these
1332    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1333    calling \fBpcre_exec()\fP. You might want to do this for the second and
1334    subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1335    all the matches in a single subject string. However, you should be sure that
1336    the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1337    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1338    subject, or a value of \fIstartoffset\fP that does not point to the start of a
1339    UTF-8 character, is undefined. Your program may crash.
1340    .sp
1341      PCRE_PARTIAL
1342    .sp
1343    This option turns on the partial matching feature. If the subject string fails
1344    to match the pattern, but at some point during the matching process the end of
1345    the subject was reached (that is, the subject partially matches the pattern and
1346    the failure to match occurred only because there were not enough subject
1347    characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1348    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1349    may appear in the pattern. These are discussed in the
1350    .\" HREF
1351    \fBpcrepartial\fP
1352    .\"
1353    documentation.
1354    .
1355    .SS "The string to be matched by \fBpcre_exec()\fP"
1356    .rs
1357    .sp
1358    The subject string is passed to \fBpcre_exec()\fP as a pointer in
1359    \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
1360    \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
1361    UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1362    bytes. When the starting offset is zero, the search for a match starts at the
1363    beginning of the subject, and this is by far the most common case.
1364    .P
1365  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
1366  same subject by calling \fBpcre_exec()\fR again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
1367  Setting \fIstartoffset\fR differs from just passing over a shortened string and  Setting \fIstartoffset\fP differs from just passing over a shortened string and
1368  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
1369  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
1370    .sp
1371    \\Biss\\B    \eBiss\eB
1372    .sp
1373  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
1374  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
1375  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
1376  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
1377  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
1378  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
1379  \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
1380  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
1381  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.
1382    .P
1383  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
1384  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
1385  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.
1386    .
1387    .SS "How \fBpcre_exec()\fP returns captured substrings"
1388    .rs
1389    .sp
1390  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
1391  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
1392  pattern. Following the usage in Jeffrey Friedl's book, this is called  pattern. Following the usage in Jeffrey Friedl's book, this is called
1393  "capturing" in what follows, and the phrase "capturing subpattern" is used for  "capturing" in what follows, and the phrase "capturing subpattern" is used for
1394  a fragment of a pattern that picks out a substring. PCRE supports several other  a fragment of a pattern that picks out a substring. PCRE supports several other
1395  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1396    .P
1397  Captured substrings are returned to the caller via a vector of integer offsets  Captured substrings are returned to the caller via a vector of integer offsets
1398  whose address is passed in \fIovector\fR. The number of elements in the vector  whose address is passed in \fIovector\fP. The number of elements in the vector
1399  is passed in \fIovecsize\fR. The first two-thirds of the vector is used to pass  is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:
1400  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of \fIovector\fP in bytes.
1401  remaining third of the vector is used as workspace by \fBpcre_exec()\fR while  .P
1402  matching capturing subpatterns, and is not available for passing back  The first two-thirds of the vector is used to pass back captured substrings,
1403  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
1404  three. If it is not, it is rounded down.  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1405    and is not available for passing back information. The length passed in
1406  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
1407  returned in pairs of integers, starting at the beginning of \fIovector\fR, and  rounded down.
1408    .P
1409    When a match is successful, information about captured substrings is returned
1410    in pairs of integers, starting at the beginning of \fIovector\fP, and
1411  continuing up to two-thirds of its length at the most. The first element of a  continuing up to two-thirds of its length at the most. The first element of a
1412  pair is set to the offset of the first character in a substring, and the second  pair is set to the offset of the first character in a substring, and the second
1413  is set to the offset of the first character after the end of a substring. The  is set to the offset of the first character after the end of a substring. The
1414  first pair, \fIovector[0]\fR and \fIovector[1]\fR, identify the portion of the  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
1415  subject string matched by the entire pattern. The next pair is used for the  subject string matched by the entire pattern. The next pair is used for the
1416  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fR  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
1417  is the number of pairs that have been set. If there are no capturing  is one more than the highest numbered pair that has been set. For example, if
1418  subpatterns, the return value from a successful match is 1, indicating that  two substrings have been captured, the returned value is 3. If there are no
1419  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1420    indicating that just the first pair of offsets has been set.
1421  Some convenience functions are provided for extracting the captured substrings  .P
 as separate strings. These are described in the following section.  
   
 It is possible for an capturing subpattern number \fIn+1\fR to match some  
 part of the subject when subpattern \fIn\fR has not been used at all. For  
 example, if the string "abc" is matched against the pattern (a|(z))(bc)  
 subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset  
 values corresponding to the unused subpattern are set to -1.  
   
1422  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
1423  string that it matched that gets returned.  string that it matched that is returned.
1424    .P
1425  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
1426  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
1427  value of zero. In particular, if the substring offsets are not of interest,  returns a value of zero. In particular, if the substring offsets are not of
1428  \fBpcre_exec()\fR may be called with \fIovector\fR passed as NULL and  interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1429  \fIovecsize\fR as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1430  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
1431  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
1432  to supply an \fIovector\fR.  advisable to supply an \fIovector\fP.
1433    .P
1434  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
1435  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1436  \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
1437  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.
1438    .P
1439  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
1440    the subject when subpattern \fIn\fP has not been used at all. For example, if
1441    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1442    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1443    happens, both values in the offset pairs corresponding to unused subpatterns
1444    are set to -1.
1445    .P
1446    Offset values that correspond to unused subpatterns at the end of the
1447    expression are also set to -1. For example, if the string "abc" is matched
1448    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1449    return from the function is 2, because the highest used capturing subpattern
1450    number is 1. However, you can refer to the offsets for the second and third
1451    capturing subpatterns if you wish (assuming the vector is large enough, of
1452    course).
1453    .P
1454    Some convenience functions are provided for extracting the captured substrings
1455    as separate strings. These are described below.
1456    .
1457    .\" HTML <a name="errorlist"></a>
1458    .SS "Error return values from \fBpcre_exec()\fP"
1459    .rs
1460    .sp
1461    If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1462  defined in the header file:  defined in the header file:
1463    .sp
1464    PCRE_ERROR_NOMATCH        (-1)    PCRE_ERROR_NOMATCH        (-1)
1465    .sp
1466  The subject string did not match the pattern.  The subject string did not match the pattern.
1467    .sp
1468    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1469    .sp
1470  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
1471  NULL and \fIovecsize\fR was not zero.  NULL and \fIovecsize\fP was not zero.
1472    .sp
1473    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1474    .sp
1475  An unrecognized bit was set in the \fIoptions\fR argument.  An unrecognized bit was set in the \fIoptions\fP argument.
1476    .sp
1477    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1478    .sp
1479  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
1480  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
1481  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1482    other endianness. This is the error that PCRE gives when the magic number is
1483    PCRE_ERROR_UNKNOWN_NODE   (-5)  not present.
1484    .sp
1485      PCRE_ERROR_UNKNOWN_OPCODE (-5)
1486    .sp
1487  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1488  compiled pattern. This error could be caused by a bug in PCRE or by overwriting  compiled pattern. This error could be caused by a bug in PCRE or by overwriting
1489  of the compiled pattern.  of the compiled pattern.
1490    .sp
1491    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1492    .sp
1493  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
1494  \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
1495  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
1496  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
1497  the end of matching.  automatically freed at the end of matching.
1498    .sp
1499    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1500    .sp
1501  This error is used by the \fBpcre_copy_substring()\fR,  This error is used by the \fBpcre_copy_substring()\fP,
1502  \fBpcre_get_substring()\fR, and \fBpcre_get_substring_list()\fR functions (see  \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1503  below). It is never returned by \fBpcre_exec()\fR.  below). It is never returned by \fBpcre_exec()\fP.
1504    .sp
1505    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1506    .sp
1507  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
1508  field in a \fBpcre_extra\fR structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1509  description above.  above.
1510    .sp
1511    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1512    .sp
1513  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
1514  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
1515  \fBpcrecallout\fR documentation for details.  .\" HREF
1516    \fBpcrecallout\fP
1517  .SH EXTRACTING CAPTURED SUBSTRINGS BY NUMBER  .\"
1518    documentation for details.
1519    .sp
1520      PCRE_ERROR_BADUTF8        (-10)
1521    .sp
1522    A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1523    .sp
1524      PCRE_ERROR_BADUTF8_OFFSET (-11)
1525    .sp
1526    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1527    of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1528    .sp
1529      PCRE_ERROR_PARTIAL        (-12)
1530    .sp
1531    The subject string did not match, but it did match partially. See the
1532    .\" HREF
1533    \fBpcrepartial\fP
1534    .\"
1535    documentation for details of partial matching.
1536    .sp
1537      PCRE_ERROR_BADPARTIAL     (-13)
1538    .sp
1539    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1540    are not supported for partial matching. See the
1541    .\" HREF
1542    \fBpcrepartial\fP
1543    .\"
1544    documentation for details of partial matching.
1545    .sp
1546      PCRE_ERROR_INTERNAL       (-14)
1547    .sp
1548    An unexpected internal error has occurred. This error could be caused by a bug
1549    in PCRE or by overwriting of the compiled pattern.
1550    .sp
1551      PCRE_ERROR_BADCOUNT       (-15)
1552    .sp
1553    This error is given if the value of the \fIovecsize\fP argument is negative.
1554    .sp
1555      PCRE_ERROR_RECURSIONLIMIT (-21)
1556    .sp
1557    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1558    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1559    description above.
1560    .sp
1561      PCRE_ERROR_BADNEWLINE     (-23)
1562    .sp
1563    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1564    .P
1565    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1566    .
1567    .
1568    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1569  .rs  .rs
1570  .sp  .sp
1571  .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1572  .ti +5n  .ti +5n
1573  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1574  .ti +5n  .ti +5n
1575  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
1576  .PP  .PP
1577  .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,  
1578  .ti +5n  .ti +5n
1579  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1580  .ti +5n  .ti +5n
1581  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1582  .PP  .PP
1583  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
1584  .ti +5n  .ti +5n
1585  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1586  .PP  .PP
1587  Captured substrings can be accessed directly by using the offsets returned by  Captured substrings can be accessed directly by using the offsets returned by
1588  \fBpcre_exec()\fR in \fIovector\fR. For convenience, the functions  \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1589  \fBpcre_copy_substring()\fR, \fBpcre_get_substring()\fR, and  \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1590  \fBpcre_get_substring_list()\fR are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1591  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1592  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1593  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1594  has a further zero added on the end, but the result is not, of course,  .P
1595  a C string.  A substring that contains a binary zero is correctly extracted and has a
1596    further zero added on the end, but the result is not, of course, a C string.
1597    However, you can process such a string by referring to the length that is
1598    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1599    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1600    for handling strings containing binary zeros, because the end of the final
1601    string is not independently indicated.
1602    .P
1603  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:
1604  \fIsubject\fR is the subject string which has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
1605  \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
1606  \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
1607  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1608  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
1609  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
1610  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
1611  the vector divided by three.  number of elements in the vector divided by three.
1612    .P
1613  The functions \fBpcre_copy_substring()\fR and \fBpcre_get_substring()\fR  The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1614  extract a single substring, whose number is given as \fIstringnumber\fR. A  extract a single substring, whose number is given as \fIstringnumber\fP. A
1615  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1616  higher values extract the captured substrings. For \fBpcre_copy_substring()\fR,  higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1617  the string is placed in \fIbuffer\fR, whose length is given by  the string is placed in \fIbuffer\fP, whose length is given by
1618  \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
1619  obtained via \fBpcre_malloc\fR, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1620  \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
1621  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1622    .sp
1623    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1624    .sp
1625  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
1626  memory failed for \fBpcre_get_substring()\fR.  memory failed for \fBpcre_get_substring()\fP.
1627    .sp
1628    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1629    .sp
1630  There is no substring whose number is \fIstringnumber\fR.  There is no substring whose number is \fIstringnumber\fP.
1631    .P
1632  The \fBpcre_get_substring_list()\fR function extracts all available substrings  The \fBpcre_get_substring_list()\fP function extracts all available substrings
1633  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
1634  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
1635  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
1636  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
1637  function is zero if all went well, or  function is zero if all went well, or the error code
1638    .sp
1639    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1640    .sp
1641  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1642    .P
1643  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
1644  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
1645  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
1646  string. This can be distinguished from a genuine zero-length substring by  string. This can be distinguished from a genuine zero-length substring by
1647  inspecting the appropriate offset in \fIovector\fR, which is negative for unset  inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1648  substrings.  substrings.
1649    .P
1650  The two convenience functions \fBpcre_free_substring()\fR and  The two convenience functions \fBpcre_free_substring()\fP and
1651  \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
1652  a previous call of \fBpcre_get_substring()\fR or  a previous call of \fBpcre_get_substring()\fP or
1653  \fBpcre_get_substring_list()\fR, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1654  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
1655  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
1656  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1657  \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
1658  provided.  provided.
1659    .
1660  .SH EXTRACTING CAPTURED SUBSTRINGS BY NAME  .
1661    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1662  .rs  .rs
1663  .sp  .sp
1664  .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1665  .ti +5n  .ti +5n
1666  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIname\fP);
1667    .PP
1668    .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1669  .ti +5n  .ti +5n
1670  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1671  .ti +5n  .ti +5n
1672  .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,  
1673  .ti +5n  .ti +5n
1674  .B const char *\fIname\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1675  .PP  .PP
1676  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
 .B int pcre_get_named_substring(const pcre *\fIcode\fR,  
1677  .ti +5n  .ti +5n
1678  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1679  .ti +5n  .ti +5n
1680  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
1681  .ti +5n  .ti +5n
1682  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1683  .PP  .PP
1684  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.
1685  can be done by calling \fBpcre_get_stringnumber()\fR. The first argument is the  For example, for this pattern
1686  compiled pattern, and the second is the name. For example, for this pattern  .sp
1687      (a+)b(?<xxx>\ed+)...
1688    ab(?<xxx>\\d+)...  .sp
1689    the number of the subpattern called "xxx" is 2. If the name is known to be
1690  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
1691  extract the substring directly, or use one of the functions described in the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1692  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
1693  whole job.  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1694    that name.
1695  Most of the arguments of \fIpcre_copy_named_substring()\fR and  .P
1696  \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
1697  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
1698  differences.  two functions that do the whole job.
1699    .P
1700    Most of the arguments of \fBpcre_copy_named_substring()\fP and
1701    \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1702    functions that extract by number. As these are described in the previous
1703    section, they are not re-described here. There are just two differences:
1704    .P
1705  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
1706  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
1707  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
1708  translation table.  translation table.
1709    .P
1710  These functions call \fBpcre_get_stringnumber()\fR, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1711  then call \fIpcre_copy_substring()\fR or \fIpcre_get_substring()\fR, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1712  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1713    the behaviour may not be what you want (see the next section).
1714  .in 0  .
1715  Last updated: 03 February 2003  .
1716  .br  .SH "DUPLICATE SUBPATTERN NAMES"
1717  Copyright (c) 1997-2003 University of Cambridge.  .rs
1718    .sp
1719    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1720    .ti +5n
1721    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1722    .PP
1723    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1724    are not required to be unique. Normally, patterns with duplicate names are such
1725    that in any one match, only one of the named subpatterns participates. An
1726    example is shown in the
1727    .\" HREF
1728    \fBpcrepattern\fP
1729    .\"
1730    documentation.
1731    .P
1732    When duplicates are present, \fBpcre_copy_named_substring()\fP and
1733    \fBpcre_get_named_substring()\fP return the first substring corresponding to
1734    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1735    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1736    returns one of the numbers that are associated with the name, but it is not
1737    defined which it is.
1738    .P
1739    If you want to get full details of all captured substrings for a given name,
1740    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1741    argument is the compiled pattern, and the second is the name. The third and
1742    fourth are pointers to variables which are updated by the function. After it
1743    has run, they point to the first and last entries in the name-to-number table
1744    for the given name. The function itself returns the length of each entry, or
1745    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1746    described above in the section entitled \fIInformation about a pattern\fP.
1747    Given all the relevant entries for the name, you can extract each of their
1748    numbers, and hence the captured data, if any.
1749    .
1750    .
1751    .SH "FINDING ALL POSSIBLE MATCHES"
1752    .rs
1753    .sp
1754    The traditional matching function uses a similar algorithm to Perl, which stops
1755    when it finds the first match, starting at a given point in the subject. If you
1756    want to find all possible matches, or the longest possible match, consider
1757    using the alternative matching function (see below) instead. If you cannot use
1758    the alternative function, but still need to find all possible matches, you
1759    can kludge it up by making use of the callout facility, which is described in
1760    the
1761    .\" HREF
1762    \fBpcrecallout\fP
1763    .\"
1764    documentation.
1765    .P
1766    What you have to do is to insert a callout right at the end of the pattern.
1767    When your callout function is called, extract and save the current matched
1768    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1769    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1770    will yield PCRE_ERROR_NOMATCH.
1771    .
1772    .
1773    .\" HTML <a name="dfamatch"></a>
1774    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1775    .rs
1776    .sp
1777    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1778    .ti +5n
1779    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1780    .ti +5n
1781    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1782    .ti +5n
1783    .B int *\fIworkspace\fP, int \fIwscount\fP);
1784    .P
1785    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1786    a compiled pattern, using a matching algorithm that scans the subject string
1787    just once, and does not backtrack. This has different characteristics to the
1788    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1789    patterns are not supported. Nevertheless, there are times when this kind of
1790    matching can be useful. For a discussion of the two matching algorithms, see
1791    the
1792    .\" HREF
1793    \fBpcrematching\fP
1794    .\"
1795    documentation.
1796    .P
1797    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1798    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1799    different way, and this is described below. The other common arguments are used
1800    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1801    here.
1802    .P
1803    The two additional arguments provide workspace for the function. The workspace
1804    vector should contain at least 20 elements. It is used for keeping track of
1805    multiple paths through the pattern tree. More workspace will be needed for
1806    patterns and subjects where there are a lot of potential matches.
1807    .P
1808    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1809    .sp
1810      int rc;
1811      int ovector[10];
1812      int wspace[20];
1813      rc = pcre_dfa_exec(
1814        re,             /* result of pcre_compile() */
1815        NULL,           /* we didn't study the pattern */
1816        "some string",  /* the subject string */
1817        11,             /* the length of the subject string */
1818        0,              /* start at offset 0 in the subject */
1819        0,              /* default options */
1820        ovector,        /* vector of integers for substring information */
1821        10,             /* number of elements (NOT size in bytes) */
1822        wspace,         /* working space vector */
1823        20);            /* number of elements (NOT size in bytes) */
1824    .
1825    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1826    .rs
1827    .sp
1828    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1829    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1830    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1831    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1832    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1833    .sp
1834      PCRE_PARTIAL
1835    .sp
1836    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1837    details are slightly different. When PCRE_PARTIAL is set for
1838    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1839    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1840    complete matches, but there is still at least one matching possibility. The
1841    portion of the string that provided the partial match is set as the first
1842    matching string.
1843    .sp
1844      PCRE_DFA_SHORTEST
1845    .sp
1846    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1847    soon as it has found one match. Because of the way the alternative algorithm
1848    works, this is necessarily the shortest possible match at the first possible
1849    matching point in the subject string.
1850    .sp
1851      PCRE_DFA_RESTART
1852    .sp
1853    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1854    a partial match, it is possible to call it again, with additional subject
1855    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1856    option requests this action; when it is set, the \fIworkspace\fP and
1857    \fIwscount\fP options must reference the same vector as before because data
1858    about the match so far is left in them after a partial match. There is more
1859    discussion of this facility in the
1860    .\" HREF
1861    \fBpcrepartial\fP
1862    .\"
1863    documentation.
1864    .
1865    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1866    .rs
1867    .sp
1868    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1869    substring in the subject. Note, however, that all the matches from one run of
1870    the function start at the same point in the subject. The shorter matches are
1871    all initial substrings of the longer matches. For example, if the pattern
1872    .sp
1873      <.*>
1874    .sp
1875    is matched against the string
1876    .sp
1877      This is <something> <something else> <something further> no more
1878    .sp
1879    the three matched strings are
1880    .sp
1881      <something>
1882      <something> <something else>
1883      <something> <something else> <something further>
1884    .sp
1885    On success, the yield of the function is a number greater than zero, which is
1886    the number of matched substrings. The substrings themselves are returned in
1887    \fIovector\fP. Each string uses two elements; the first is the offset to the
1888    start, and the second is the offset to the end. In fact, all the strings have
1889    the same start offset. (Space could have been saved by giving this only once,
1890    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1891    returns data, even though the meaning of the strings is different.)
1892    .P
1893    The strings are returned in reverse order of length; that is, the longest
1894    matching string is given first. If there were too many matches to fit into
1895    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1896    the longest matches.
1897    .
1898    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1899    .rs
1900    .sp
1901    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1902    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1903    described
1904    .\" HTML <a href="#errorlist">
1905    .\" </a>
1906    above.
1907    .\"
1908    There are in addition the following errors that are specific to
1909    \fBpcre_dfa_exec()\fP:
1910    .sp
1911      PCRE_ERROR_DFA_UITEM      (-16)
1912    .sp
1913    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1914    that it does not support, for instance, the use of \eC or a back reference.
1915    .sp
1916      PCRE_ERROR_DFA_UCOND      (-17)
1917    .sp
1918    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1919    uses a back reference for the condition, or a test for recursion in a specific
1920    group. These are not supported.
1921    .sp
1922      PCRE_ERROR_DFA_UMLIMIT    (-18)
1923    .sp
1924    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1925    block that contains a setting of the \fImatch_limit\fP field. This is not
1926    supported (it is meaningless).
1927    .sp
1928      PCRE_ERROR_DFA_WSSIZE     (-19)
1929    .sp
1930    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1931    \fIworkspace\fP vector.
1932    .sp
1933      PCRE_ERROR_DFA_RECURSE    (-20)
1934    .sp
1935    When a recursive subpattern is processed, the matching function calls itself
1936    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1937    error is given if the output vector is not large enough. This should be
1938    extremely rare, as a vector of size 1000 is used.
1939    .
1940    .
1941    .SH "SEE ALSO"
1942    .rs
1943    .sp
1944    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1945    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1946    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1947    .
1948    .
1949    .SH AUTHOR
1950    .rs
1951    .sp
1952    .nf
1953    Philip Hazel
1954    University Computing Service
1955    Cambridge CB2 3QH, England.
1956    .fi
1957    .
1958    .
1959    .SH REVISION
1960    .rs
1961    .sp
1962    .nf
1963    Last updated: 23 January 2008
1964    Copyright (c) 1997-2008 University of Cambridge.
1965    .fi

Legend:
Removed from v.63  
changed lines
  Added in v.312

  ViewVC Help
Powered by ViewVC 1.1.5