/[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 73 by nigel, Sat Feb 24 21:40:30 2007 UTC revision 155 by ph10, Tue Apr 24 13:36:11 2007 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  .PP
28  .br  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
 .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  
29  .ti +5n  .ti +5n
30  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31  .ti +5n  .ti +5n
32  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .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,"
 .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  
35  .ti +5n  .ti +5n
36  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37  .ti +5n  .ti +5n
38  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39  .ti +5n  .ti +5n
40  .B char *\fIbuffer\fR, int \fIbuffersize\fR);  .B int *\fIworkspace\fP, int \fIwscount\fP);
41  .PP  .PP
42  .br  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
43  .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  .ti +5n
44    .B const char *\fIsubject\fP, int *\fIovector\fP,
45  .ti +5n  .ti +5n
46  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
47  .ti +5n  .ti +5n
48  .B 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_get_named_substring(const pcre *\fIcode\fR,  
51  .ti +5n  .ti +5n
52  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53  .ti +5n  .ti +5n
54  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIbuffersize\fP);
55    .PP
56    .B int pcre_get_named_substring(const pcre *\fIcode\fP,
57  .ti +5n  .ti +5n
58  .B const char **\fIstringptr\fR);  .B const char *\fIsubject\fP, int *\fIovector\fP,
59    .ti +5n
60    .B int \fIstringcount\fP, const char *\fIstringname\fP,
61    .ti +5n
62    .B const char **\fIstringptr\fP);
63  .PP  .PP
64  .br  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
 .B int pcre_get_stringnumber(const pcre *\fIcode\fR,  
65  .ti +5n  .ti +5n
66  .B const char *\fIname\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_substring(const char *\fIsubject\fR, int *\fIovector\fR,  
69  .ti +5n  .ti +5n
70  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71    .PP
72    .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
73  .ti +5n  .ti +5n
74  .B const char **\fIstringptr\fR);  .B int \fIstringcount\fP, int \fIstringnumber\fP,
75    .ti +5n
76    .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
 .br  
105  .B void *(*pcre_stack_malloc)(size_t);  .B void *(*pcre_stack_malloc)(size_t);
106  .PP  .PP
 .br  
107  .B void (*pcre_stack_free)(void *);  .B void (*pcre_stack_free)(void *);
108  .PP  .PP
 .br  
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 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 variables \fBpcre_stack_malloc\fR and \fBpcre_stack_free\fR are also  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  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  only when PCRE is compiled to use the heap for remembering data, instead of
196  recursive function calls. This is a non-standard way of building PCRE, for use  recursive function calls, when running the \fBpcre_exec()\fP function. See the
197  in environments that have limited stacks. Because of the greater use of memory  .\" HREF
198  management, it runs more slowly. Separate functions are provided so that  \fBpcrebuild\fP
199  special-purpose external code can be used for this case. When used, these  .\"
200  functions are always called in a stack-like manner (last obtained, first  documentation for details of how to do this. It is a non-standard way of
201  freed), and always for memory blocks of the same size.  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  The global variable \fBpcre_callout\fR initially contains NULL. It can be set  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    .SH NEWLINES
222    .rs
223    .sp
224    PCRE supports five different conventions for indicating line breaks in
225    strings: a single CR (carriage return) character, a single LF (linefeed)
226    character, the two-character sequence CRLF, any of the three preceding, or any
227    Unicode newline sequence. The Unicode newline sequences are the three just
228    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
229    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
230    (paragraph separator, U+2029).
231    .P
232    Each of the first three conventions is used by at least one operating system as
233    its standard newline sequence. When PCRE is built, a default can be specified.
234    The default default is LF, which is the Unix standard. When PCRE is run, the
235    default can be overridden, either when a pattern is compiled, or when it is
236    matched.
237    .P
238    In the PCRE documentation the word "newline" is used to mean "the character or
239    pair of characters that indicate a line break". The choice of newline
240    convention affects the handling of the dot, circumflex, and dollar
241    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
242    recognized line ending sequence, the match position advancement for a
243    non-anchored pattern. The choice of newline convention does not affect the
244    interpretation of the \en or \er escape sequences.
245    .
246    .
247  .SH MULTITHREADING  .SH MULTITHREADING
248  .rs  .rs
249  .sp  .sp
250  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
251  proviso that the memory management functions pointed to by \fBpcre_malloc\fR,  proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
252  \fBpcre_free\fR, \fBpcre_stack_malloc\fR, and \fBpcre_stack_free\fR, and the  \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
253  callout function pointed to by \fBpcre_callout\fR, are shared by all threads.  callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
254    .P
255  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
256  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.
257    .
258  .SH CHECKING BUILD-TIME OPTIONS  .
259    .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
260  .rs  .rs
261  .sp  .sp
262  .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
263    time, possibly by a different program, and even on a host other than the one on
264    which it was compiled. Details are given in the
265    .\" HREF
266    \fBpcreprecompile\fP
267    .\"
268    documentation. However, compiling a regular expression with one version of PCRE
269    for use with a different version is not guaranteed to work and may cause
270    crashes.
271    .
272    .
273    .SH "CHECKING BUILD-TIME OPTIONS"
274    .rs
275    .sp
276    .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
277  .PP  .PP
278  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
279  discover which optional features have been compiled into the PCRE library. The  discover which optional features have been compiled into the PCRE library. The
280  .\" HREF  .\" HREF
281  \fBpcrebuild\fR  \fBpcrebuild\fP
282  .\"  .\"
283  documentation has more details about these optional features.  documentation has more details about these optional features.
284    .P
285  The first argument for \fBpcre_config()\fR is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
286  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
287  which the information is placed. The following information is available:  which the information is placed. The following information is available:
288    .sp
289    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
290    .sp
291  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;
292  otherwise it is set to zero.  otherwise it is set to zero.
293    .sp
294      PCRE_CONFIG_UNICODE_PROPERTIES
295    .sp
296    The output is an integer that is set to one if support for Unicode character
297    properties is available; otherwise it is set to zero.
298    .sp
299    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
300    .sp
301  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
302  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
303  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
304    default should normally be the standard sequence for your operating system.
305    .sp
306    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
307    .sp
308  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
309  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
310  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
311  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
312  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.
313    .sp
314    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
315    .sp
316  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
317  interface uses \fBmalloc()\fR for output vectors. Further details are given in  interface uses \fBmalloc()\fP for output vectors. Further details are given in
318  the \fBpcreposix\fR documentation.  the
319    .\" HREF
320    \fBpcreposix\fP
321    .\"
322    documentation.
323    .sp
324    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
325    .sp
326  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
327  internal matching function calls in a \fBpcre_exec()\fR execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
328  details are given with \fBpcre_exec()\fR below.  details are given with \fBpcre_exec()\fP below.
329    .sp
330      PCRE_CONFIG_MATCH_LIMIT_RECURSION
331    .sp
332    The output is an integer that gives the default limit for the depth of
333    recursion when calling the internal matching function in a \fBpcre_exec()\fP
334    execution. Further details are given with \fBpcre_exec()\fP below.
335    .sp
336    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
337    .sp
338  The output is an integer that is set to one if internal recursion is  The output is an integer that is set to one if internal recursion when running
339  implemented by recursive function calls that use the stack to remember their  \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
340  state. This is the usual way that PCRE is compiled. The output is zero if PCRE  to remember their state. This is the usual way that PCRE is compiled. The
341  was compiled to use blocks of data on the heap instead of recursive function  output is zero if PCRE was compiled to use blocks of data on the heap instead
342  calls. In this case, \fBpcre_stack_malloc\fR and \fBpcre_stack_free\fR are  of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
343  called to manage memory blocks on the heap, thus avoiding the use of the stack.  \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
344    avoiding the use of the stack.
345  .SH COMPILING A PATTERN  .
346  .rs  .
347  .sp  .SH "COMPILING A PATTERN"
348  .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  .rs
349  .ti +5n  .sp
350  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
351  .ti +5n  .ti +5n
352  .B const unsigned char *\fItableptr\fR);  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
353  .PP  .ti +5n
354    .B const unsigned char *\fItableptr\fP);
355  The function \fBpcre_compile()\fR is called to compile a pattern into an  .sp
356  internal form. The pattern is a C string terminated by a binary zero, and  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
357  is passed in the argument \fIpattern\fR. A pointer to a single block of memory  .ti +5n
358  that is obtained via \fBpcre_malloc\fR is returned. This contains the compiled  .B int *\fIerrorcodeptr\fP,
359  code and related data. The \fBpcre\fR type is defined for the returned block;  .ti +5n
360  this is a typedef for a structure whose contents are not externally defined. It  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
361  is up to the caller to free the memory when it is no longer required.  .ti +5n
362    .B const unsigned char *\fItableptr\fP);
363    .P
364    Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
365    called to compile a pattern into an internal form. The only difference between
366    the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
367    \fIerrorcodeptr\fP, via which a numerical error code can be returned.
368    .P
369    The pattern is a C string terminated by a binary zero, and is passed in the
370    \fIpattern\fP argument. A pointer to a single block of memory that is obtained
371    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
372    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
373    for a structure whose contents are not externally defined. It is up to the
374    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
375    .P
376  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
377  depend on memory location, the complete \fBpcre\fR data block is not  depend on memory location, the complete \fBpcre\fP data block is not
378  fully relocatable, because it contains a copy of the \fItableptr\fR argument,  fully relocatable, because it may contain a copy of the \fItableptr\fP
379  which is an address (see below).  argument, which is an address (see below).
380    .P
381  The \fIoptions\fR argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
382  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
383  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
384  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
385  in the \fBpcrepattern\fR documentation). For these options, the contents of the  the detailed description in the
386  \fIoptions\fR argument specifies their initial settings at the start of  .\" HREF
387  compilation and execution. The PCRE_ANCHORED option can be set at the time of  \fBpcrepattern\fP
388    .\"
389    documentation). For these options, the contents of the \fIoptions\fP argument
390    specifies their initial settings at the start of compilation and execution. The
391    PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
392  matching as well as at compile time.  matching as well as at compile time.
393    .P
394  If \fIerrptr\fR is NULL, \fBpcre_compile()\fR returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
395  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fR returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
396  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
397  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
398  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
399  \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
400    \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
401  If the final argument, \fItableptr\fR, is NULL, PCRE uses a default set of  .P
402  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
403  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
404  \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
405    textual error message. Error codes and messages are listed below.
406  This code fragment shows a typical straightforward call to \fBpcre_compile()\fR:  .P
407    If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
408    character tables that are built when PCRE is compiled, using the default C
409    locale. Otherwise, \fItableptr\fP must be an address that is the result of a
410    call to \fBpcre_maketables()\fP. This value is stored with the compiled
411    pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
412    passed to it. For more discussion, see the section on locale support below.
413    .P
414    This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
415    .sp
416    pcre *re;    pcre *re;
417    const char *error;    const char *error;
418    int erroffset;    int erroffset;
# Line 290  This code fragment shows a typical strai Line 422  This code fragment shows a typical strai
422      &error,           /* for error message */      &error,           /* for error message */
423      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
424      NULL);            /* use default character tables */      NULL);            /* use default character tables */
425    .sp
426  The following option bits are defined:  The following names for option bits are defined in the \fBpcre.h\fP header
427    file:
428    .sp
429    PCRE_ANCHORED    PCRE_ANCHORED
430    .sp
431  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
432  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
433  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
434  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
435  Perl.  Perl.
436    .sp
437      PCRE_AUTO_CALLOUT
438    .sp
439    If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
440    all with number 255, before each pattern item. For discussion of the callout
441    facility, see the
442    .\" HREF
443    \fBpcrecallout\fP
444    .\"
445    documentation.
446    .sp
447    PCRE_CASELESS    PCRE_CASELESS
448    .sp
449  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
450  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
451  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
452    concept of case for characters whose values are less than 128, so caseless
453    matching is always possible. For characters with higher values, the concept of
454    case is supported if PCRE is compiled with Unicode property support, but not
455    otherwise. If you want to use caseless matching for characters 128 and above,
456    you must ensure that PCRE is compiled with Unicode property support as well as
457    with UTF-8 support.
458    .sp
459    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
460    .sp
461  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
462  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
463  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
464  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.
465  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
466  a pattern.  pattern.
467    .sp
468    PCRE_DOTALL    PCRE_DOTALL
469    .sp
470  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,
471  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
472  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
473  (?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
474  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
475    the setting of this option.
476    .sp
477      PCRE_DUPNAMES
478    .sp
479    If this bit is set, names used to identify capturing subpatterns need not be
480    unique. This can be helpful for certain types of pattern when it is known that
481    only one instance of the named subpattern can ever be matched. There are more
482    details of named subpatterns below; see also the
483    .\" HREF
484    \fBpcrepattern\fP
485    .\"
486    documentation.
487    .sp
488    PCRE_EXTENDED    PCRE_EXTENDED
489    .sp
490  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
491  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
492  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
493  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
494  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
495  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
496    .P
497  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
498  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
499  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
500  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
501    .sp
502    PCRE_EXTRA    PCRE_EXTRA
503    .sp
504  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
505  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
506  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
507  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
508  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
509  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
510  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
511  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
512    .sp
513      PCRE_FIRSTLINE
514    .sp
515    If this option is set, an unanchored pattern is required to match before or at
516    the first newline in the subject string, though the matched text may continue
517    over the newline.
518    .sp
519    PCRE_MULTILINE    PCRE_MULTILINE
520    .sp
521  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
522  characters (even if it actually contains several newlines). The "start of line"  characters (even if it actually contains newlines). The "start of line"
523  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
524  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
525  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
526  Perl.  Perl.
527    .P
528  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
529  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
530  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
531  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
532  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
533  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
534    .sp
535      PCRE_NEWLINE_CR
536      PCRE_NEWLINE_LF
537      PCRE_NEWLINE_CRLF
538      PCRE_NEWLINE_ANYCRLF
539      PCRE_NEWLINE_ANY
540    .sp
541    These options override the default newline definition that was chosen when PCRE
542    was built. Setting the first or the second specifies that a newline is
543    indicated by a single character (CR or LF, respectively). Setting
544    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
545    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
546    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
547    that any Unicode newline sequence should be recognized. The Unicode newline
548    sequences are the three just mentioned, plus the single characters VT (vertical
549    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
550    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
551    recognized only in UTF-8 mode.
552    .P
553    The newline setting in the options word uses three bits that are treated
554    as a number, giving eight possibilities. Currently only six are used (default
555    plus the five values above). This means that if you set more than one newline
556    option, the combination may or may not be sensible. For example,
557    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
558    other combinations may yield unused numbers and cause an error.
559    .P
560    The only time that a line break is specially recognized when compiling a
561    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
562    class is encountered. This indicates a comment that lasts until after the next
563    line break sequence. In other circumstances, line break sequences are treated
564    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
565    as whitespace characters and are therefore ignored.
566    .P
567    The newline option that is set at compile time becomes the default that is used
568    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
569    .sp
570    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
571    .sp
572  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
573  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
574  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
575  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
576  in Perl.  in Perl.
577    .sp
578    PCRE_UNGREEDY    PCRE_UNGREEDY
579    .sp
580  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
581  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
582  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.
583    .sp
584    PCRE_UTF8    PCRE_UTF8
585    .sp
586  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
587  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
588  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
589  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
590  behaviour of PCRE are given in the  behaviour of PCRE are given in the
591  .\" HTML <a href="pcre.html#utf8support">  .\" HTML <a href="pcre.html#utf8support">
# Line 392  section on UTF-8 support Line 594  section on UTF-8 support
594  .\"  .\"
595  in the main  in the main
596  .\" HREF  .\" HREF
597  \fBpcre\fR  \fBpcre\fP
598  .\"  .\"
599  page.  page.
600    .sp
601    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
602    .sp
603  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
604  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. If an invalid UTF-8 sequence of bytes is found,
605  \fBpcre_compile()\fR returns an error. If you already know that your pattern is  \fBpcre_compile()\fP returns an error. If you already know that your pattern is
606  valid, and you want to skip this check for performance reasons, you can set the  valid, and you want to skip this check for performance reasons, you can set the
607  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
608  UTF-8 string as a pattern is undefined. It may cause your program to crash.  UTF-8 string as a pattern is undefined. It may cause your program to crash.
609  Note that there is a similar option for suppressing the checking of subject  Note that this option can also be passed to \fBpcre_exec()\fP and
610  strings passed to \fBpcre_exec()\fR.  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject
611    strings.
612    .
613  .SH STUDYING A PATTERN  .
614  .rs  .SH "COMPILATION ERROR CODES"
615  .sp  .rs
616  .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  .sp
617  .ti +5n  The following table lists the error codes than may be returned by
618  .B const char **\fIerrptr\fR);  \fBpcre_compile2()\fP, along with the error messages that may be returned by
619  .PP  both compiling functions. As PCRE has developed, some error codes have fallen
620  When a pattern is going to be used several times, it is worth spending more  out of use. To avoid confusion, they have not been re-used.
621  time analyzing it in order to speed up the time taken for matching. The  .sp
622  function \fBpcre_study()\fR takes a pointer to a compiled pattern as its first     0  no error
623  argument. If studing the pattern produces additional information that will help     1  \e at end of pattern
624  speed up matching, \fBpcre_study()\fR returns a pointer to a \fBpcre_extra\fR     2  \ec at end of pattern
625  block, in which the \fIstudy_data\fR field points to the results of the study.     3  unrecognized character follows \e
626       4  numbers out of order in {} quantifier
627  The returned value from a \fBpcre_study()\fR can be passed directly to     5  number too big in {} quantifier
628  \fBpcre_exec()\fR. However, the \fBpcre_extra\fR block also contains other     6  missing terminating ] for character class
629       7  invalid escape sequence in character class
630       8  range out of order in character class
631       9  nothing to repeat
632      10  [this code is not in use]
633      11  internal error: unexpected repeat
634      12  unrecognized character after (?
635      13  POSIX named classes are supported only within a class
636      14  missing )
637      15  reference to non-existent subpattern
638      16  erroffset passed as NULL
639      17  unknown option bit(s) set
640      18  missing ) after comment
641      19  [this code is not in use]
642      20  regular expression too large
643      21  failed to get memory
644      22  unmatched parentheses
645      23  internal error: code overflow
646      24  unrecognized character after (?<
647      25  lookbehind assertion is not fixed length
648      26  malformed number or name after (?(
649      27  conditional group contains more than two branches
650      28  assertion expected after (?(
651      29  (?R or (?digits must be followed by )
652      30  unknown POSIX class name
653      31  POSIX collating elements are not supported
654      32  this version of PCRE is not compiled with PCRE_UTF8 support
655      33  [this code is not in use]
656      34  character value in \ex{...} sequence is too large
657      35  invalid condition (?(0)
658      36  \eC not allowed in lookbehind assertion
659      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
660      38  number after (?C is > 255
661      39  closing ) for (?C expected
662      40  recursive call could loop indefinitely
663      41  unrecognized character after (?P
664      42  syntax error in subpattern name (missing terminator)
665      43  two named subpatterns have the same name
666      44  invalid UTF-8 string
667      45  support for \eP, \ep, and \eX has not been compiled
668      46  malformed \eP or \ep sequence
669      47  unknown property name after \eP or \ep
670      48  subpattern name is too long (maximum 32 characters)
671      49  too many named subpatterns (maximum 10,000)
672      50  repeated subpattern is too long
673      51  octal value is greater than \e377 (not in UTF-8 mode)
674      52  internal error: overran compiling workspace
675      53  internal error: previously-checked referenced subpattern not found
676      54  DEFINE group contains more than one branch
677      55  repeating a DEFINE group is not allowed
678      56  inconsistent NEWLINE options"
679    .
680    .
681    .SH "STUDYING A PATTERN"
682    .rs
683    .sp
684    .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
685    .ti +5n
686    .B const char **\fIerrptr\fP);
687    .PP
688    If a compiled pattern is going to be used several times, it is worth spending
689    more time analyzing it in order to speed up the time taken for matching. The
690    function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
691    argument. If studying the pattern produces additional information that will
692    help speed up matching, \fBpcre_study()\fP returns a pointer to a
693    \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
694    results of the study.
695    .P
696    The returned value from \fBpcre_study()\fP can be passed directly to
697    \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
698  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
699  described below. If studying the pattern does not produce any additional  described
700  information, \fBpcre_study()\fR returns NULL. In that circumstance, if the  .\" HTML <a href="#extradata">
701  calling program wants to pass some of the other fields to \fBpcre_exec()\fR, it  .\" </a>
702  must set up its own \fBpcre_extra\fR block.  below
703    .\"
704  The second argument contains option bits. At present, no options are defined  in the section on matching a pattern.
705  for \fBpcre_study()\fR, and this argument should always be zero.  .P
706    If studying the pattern does not produce any additional information
707  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
708    wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
709    own \fBpcre_extra\fP block.
710    .P
711    The second argument of \fBpcre_study()\fP contains option bits. At present, no
712    options are defined, and this argument should always be zero.
713    .P
714    The third argument for \fBpcre_study()\fP is a pointer for an error message. If
715  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
716  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
717  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
718  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
719    sure that it has run successfully.
720  This is a typical call to \fBpcre_study\fR():  .P
721    This is a typical call to \fBpcre_study\fP():
722    .sp
723    pcre_extra *pe;    pcre_extra *pe;
724    pe = pcre_study(    pe = pcre_study(
725      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
726      0,              /* no options exist */      0,              /* no options exist */
727      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
728    .sp
729  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
730  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
731  characters is created.  bytes is created.
732    .
733    .
734  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
735  .SH LOCALE SUPPORT  .SH "LOCALE SUPPORT"
736  .rs  .rs
737  .sp  .sp
738  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
739  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
740  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
741  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
742  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
743  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
744    characters with codes greater than 128, you should either use UTF-8 and
745  An alternative set of tables can, however, be supplied. Such tables are built  Unicode, or use locales, but not try to mix the two.
746  by calling the \fBpcre_maketables()\fR function, which has no arguments, in the  .P
747  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
748  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.
749  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
750  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
751    default "C" locale of the local system, which may cause them to be different.
752    setlocale(LC_CTYPE, "fr");  .P
753    The internal tables can always be overridden by tables supplied by the
754    application that calls PCRE. These may be created in a different locale from
755    the default. As more and more applications change to using Unicode, the need
756    for this locale support is expected to die away.
757    .P
758    External tables are built by calling the \fBpcre_maketables()\fP function,
759    which has no arguments, in the relevant locale. The result can then be passed
760    to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
761    example, to build and use tables that are appropriate for the French locale
762    (where accented characters with values greater than 128 are treated as letters),
763    the following code could be used:
764    .sp
765      setlocale(LC_CTYPE, "fr_FR");
766    tables = pcre_maketables();    tables = pcre_maketables();
767    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
768    .sp
769  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
770  pointer that is passed to \fBpcre_compile\fR is saved with the compiled  are using Windows, the name for the French locale is "french".
771  pattern, and the same tables are used via this pointer by \fBpcre_study()\fR  .P
772  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
773  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
774  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
775  memory containing the tables remains available for as long as it is needed.  needed.
776    .P
777  .SH INFORMATION ABOUT A PATTERN  The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
778    pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
779    and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
780    pattern, compilation, studying and matching all happen in the same locale, but
781    different patterns can be compiled in different locales.
782    .P
783    It is possible to pass a table pointer or NULL (indicating the use of the
784    internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
785    this facility could be used to match a pattern in a different locale from the
786    one in which it was compiled. Passing table pointers at run time is discussed
787    below in the section on matching a pattern.
788    .
789    .
790    .SH "INFORMATION ABOUT A PATTERN"
791  .rs  .rs
792  .sp  .sp
793  .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,"
794  .ti +5n  .ti +5n
795  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
796  .PP  .PP
797  The \fBpcre_fullinfo()\fR function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
798  pattern. It replaces the obsolete \fBpcre_info()\fR function, which is  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
799  nevertheless retained for backwards compability (and is documented below).  nevertheless retained for backwards compability (and is documented below).
800    .P
801  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
802  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
803  the pattern was not studied. The third argument specifies which piece of  the pattern was not studied. The third argument specifies which piece of
804  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
805  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
806  the following negative numbers:  the following negative numbers:
807    .sp
808    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
809                          the argument \fIwhere\fR was NULL                          the argument \fIwhere\fP was NULL
810    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
811    PCRE_ERROR_BADOPTION  the value of \fIwhat\fR was invalid    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid
812    .sp
813  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
814  compiled pattern:  check against passing an arbitrary memory pointer. Here is a typical call of
815    \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
816    .sp
817    int rc;    int rc;
818    unsigned long int length;    size_t length;
819    rc = pcre_fullinfo(    rc = pcre_fullinfo(
820      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
821      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
822      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
823      &length);         /* where to put the data */      &length);         /* where to put the data */
824    .sp
825  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
826  as follows:  as follows:
827    .sp
828    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
829    .sp
830  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
831  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
832  no back references.  no back references.
833    .sp
834    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
835    .sp
836  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
837  should point to an \fbint\fR variable.  should point to an \fBint\fP variable.
838    .sp
839      PCRE_INFO_DEFAULT_TABLES
840    .sp
841    Return a pointer to the internal default character tables within PCRE. The
842    fourth argument should point to an \fBunsigned char *\fP variable. This
843    information call is provided for internal use by the \fBpcre_study()\fP
844    function. External callers can cause PCRE to use its internal tables by passing
845    a NULL table pointer.
846    .sp
847    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
848    .sp
849  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
850  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
851  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
852    still recognized for backwards compatibility.)
853  If there is a fixed first byte, e.g. from a pattern such as (cat|cow|coyote),  .P
854  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
855    (cat|cow|coyote), its value is returned. Otherwise, if either
856    .sp
857  (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
858  starts with "^", or  starts with "^", or
859    .sp
860  (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
861  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
862    .sp
863  -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
864  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
865  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
866    .sp
867    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
868    .sp
869  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
870  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
871  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
872  fourth argument should point to an \fBunsigned char *\fR variable.  fourth argument should point to an \fBunsigned char *\fP variable.
873    .sp
874    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
875    .sp
876  Return the value of the rightmost literal byte that must exist in any matched  Return the value of the rightmost literal byte that must exist in any matched
877  string, other than at its start, if such a byte has been recorded. The fourth  string, other than at its start, if such a byte has been recorded. The fourth
878  argument should point to an \fBint\fR variable. If there is no such byte, -1 is  argument should point to an \fBint\fP variable. If there is no such byte, -1 is
879  returned. For anchored patterns, a last literal byte is recorded only if it  returned. For anchored patterns, a last literal byte is recorded only if it
880  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
881  /^a\\d+z\\d+/ the returned value is "z", but for /^a\\dz\\d/ the returned value  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
882  is -1.  is -1.
883    .sp
884    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
885    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
886    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
887    .sp
888  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
889  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
890  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
891  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
892  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
893  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
894  table for the pattern.  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
895    you need to use the name-to-number map, which is described by these three
896    values.
897    .P
898  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
899  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
900  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
901  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
902  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
903  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
904  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
905  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
906    their parentheses numbers. For example, consider the following pattern (assume
907  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
908    .sp
909    (?P<date> (?P<year>(\\d\\d)?\\d\\d) -  .\" JOIN
910    (?P<month>\\d\\d) - (?P<day>\\d\\d) )    (?<date> (?<year>(\ed\ed)?\ed\ed) -
911      (?<month>\ed\ed) - (?<day>\ed\ed) )
912    .sp
913  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
914  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
915  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
916    .sp
917    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
918    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
919    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
920    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
921    .sp
922  When writing code to extract data from named subpatterns, remember that the  When writing code to extract data from named subpatterns using the
923  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
924    different for each compiled pattern.
925    .sp
926    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
927    .sp
928  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
929  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
930  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
931  top-level option settings within the pattern itself.  top-level option settings within the pattern itself.
932    .P
933  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
934  alternatives begin with one of the following:  alternatives begin with one of the following:
935    .sp
936    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
937    \\A    always    \eA    always
938    \\G    always    \eG    always
939    .\" JOIN
940    .*    if PCRE_DOTALL is set and there are no back    .*    if PCRE_DOTALL is set and there are no back
941            references to the subpattern in which .* appears            references to the subpattern in which .* appears
942    .sp
943  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
944  \fBpcre_fullinfo()\fR.  \fBpcre_fullinfo()\fP.
945    .sp
946    PCRE_INFO_SIZE    PCRE_INFO_SIZE
947    .sp
948  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
949  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
950  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
951  variable.  variable.
952    .sp
953    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
954    .sp
955  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
956  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
957  \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
958  created by \fBpcre_study()\fR. The fourth argument should point to a  created by \fBpcre_study()\fP. The fourth argument should point to a
959  \fBsize_t\fR variable.  \fBsize_t\fP variable.
960    .
961  .SH OBSOLETE INFO FUNCTION  .
962    .SH "OBSOLETE INFO FUNCTION"
963  .rs  .rs
964  .sp  .sp
965  .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
966  .B *\fIfirstcharptr\fR);  .B *\fIfirstcharptr\fP);
967  .PP  .PP
968  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
969  restrictive to return all the available data about a compiled pattern. New  restrictive to return all the available data about a compiled pattern. New
970  programs should use \fBpcre_fullinfo()\fR instead. The yield of  programs should use \fBpcre_fullinfo()\fP instead. The yield of
971  \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
972  following negative numbers:  following negative numbers:
973    .sp
974    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
975    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
976    .sp
977  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
978  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
979  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
980    .P
981  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,
982  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
983  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
984    .
985  .SH MATCHING A PATTERN  .
986    .SH "REFERENCE COUNTS"
987  .rs  .rs
988  .sp  .sp
989  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
990    .PP
991    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
992    data block that contains a compiled pattern. It is provided for the benefit of
993    applications that operate in an object-oriented manner, where different parts
994    of the application may be using the same compiled pattern, but you want to free
995    the block when they are all done.
996    .P
997    When a pattern is compiled, the reference count field is initialized to zero.
998    It is changed only by calling this function, whose action is to add the
999    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1000    function is the new value. However, the value of the count is constrained to
1001    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1002    it is forced to the appropriate limit value.
1003    .P
1004    Except when it is zero, the reference count is not correctly preserved if a
1005    pattern is compiled on one host and then transferred to a host whose byte-order
1006    is different. (This seems a highly unlikely scenario.)
1007    .
1008    .
1009    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1010    .rs
1011    .sp
1012    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1013  .ti +5n  .ti +5n
1014  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1015  .ti +5n  .ti +5n
1016  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1017  .PP  .P
1018  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
1019  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
1020  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
1021  \fIextra\fR argument.  \fIextra\fP argument. This function is the main matching facility of the
1022    library, and it operates in a Perl-like manner. For specialist use there is
1023  Here is an example of a simple call to \fBpcre_exec()\fR:  also an alternative matching function, which is described
1024    .\" HTML <a href="#dfamatch">
1025    .\" </a>
1026    below
1027    .\"
1028    in the section about the \fBpcre_dfa_exec()\fP function.
1029    .P
1030    In most applications, the pattern will have been compiled (and optionally
1031    studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1032    possible to save compiled patterns and study data, and then use them later
1033    in different processes, possibly even on different hosts. For a discussion
1034    about this, see the
1035    .\" HREF
1036    \fBpcreprecompile\fP
1037    .\"
1038    documentation.
1039    .P
1040    Here is an example of a simple call to \fBpcre_exec()\fP:
1041    .sp
1042    int rc;    int rc;
1043    int ovector[30];    int ovector[30];
1044    rc = pcre_exec(    rc = pcre_exec(
# Line 684  Here is an example of a simple call to \ Line 1048  Here is an example of a simple call to \
1048      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1049      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1050      0,              /* default options */      0,              /* default options */
1051      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1052      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1053    .
1054  If the \fIextra\fR argument is not NULL, it must point to a \fBpcre_extra\fR  .\" HTML <a name="extradata"></a>
1055  data block. The \fBpcre_study()\fR function returns such a block (when it  .SS "Extra data for \fBpcre_exec()\fR"
1056    .rs
1057    .sp
1058    If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1059    data block. The \fBpcre_study()\fP function returns such a block (when it
1060  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
1061  additional information in it. The fields in the block are as follows:  additional information in it. The \fBpcre_extra\fP block contains the following
1062    fields (not necessarily in this order):
1063    unsigned long int \fIflags\fR;  .sp
1064    void *\fIstudy_data\fR;    unsigned long int \fIflags\fP;
1065    unsigned long int \fImatch_limit\fR;    void *\fIstudy_data\fP;
1066    void *\fIcallout_data\fR;    unsigned long int \fImatch_limit\fP;
1067      unsigned long int \fImatch_limit_recursion\fP;
1068  The \fIflags\fR field is a bitmap that specifies which of the other fields    void *\fIcallout_data\fP;
1069      const unsigned char *\fItables\fP;
1070    .sp
1071    The \fIflags\fP field is a bitmap that specifies which of the other fields
1072  are set. The flag bits are:  are set. The flag bits are:
1073    .sp
1074    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1075    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1076      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1077    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1078      PCRE_EXTRA_TABLES
1079  Other flag bits should be set to zero. The \fIstudy_data\fR field is set in the  .sp
1080  \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
1081  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
1082  the block by setting the other fields.  the appropriate flag bit. You should not set this yourself, but you may add to
1083    the block by setting the other fields and their corresponding flag bits.
1084  The \fImatch_limit\fR field provides a means of preventing PCRE from using up a  .P
1085    The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1086  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,
1087  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
1088  classic example is the use of nested unlimited repeats. Internally, PCRE uses a  classic example is the use of nested unlimited repeats.
1089  function called \fBmatch()\fR which it calls repeatedly (sometimes  .P
1090  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
1091  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
1092  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
1093  count starts from zero for each position in the subject string.  limiting the amount of backtracking that can take place. For patterns that are
1094    not anchored, the count restarts from zero for each position in the subject
1095  The default limit for the library can be set when PCRE is built; the default  string.
1096    .P
1097    The default value for the limit can be set when PCRE is built; the default
1098  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
1099  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
1100  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
1101  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
1102  exceeded, \fBpcre_exec()\fR returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1103    .P
1104  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
1105  which is described in the \fBpcrecallout\fR documentation.  instead of limiting the total number of times that \fBmatch()\fP is called, it
1106    limits the depth of recursion. The recursion depth is a smaller number than the
1107  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.
1108  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.
1109  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  .P
1110  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,
1111  unachored at matching time.  when PCRE has been compiled to use memory on the heap instead of the stack, the
1112    amount of heap memory that can be used.
1113  When PCRE_UTF8 was set at compile time, the validity of the subject as a UTF-8  .P
1114  string is automatically checked, and the value of \fIstartoffset\fR is also  The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1115  checked to ensure that it points to the start of a UTF-8 character. If an  built; the default default is the same value as the default for
1116  invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fR returns the error  \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1117  PCRE_ERROR_BADUTF8. If \fIstartoffset\fR contains an invalid value,  with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1118  PCRE_ERROR_BADUTF8_OFFSET is returned.  PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1119    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1120  If you already know that your subject is valid, and you want to skip these  .P
1121  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1122  calling \fBpcre_exec()\fR. You might want to do this for the second and  which is described in the
1123  subsequent calls to \fBpcre_exec()\fR if you are making repeated calls to find  .\" HREF
1124  all the matches in a single subject string. However, you should be sure that  \fBpcrecallout\fP
1125  the value of \fIstartoffset\fR points to the start of a UTF-8 character. When  .\"
1126  PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a  documentation.
1127  subject, or a value of \fIstartoffset\fR that does not point to the start of a  .P
1128  UTF-8 character, is undefined. Your program may crash.  The \fItables\fP field is used to pass a character tables pointer to
1129    \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1130  There are also three further options that can be set only at matching time:  pattern. A non-NULL value is stored with the compiled pattern only if custom
1131    tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1132    If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1133    internal tables to be used. This facility is helpful when re-using patterns
1134    that have been saved after compiling with an external set of tables, because
1135    the external tables might be at a different address when \fBpcre_exec()\fP is
1136    called. See the
1137    .\" HREF
1138    \fBpcreprecompile\fP
1139    .\"
1140    documentation for a discussion of saving compiled patterns for later use.
1141    .
1142    .SS "Option bits for \fBpcre_exec()\fP"
1143    .rs
1144    .sp
1145    The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1146    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1147    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1148    .sp
1149      PCRE_ANCHORED
1150    .sp
1151    The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1152    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1153    to be anchored by virtue of its contents, it cannot be made unachored at
1154    matching time.
1155    .sp
1156      PCRE_NEWLINE_CR
1157      PCRE_NEWLINE_LF
1158      PCRE_NEWLINE_CRLF
1159      PCRE_NEWLINE_ANYCRLF
1160      PCRE_NEWLINE_ANY
1161    .sp
1162    These options override the newline definition that was chosen or defaulted when
1163    the pattern was compiled. For details, see the description of
1164    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1165    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1166    the way the match position is advanced after a match failure for an unanchored
1167    pattern. When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
1168    set, and a match attempt fails when the current position is at a CRLF sequence,
1169    the match position is advanced by two characters instead of one, in other
1170    words, to after the CRLF.
1171    .sp
1172    PCRE_NOTBOL    PCRE_NOTBOL
1173    .sp
1174  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
1175  circumflex metacharacter should not match before it. Setting this without  beginning of a line, so the circumflex metacharacter should not match before
1176  PCRE_MULTILINE (at compile time) causes circumflex never to match.  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1177    never to match. This option affects only the behaviour of the circumflex
1178    metacharacter. It does not affect \eA.
1179    .sp
1180    PCRE_NOTEOL    PCRE_NOTEOL
1181    .sp
1182  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
1183  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
1184  it. Setting this without PCRE_MULTILINE (at compile time) causes dollar never  mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1185  to match.  compile time) causes dollar never to match. This option affects only the
1186    behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1187    .sp
1188    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1189    .sp
1190  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
1191  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
1192  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
1193    .sp
1194    a?b?    a?b?
1195    .sp
1196  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
1197  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
1198  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".
1199    .P
1200  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
1201  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
1202  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
1203  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
1204  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
1205  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1206    code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1207  The subject string is passed to \fBpcre_exec()\fR as a pointer in  .sp
1208  \fIsubject\fR, a length in \fIlength\fR, and a starting byte offset in    PCRE_NO_UTF8_CHECK
1209  \fIstartoffset\fR. Unlike the pattern string, the subject may contain binary  .sp
1210  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
1211  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.
1212    The value of \fIstartoffset\fP is also checked to ensure that it points to the
1213  If the pattern was compiled with the PCRE_UTF8 option, the subject must be a  start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,
1214  sequence of bytes that is a valid UTF-8 string, and the starting offset must  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP
1215  point to the beginning of a UTF-8 character. If an invalid UTF-8 string or  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
1216  offset is passed, an error (either PCRE_ERROR_BADUTF8 or  .P
1217  PCRE_ERROR_BADUTF8_OFFSET) is returned, unless the option PCRE_NO_UTF8_CHECK is  If you already know that your subject is valid, and you want to skip these
1218  set, in which case PCRE's behaviour is not defined.  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1219    calling \fBpcre_exec()\fP. You might want to do this for the second and
1220    subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1221    all the matches in a single subject string. However, you should be sure that
1222    the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1223    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1224    subject, or a value of \fIstartoffset\fP that does not point to the start of a
1225    UTF-8 character, is undefined. Your program may crash.
1226    .sp
1227      PCRE_PARTIAL
1228    .sp
1229    This option turns on the partial matching feature. If the subject string fails
1230    to match the pattern, but at some point during the matching process the end of
1231    the subject was reached (that is, the subject partially matches the pattern and
1232    the failure to match occurred only because there were not enough subject
1233    characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1234    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1235    may appear in the pattern. These are discussed in the
1236    .\" HREF
1237    \fBpcrepartial\fP
1238    .\"
1239    documentation.
1240    .
1241    .SS "The string to be matched by \fBpcre_exec()\fP"
1242    .rs
1243    .sp
1244    The subject string is passed to \fBpcre_exec()\fP as a pointer in
1245    \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
1246    \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
1247    UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1248    bytes. When the starting offset is zero, the search for a match starts at the
1249    beginning of the subject, and this is by far the most common case.
1250    .P
1251  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
1252  same subject by calling \fBpcre_exec()\fR again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
1253  Setting \fIstartoffset\fR differs from just passing over a shortened string and  Setting \fIstartoffset\fP differs from just passing over a shortened string and
1254  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
1255  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
1256    .sp
1257    \\Biss\\B    \eBiss\eB
1258    .sp
1259  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
1260  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
1261  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
1262  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
1263  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
1264  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
1265  \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
1266  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
1267  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.
1268    .P
1269  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
1270  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
1271  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.
1272    .
1273    .SS "How \fBpcre_exec()\fP returns captured substrings"
1274    .rs
1275    .sp
1276  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
1277  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
1278  pattern. Following the usage in Jeffrey Friedl's book, this is called  pattern. Following the usage in Jeffrey Friedl's book, this is called
1279  "capturing" in what follows, and the phrase "capturing subpattern" is used for  "capturing" in what follows, and the phrase "capturing subpattern" is used for
1280  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
1281  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1282    .P
1283  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
1284  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
1285  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:
1286  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of \fIovector\fP in bytes.
1287  remaining third of the vector is used as workspace by \fBpcre_exec()\fR while  .P
1288  matching capturing subpatterns, and is not available for passing back  The first two-thirds of the vector is used to pass back captured substrings,
1289  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
1290  three. If it is not, it is rounded down.  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1291    and is not available for passing back information. The length passed in
1292  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
1293  returned in pairs of integers, starting at the beginning of \fIovector\fR, and  rounded down.
1294    .P
1295    When a match is successful, information about captured substrings is returned
1296    in pairs of integers, starting at the beginning of \fIovector\fP, and
1297  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
1298  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
1299  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
1300  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
1301  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
1302  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
1303  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
1304  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
1305  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1306    indicating that just the first pair of offsets has been set.
1307  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.  
   
1308  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
1309  string that it matched that gets returned.  string that it matched that is returned.
1310    .P
1311  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
1312  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
1313  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
1314  \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
1315  \fIovecsize\fR as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1316  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
1317  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
1318  to supply an \fIovector\fR.  advisable to supply an \fIovector\fP.
1319    .P
1320  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
1321  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1322  \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
1323  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.
1324    .P
1325  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
1326    the subject when subpattern \fIn\fP has not been used at all. For example, if
1327    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1328    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1329    happens, both values in the offset pairs corresponding to unused subpatterns
1330    are set to -1.
1331    .P
1332    Offset values that correspond to unused subpatterns at the end of the
1333    expression are also set to -1. For example, if the string "abc" is matched
1334    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1335    return from the function is 2, because the highest used capturing subpattern
1336    number is 1. However, you can refer to the offsets for the second and third
1337    capturing subpatterns if you wish (assuming the vector is large enough, of
1338    course).
1339    .P
1340    Some convenience functions are provided for extracting the captured substrings
1341    as separate strings. These are described below.
1342    .
1343    .\" HTML <a name="errorlist"></a>
1344    .SS "Error return values from \fBpcre_exec()\fP"
1345    .rs
1346    .sp
1347    If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1348  defined in the header file:  defined in the header file:
1349    .sp
1350    PCRE_ERROR_NOMATCH        (-1)    PCRE_ERROR_NOMATCH        (-1)
1351    .sp
1352  The subject string did not match the pattern.  The subject string did not match the pattern.
1353    .sp
1354    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1355    .sp
1356  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
1357  NULL and \fIovecsize\fR was not zero.  NULL and \fIovecsize\fP was not zero.
1358    .sp
1359    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1360    .sp
1361  An unrecognized bit was set in the \fIoptions\fR argument.  An unrecognized bit was set in the \fIoptions\fP argument.
1362    .sp
1363    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1364    .sp
1365  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
1366  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
1367  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1368    other endianness. This is the error that PCRE gives when the magic number is
1369    PCRE_ERROR_UNKNOWN_NODE   (-5)  not present.
1370    .sp
1371      PCRE_ERROR_UNKNOWN_OPCODE (-5)
1372    .sp
1373  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1374  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
1375  of the compiled pattern.  of the compiled pattern.
1376    .sp
1377    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1378    .sp
1379  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
1380  \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
1381  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
1382  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
1383  the end of matching.  automatically freed at the end of matching.
1384    .sp
1385    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1386    .sp
1387  This error is used by the \fBpcre_copy_substring()\fR,  This error is used by the \fBpcre_copy_substring()\fP,
1388  \fBpcre_get_substring()\fR, and \fBpcre_get_substring_list()\fR functions (see  \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1389  below). It is never returned by \fBpcre_exec()\fR.  below). It is never returned by \fBpcre_exec()\fP.
1390    .sp
1391    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1392    .sp
1393  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
1394  field in a \fBpcre_extra\fR structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1395  description above.  above.
1396    .sp
1397    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1398    .sp
1399  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
1400  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
1401  \fBpcrecallout\fR documentation for details.  .\" HREF
1402    \fBpcrecallout\fP
1403    .\"
1404    documentation for details.
1405    .sp
1406    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
1407    .sp
1408  A string that contains an invalid UTF-8 byte sequence was passed as a subject.  A string that contains an invalid UTF-8 byte sequence was passed as a subject.
1409    .sp
1410    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
1411    .sp
1412  The UTF-8 byte sequence that was passed as a subject was valid, but the value  The UTF-8 byte sequence that was passed as a subject was valid, but the value
1413  of \fIstartoffset\fR did not point to the beginning of a UTF-8 character.  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1414    .sp
1415  .SH EXTRACTING CAPTURED SUBSTRINGS BY NUMBER    PCRE_ERROR_PARTIAL        (-12)
1416    .sp
1417    The subject string did not match, but it did match partially. See the
1418    .\" HREF
1419    \fBpcrepartial\fP
1420    .\"
1421    documentation for details of partial matching.
1422    .sp
1423      PCRE_ERROR_BADPARTIAL     (-13)
1424    .sp
1425    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1426    are not supported for partial matching. See the
1427    .\" HREF
1428    \fBpcrepartial\fP
1429    .\"
1430    documentation for details of partial matching.
1431    .sp
1432      PCRE_ERROR_INTERNAL       (-14)
1433    .sp
1434    An unexpected internal error has occurred. This error could be caused by a bug
1435    in PCRE or by overwriting of the compiled pattern.
1436    .sp
1437      PCRE_ERROR_BADCOUNT       (-15)
1438    .sp
1439    This error is given if the value of the \fIovecsize\fP argument is negative.
1440    .sp
1441      PCRE_ERROR_RECURSIONLIMIT (-21)
1442    .sp
1443    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1444    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1445    description above.
1446    .sp
1447      PCRE_ERROR_NULLWSLIMIT    (-22)
1448    .sp
1449    When a group that can match an empty substring is repeated with an unbounded
1450    upper limit, the subject position at the start of the group must be remembered,
1451    so that a test for an empty string can be made when the end of the group is
1452    reached. Some workspace is required for this; if it runs out, this error is
1453    given.
1454    .sp
1455      PCRE_ERROR_BADNEWLINE     (-23)
1456    .sp
1457    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1458    .P
1459    Error numbers -16 to -20 are not used by \fBpcre_exec()\fP.
1460    .
1461    .
1462    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1463  .rs  .rs
1464  .sp  .sp
1465  .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1466  .ti +5n  .ti +5n
1467  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1468  .ti +5n  .ti +5n
1469  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
1470  .PP  .PP
1471  .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,  
1472  .ti +5n  .ti +5n
1473  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1474  .ti +5n  .ti +5n
1475  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1476  .PP  .PP
1477  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
1478  .ti +5n  .ti +5n
1479  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1480  .PP  .PP
1481  Captured substrings can be accessed directly by using the offsets returned by  Captured substrings can be accessed directly by using the offsets returned by
1482  \fBpcre_exec()\fR in \fIovector\fR. For convenience, the functions  \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1483  \fBpcre_copy_substring()\fR, \fBpcre_get_substring()\fR, and  \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1484  \fBpcre_get_substring_list()\fR are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1485  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1486  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1487  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1488  has a further zero added on the end, but the result is not, of course,  .P
1489  a C string.  A substring that contains a binary zero is correctly extracted and has a
1490    further zero added on the end, but the result is not, of course, a C string.
1491    However, you can process such a string by referring to the length that is
1492    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1493    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1494    for handling strings containing binary zeros, because the end of the final
1495    string is not independently indicated.
1496    .P
1497  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:
1498  \fIsubject\fR is the subject string which has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
1499  \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
1500  \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
1501  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1502  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
1503  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
1504  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
1505  the vector divided by three.  number of elements in the vector divided by three.
1506    .P
1507  The functions \fBpcre_copy_substring()\fR and \fBpcre_get_substring()\fR  The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1508  extract a single substring, whose number is given as \fIstringnumber\fR. A  extract a single substring, whose number is given as \fIstringnumber\fP. A
1509  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1510  higher values extract the captured substrings. For \fBpcre_copy_substring()\fR,  higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1511  the string is placed in \fIbuffer\fR, whose length is given by  the string is placed in \fIbuffer\fP, whose length is given by
1512  \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
1513  obtained via \fBpcre_malloc\fR, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1514  \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
1515  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1516    .sp
1517    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1518    .sp
1519  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
1520  memory failed for \fBpcre_get_substring()\fR.  memory failed for \fBpcre_get_substring()\fP.
1521    .sp
1522    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1523    .sp
1524  There is no substring whose number is \fIstringnumber\fR.  There is no substring whose number is \fIstringnumber\fP.
1525    .P
1526  The \fBpcre_get_substring_list()\fR function extracts all available substrings  The \fBpcre_get_substring_list()\fP function extracts all available substrings
1527  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
1528  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
1529  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
1530  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
1531  function is zero if all went well, or  function is zero if all went well, or the error code
1532    .sp
1533    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1534    .sp
1535  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1536    .P
1537  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
1538  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
1539  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
1540  string. This can be distinguished from a genuine zero-length substring by  string. This can be distinguished from a genuine zero-length substring by
1541  inspecting the appropriate offset in \fIovector\fR, which is negative for unset  inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1542  substrings.  substrings.
1543    .P
1544  The two convenience functions \fBpcre_free_substring()\fR and  The two convenience functions \fBpcre_free_substring()\fP and
1545  \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
1546  a previous call of \fBpcre_get_substring()\fR or  a previous call of \fBpcre_get_substring()\fP or
1547  \fBpcre_get_substring_list()\fR, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1548  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
1549  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
1550  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1551  \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
1552  provided.  provided.
1553    .
1554  .SH EXTRACTING CAPTURED SUBSTRINGS BY NAME  .
1555    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1556  .rs  .rs
1557  .sp  .sp
1558  .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1559  .ti +5n  .ti +5n
1560  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIname\fP);
1561    .PP
1562    .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1563  .ti +5n  .ti +5n
1564  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1565  .ti +5n  .ti +5n
1566  .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,  
1567  .ti +5n  .ti +5n
1568  .B const char *\fIname\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1569  .PP  .PP
1570  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
 .B int pcre_get_named_substring(const pcre *\fIcode\fR,  
1571  .ti +5n  .ti +5n
1572  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1573  .ti +5n  .ti +5n
1574  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
1575  .ti +5n  .ti +5n
1576  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1577  .PP  .PP
1578  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.
1579  can be done by calling \fBpcre_get_stringnumber()\fR. The first argument is the  For example, for this pattern
1580  compiled pattern, and the second is the name. For example, for this pattern  .sp
1581      (a+)b(?<xxx>\ed+)...
1582    ab(?<xxx>\\d+)...  .sp
1583    the number of the subpattern called "xxx" is 2. If the name is known to be
1584  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
1585  extract the substring directly, or use one of the functions described in the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1586  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
1587  whole job.  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1588    that name.
1589  Most of the arguments of \fIpcre_copy_named_substring()\fR and  .P
1590  \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
1591  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
1592  differences.  two functions that do the whole job.
1593    .P
1594    Most of the arguments of \fBpcre_copy_named_substring()\fP and
1595    \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1596    functions that extract by number. As these are described in the previous
1597    section, they are not re-described here. There are just two differences:
1598    .P
1599  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
1600  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
1601  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
1602  translation table.  translation table.
1603    .P
1604  These functions call \fBpcre_get_stringnumber()\fR, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1605  then call \fIpcre_copy_substring()\fR or \fIpcre_get_substring()\fR, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1606  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1607    the behaviour may not be what you want (see the next section).
1608  .in 0  .
1609  Last updated: 09 December 2003  .
1610  .br  .SH "DUPLICATE SUBPATTERN NAMES"
1611  Copyright (c) 1997-2003 University of Cambridge.  .rs
1612    .sp
1613    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1614    .ti +5n
1615    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1616    .PP
1617    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1618    are not required to be unique. Normally, patterns with duplicate names are such
1619    that in any one match, only one of the named subpatterns participates. An
1620    example is shown in the
1621    .\" HREF
1622    \fBpcrepattern\fP
1623    .\"
1624    documentation. When duplicates are present, \fBpcre_copy_named_substring()\fP
1625    and \fBpcre_get_named_substring()\fP return the first substring corresponding
1626    to the given name that is set. If none are set, an empty string is returned.
1627    The \fBpcre_get_stringnumber()\fP function returns one of the numbers that are
1628    associated with the name, but it is not defined which it is.
1629    .sp
1630    If you want to get full details of all captured substrings for a given name,
1631    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1632    argument is the compiled pattern, and the second is the name. The third and
1633    fourth are pointers to variables which are updated by the function. After it
1634    has run, they point to the first and last entries in the name-to-number table
1635    for the given name. The function itself returns the length of each entry, or
1636    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1637    described above in the section entitled \fIInformation about a pattern\fP.
1638    Given all the relevant entries for the name, you can extract each of their
1639    numbers, and hence the captured data, if any.
1640    .
1641    .
1642    .SH "FINDING ALL POSSIBLE MATCHES"
1643    .rs
1644    .sp
1645    The traditional matching function uses a similar algorithm to Perl, which stops
1646    when it finds the first match, starting at a given point in the subject. If you
1647    want to find all possible matches, or the longest possible match, consider
1648    using the alternative matching function (see below) instead. If you cannot use
1649    the alternative function, but still need to find all possible matches, you
1650    can kludge it up by making use of the callout facility, which is described in
1651    the
1652    .\" HREF
1653    \fBpcrecallout\fP
1654    .\"
1655    documentation.
1656    .P
1657    What you have to do is to insert a callout right at the end of the pattern.
1658    When your callout function is called, extract and save the current matched
1659    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1660    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1661    will yield PCRE_ERROR_NOMATCH.
1662    .
1663    .
1664    .\" HTML <a name="dfamatch"></a>
1665    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1666    .rs
1667    .sp
1668    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1669    .ti +5n
1670    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1671    .ti +5n
1672    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1673    .ti +5n
1674    .B int *\fIworkspace\fP, int \fIwscount\fP);
1675    .P
1676    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1677    a compiled pattern, using a matching algorithm that scans the subject string
1678    just once, and does not backtrack. This has different characteristics to the
1679    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1680    patterns are not supported. Nevertheless, there are times when this kind of
1681    matching can be useful. For a discussion of the two matching algorithms, see
1682    the
1683    .\" HREF
1684    \fBpcrematching\fP
1685    .\"
1686    documentation.
1687    .P
1688    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1689    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1690    different way, and this is described below. The other common arguments are used
1691    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1692    here.
1693    .P
1694    The two additional arguments provide workspace for the function. The workspace
1695    vector should contain at least 20 elements. It is used for keeping track of
1696    multiple paths through the pattern tree. More workspace will be needed for
1697    patterns and subjects where there are a lot of potential matches.
1698    .P
1699    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1700    .sp
1701      int rc;
1702      int ovector[10];
1703      int wspace[20];
1704      rc = pcre_dfa_exec(
1705        re,             /* result of pcre_compile() */
1706        NULL,           /* we didn't study the pattern */
1707        "some string",  /* the subject string */
1708        11,             /* the length of the subject string */
1709        0,              /* start at offset 0 in the subject */
1710        0,              /* default options */
1711        ovector,        /* vector of integers for substring information */
1712        10,             /* number of elements (NOT size in bytes) */
1713        wspace,         /* working space vector */
1714        20);            /* number of elements (NOT size in bytes) */
1715    .
1716    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1717    .rs
1718    .sp
1719    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1720    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1721    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1722    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1723    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1724    .sp
1725      PCRE_PARTIAL
1726    .sp
1727    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1728    details are slightly different. When PCRE_PARTIAL is set for
1729    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1730    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1731    complete matches, but there is still at least one matching possibility. The
1732    portion of the string that provided the partial match is set as the first
1733    matching string.
1734    .sp
1735      PCRE_DFA_SHORTEST
1736    .sp
1737    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1738    soon as it has found one match. Because of the way the alternative algorithm
1739    works, this is necessarily the shortest possible match at the first possible
1740    matching point in the subject string.
1741    .sp
1742      PCRE_DFA_RESTART
1743    .sp
1744    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1745    a partial match, it is possible to call it again, with additional subject
1746    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1747    option requests this action; when it is set, the \fIworkspace\fP and
1748    \fIwscount\fP options must reference the same vector as before because data
1749    about the match so far is left in them after a partial match. There is more
1750    discussion of this facility in the
1751    .\" HREF
1752    \fBpcrepartial\fP
1753    .\"
1754    documentation.
1755    .
1756    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1757    .rs
1758    .sp
1759    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1760    substring in the subject. Note, however, that all the matches from one run of
1761    the function start at the same point in the subject. The shorter matches are
1762    all initial substrings of the longer matches. For example, if the pattern
1763    .sp
1764      <.*>
1765    .sp
1766    is matched against the string
1767    .sp
1768      This is <something> <something else> <something further> no more
1769    .sp
1770    the three matched strings are
1771    .sp
1772      <something>
1773      <something> <something else>
1774      <something> <something else> <something further>
1775    .sp
1776    On success, the yield of the function is a number greater than zero, which is
1777    the number of matched substrings. The substrings themselves are returned in
1778    \fIovector\fP. Each string uses two elements; the first is the offset to the
1779    start, and the second is the offset to the end. In fact, all the strings have
1780    the same start offset. (Space could have been saved by giving this only once,
1781    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1782    returns data, even though the meaning of the strings is different.)
1783    .P
1784    The strings are returned in reverse order of length; that is, the longest
1785    matching string is given first. If there were too many matches to fit into
1786    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1787    the longest matches.
1788    .
1789    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1790    .rs
1791    .sp
1792    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1793    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1794    described
1795    .\" HTML <a href="#errorlist">
1796    .\" </a>
1797    above.
1798    .\"
1799    There are in addition the following errors that are specific to
1800    \fBpcre_dfa_exec()\fP:
1801    .sp
1802      PCRE_ERROR_DFA_UITEM      (-16)
1803    .sp
1804    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1805    that it does not support, for instance, the use of \eC or a back reference.
1806    .sp
1807      PCRE_ERROR_DFA_UCOND      (-17)
1808    .sp
1809    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1810    uses a back reference for the condition, or a test for recursion in a specific
1811    group. These are not supported.
1812    .sp
1813      PCRE_ERROR_DFA_UMLIMIT    (-18)
1814    .sp
1815    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1816    block that contains a setting of the \fImatch_limit\fP field. This is not
1817    supported (it is meaningless).
1818    .sp
1819      PCRE_ERROR_DFA_WSSIZE     (-19)
1820    .sp
1821    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1822    \fIworkspace\fP vector.
1823    .sp
1824      PCRE_ERROR_DFA_RECURSE    (-20)
1825    .sp
1826    When a recursive subpattern is processed, the matching function calls itself
1827    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1828    error is given if the output vector is not large enough. This should be
1829    extremely rare, as a vector of size 1000 is used.
1830    .
1831    .
1832    .SH "SEE ALSO"
1833    .rs
1834    .sp
1835    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1836    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1837    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1838    .
1839    .
1840    .SH AUTHOR
1841    .rs
1842    .sp
1843    .nf
1844    Philip Hazel
1845    University Computing Service
1846    Cambridge CB2 3QH, England.
1847    .fi
1848    .
1849    .
1850    .SH REVISION
1851    .rs
1852    .sp
1853    .nf
1854    Last updated: 24 April 2007
1855    Copyright (c) 1997-2007 University of Cambridge.
1856    .fi

Legend:
Removed from v.73  
changed lines
  Added in v.155

  ViewVC Help
Powered by ViewVC 1.1.5