/[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 71 by nigel, Sat Feb 24 21:40:24 2007 UTC revision 226 by ph10, Tue Aug 21 11:46:08 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
105  .br  .B void *(*pcre_stack_malloc)(size_t);
106    .PP
107    .B void (*pcre_stack_free)(void *);
108    .PP
109  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
110    .
111  .SH PCRE API  .
112    .SH "PCRE API OVERVIEW"
113  .rs  .rs
114  .sp  .sp
115  PCRE has its own native API, which is described in this document. There is also  PCRE has its own native API, which is described in this document. There are
116  a set of wrapper functions that correspond to the POSIX regular expression API.  also some wrapper functions that correspond to the POSIX regular expression
117  These are described in the \fBpcreposix\fR documentation.  API. These are described in the
118    .\" HREF
119  The native API function prototypes are defined in the header file \fBpcre.h\fR,  \fBpcreposix\fP
120  and on Unix systems the library itself is called \fBlibpcre.a\fR, so can be  .\"
121  accessed by adding \fB-lpcre\fR to the command for linking an application which  documentation. Both of these APIs define a set of C function calls. A C++
122  calls it. The header file defines the macros PCRE_MAJOR and PCRE_MINOR to  wrapper is distributed with PCRE. It is documented in the
123  contain the major and minor release numbers for the library. Applications can  .\" HREF
124  use these to include support for different releases.  \fBpcrecpp\fP
125    .\"
126  The functions \fBpcre_compile()\fR, \fBpcre_study()\fR, and \fBpcre_exec()\fR  page.
127  are used for compiling and matching regular expressions. A sample program that  .P
128  demonstrates the simplest way of using them is given in the file  The native API C function prototypes are defined in the header file
129  \fIpcredemo.c\fR. The \fBpcresample\fR documentation describes how to run it.  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130    It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131  There are convenience functions for extracting captured substrings from a  an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132  matched subject string. They are:  and PCRE_MINOR to contain the major and minor release numbers for the library.
133    Applications can use these to include support for different releases of PCRE.
134    \fBpcre_copy_substring()\fR  .P
135    \fBpcre_copy_named_substring()\fR  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136    \fBpcre_get_substring()\fR  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137    \fBpcre_get_named_substring()\fR  in a Perl-compatible manner. A sample program that demonstrates the simplest
138    \fBpcre_get_substring_list()\fR  way of using them is provided in the file called \fIpcredemo.c\fP in the source
139    distribution. The
140  \fBpcre_free_substring()\fR and \fBpcre_free_substring_list()\fR are also  .\" HREF
141    \fBpcresample\fP
142    .\"
143    documentation describes how to run it.
144    .P
145    A second matching function, \fBpcre_dfa_exec()\fP, which is not
146    Perl-compatible, is also provided. This uses a different algorithm for the
147    matching. The alternative algorithm finds all possible matches (at a given
148    point in the subject), and scans the subject just once. However, this algorithm
149    does not return captured substrings. A description of the two matching
150    algorithms and their advantages and disadvantages is given in the
151    .\" HREF
152    \fBpcrematching\fP
153    .\"
154    documentation.
155    .P
156    In addition to the main compiling and matching functions, there are convenience
157    functions for extracting captured substrings from a subject string that is
158    matched by \fBpcre_exec()\fP. They are:
159    .sp
160      \fBpcre_copy_substring()\fP
161      \fBpcre_copy_named_substring()\fP
162      \fBpcre_get_substring()\fP
163      \fBpcre_get_named_substring()\fP
164      \fBpcre_get_substring_list()\fP
165      \fBpcre_get_stringnumber()\fP
166      \fBpcre_get_stringtable_entries()\fP
167    .sp
168    \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
169  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
170    .P
171  The function \fBpcre_maketables()\fR is used (optionally) to build a set of  The function \fBpcre_maketables()\fP is used to build a set of character tables
172  character tables in the current locale for passing to \fBpcre_compile()\fR.  in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
173    or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
174  The function \fBpcre_fullinfo()\fR is used to find out information about a  specialist use. Most commonly, no special tables are passed, in which case
175  compiled pattern; \fBpcre_info()\fR is an obsolete version which returns only  internal tables that are generated when PCRE is built are used.
176    .P
177    The function \fBpcre_fullinfo()\fP is used to find out information about a
178    compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
179  some of the available information, but is retained for backwards compatibility.  some of the available information, but is retained for backwards compatibility.
180  The function \fBpcre_version()\fR returns a pointer to a string containing the  The function \fBpcre_version()\fP returns a pointer to a string containing the
181  version of PCRE and its date of release.  version of PCRE and its date of release.
182    .P
183  The global variables \fBpcre_malloc\fR and \fBpcre_free\fR initially contain  The function \fBpcre_refcount()\fP maintains a reference count in a data block
184  the entry points of the standard \fBmalloc()\fR and \fBfree()\fR functions  containing a compiled pattern. This is provided for the benefit of
185    object-oriented applications.
186    .P
187    The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
188    the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
189  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
190  so a calling program can replace them if it wishes to intercept the calls. This  so a calling program can replace them if it wishes to intercept the calls. This
191  should be done before calling any PCRE functions.  should be done before calling any PCRE functions.
192    .P
193  The global variable \fBpcre_callout\fR initially contains NULL. It can be set  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
194    indirections to memory management functions. These special functions are used
195    only when PCRE is compiled to use the heap for remembering data, instead of
196    recursive function calls, when running the \fBpcre_exec()\fP function. See the
197    .\" HREF
198    \fBpcrebuild\fP
199    .\"
200    documentation for details of how to do this. It is a non-standard way of
201    building PCRE, for use in environments that have limited stacks. Because of the
202    greater use of memory management, it runs more slowly. Separate functions are
203    provided so that special-purpose external code can be used for this case. When
204    used, these functions are always called in a stack-like manner (last obtained,
205    first freed), and always for memory blocks of the same size. There is a
206    discussion about PCRE's stack usage in the
207    .\" HREF
208    \fBpcrestack\fP
209    .\"
210    documentation.
211    .P
212    The global variable \fBpcre_callout\fP initially contains NULL. It can be set
213  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
214  points during a matching operation. Details are given in the \fBpcrecallout\fR  points during a matching operation. Details are given in the
215    .\" HREF
216    \fBpcrecallout\fP
217    .\"
218  documentation.  documentation.
219    .
220    .
221    .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. There is more detail about this in the
244    .\" HTML <a href="#execoptions">
245    .\" </a>
246    section on \fBpcre_exec()\fP options
247    .\"
248    below. The choice of newline convention does not affect the interpretation of
249    the \en or \er escape sequences.
250    .
251    .
252  .SH MULTITHREADING  .SH MULTITHREADING
253  .rs  .rs
254  .sp  .sp
255  The PCRE functions can be used in multi-threading applications, with the  The PCRE functions can be used in multi-threading applications, with the
256  proviso that the memory management functions pointed to by \fBpcre_malloc\fR  proviso that the memory management functions pointed to by \fBpcre_malloc\fP,
257  and \fBpcre_free\fR, and the callout function pointed to by \fBpcre_callout\fR,  \fBpcre_free\fP, \fBpcre_stack_malloc\fP, and \fBpcre_stack_free\fP, and the
258  are shared by all threads.  callout function pointed to by \fBpcre_callout\fP, are shared by all threads.
259    .P
260  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
261  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.
262    .
263  .SH CHECKING BUILD-TIME OPTIONS  .
264    .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
265  .rs  .rs
266  .sp  .sp
267  .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
268    time, possibly by a different program, and even on a host other than the one on
269    which it was compiled. Details are given in the
270    .\" HREF
271    \fBpcreprecompile\fP
272    .\"
273    documentation. However, compiling a regular expression with one version of PCRE
274    for use with a different version is not guaranteed to work and may cause
275    crashes.
276    .
277    .
278    .SH "CHECKING BUILD-TIME OPTIONS"
279    .rs
280    .sp
281    .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
282  .PP  .PP
283  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
284  discover which optional features have been compiled into the PCRE library. The  discover which optional features have been compiled into the PCRE library. The
285  .\" HREF  .\" HREF
286  \fBpcrebuild\fR  \fBpcrebuild\fP
287  .\"  .\"
288  documentation has more details about these optional features.  documentation has more details about these optional features.
289    .P
290  The first argument for \fBpcre_config()\fR is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
291  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
292  which the information is placed. The following information is available:  which the information is placed. The following information is available:
293    .sp
294    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
295    .sp
296  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;
297  otherwise it is set to zero.  otherwise it is set to zero.
298    .sp
299      PCRE_CONFIG_UNICODE_PROPERTIES
300    .sp
301    The output is an integer that is set to one if support for Unicode character
302    properties is available; otherwise it is set to zero.
303    .sp
304    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
305    .sp
306  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
307  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
308  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
309    default should normally be the standard sequence for your operating system.
310    .sp
311    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
312    .sp
313  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
314  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
315  allow larger regular expressions to be compiled, at the expense of slower  allow larger regular expressions to be compiled, at the expense of slower
316  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
317  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.
318    .sp
319    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
320    .sp
321  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
322  interface uses \fBmalloc()\fR for output vectors. Further details are given in  interface uses \fBmalloc()\fP for output vectors. Further details are given in
323  the \fBpcreposix\fR documentation.  the
324    .\" HREF
325    \fBpcreposix\fP
326    .\"
327    documentation.
328    .sp
329    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
330    .sp
331  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
332  internal matching function calls in a \fBpcre_exec()\fR execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
333  details are given with \fBpcre_exec()\fR below.  details are given with \fBpcre_exec()\fP below.
334    .sp
335  .SH COMPILING A PATTERN    PCRE_CONFIG_MATCH_LIMIT_RECURSION
336  .rs  .sp
337  .sp  The output is an integer that gives the default limit for the depth of
338  .B pcre *pcre_compile(const char *\fIpattern\fR, int \fIoptions\fR,  recursion when calling the internal matching function in a \fBpcre_exec()\fP
339  .ti +5n  execution. Further details are given with \fBpcre_exec()\fP below.
340  .B const char **\fIerrptr\fR, int *\fIerroffset\fR,  .sp
341  .ti +5n    PCRE_CONFIG_STACKRECURSE
342  .B const unsigned char *\fItableptr\fR);  .sp
343  .PP  The output is an integer that is set to one if internal recursion when running
344    \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
345  The function \fBpcre_compile()\fR is called to compile a pattern into an  to remember their state. This is the usual way that PCRE is compiled. The
346  internal form. The pattern is a C string terminated by a binary zero, and  output is zero if PCRE was compiled to use blocks of data on the heap instead
347  is passed in the argument \fIpattern\fR. A pointer to a single block of memory  of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
348  that is obtained via \fBpcre_malloc\fR is returned. This contains the compiled  \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
349  code and related data. The \fBpcre\fR type is defined for the returned block;  avoiding the use of the stack.
350  this is a typedef for a structure whose contents are not externally defined. It  .
351  is up to the caller to free the memory when it is no longer required.  .
352    .SH "COMPILING A PATTERN"
353    .rs
354    .sp
355    .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
356    .ti +5n
357    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
358    .ti +5n
359    .B const unsigned char *\fItableptr\fP);
360    .sp
361    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
362    .ti +5n
363    .B int *\fIerrorcodeptr\fP,
364    .ti +5n
365    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
366    .ti +5n
367    .B const unsigned char *\fItableptr\fP);
368    .P
369    Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
370    called to compile a pattern into an internal form. The only difference between
371    the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
372    \fIerrorcodeptr\fP, via which a numerical error code can be returned.
373    .P
374    The pattern is a C string terminated by a binary zero, and is passed in the
375    \fIpattern\fP argument. A pointer to a single block of memory that is obtained
376    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
377    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
378    for a structure whose contents are not externally defined. It is up to the
379    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
380    .P
381  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
382  depend on memory location, the complete \fBpcre\fR data block is not  depend on memory location, the complete \fBpcre\fP data block is not
383  fully relocatable, because it contains a copy of the \fItableptr\fR argument,  fully relocatable, because it may contain a copy of the \fItableptr\fP
384  which is an address (see below).  argument, which is an address (see below).
385    .P
386  The \fIoptions\fR argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
387  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
388  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
389  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
390  in the \fBpcrepattern\fR documentation). For these options, the contents of the  the detailed description in the
391  \fIoptions\fR argument specifies their initial settings at the start of  .\" HREF
392  compilation and execution. The PCRE_ANCHORED option can be set at the time of  \fBpcrepattern\fP
393    .\"
394    documentation). For these options, the contents of the \fIoptions\fP argument
395    specifies their initial settings at the start of compilation and execution. The
396    PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
397  matching as well as at compile time.  matching as well as at compile time.
398    .P
399  If \fIerrptr\fR is NULL, \fBpcre_compile()\fR returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
400  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fR returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
401  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
402  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
403  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
404  \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
405    \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
406  If the final argument, \fItableptr\fR, is NULL, PCRE uses a default set of  .P
407  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
408  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
409  \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
410    textual error message. Error codes and messages are listed below.
411  This code fragment shows a typical straightforward call to \fBpcre_compile()\fR:  .P
412    If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
413    character tables that are built when PCRE is compiled, using the default C
414    locale. Otherwise, \fItableptr\fP must be an address that is the result of a
415    call to \fBpcre_maketables()\fP. This value is stored with the compiled
416    pattern, and used again by \fBpcre_exec()\fP, unless another table pointer is
417    passed to it. For more discussion, see the section on locale support below.
418    .P
419    This code fragment shows a typical straightforward call to \fBpcre_compile()\fP:
420    .sp
421    pcre *re;    pcre *re;
422    const char *error;    const char *error;
423    int erroffset;    int erroffset;
# Line 265  This code fragment shows a typical strai Line 427  This code fragment shows a typical strai
427      &error,           /* for error message */      &error,           /* for error message */
428      &erroffset,       /* for error offset */      &erroffset,       /* for error offset */
429      NULL);            /* use default character tables */      NULL);            /* use default character tables */
430    .sp
431  The following option bits are defined:  The following names for option bits are defined in the \fBpcre.h\fP header
432    file:
433    .sp
434    PCRE_ANCHORED    PCRE_ANCHORED
435    .sp
436  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
437  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
438  being searched (the "subject string"). This effect can also be achieved by  being searched (the "subject string"). This effect can also be achieved by
439  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
440  Perl.  Perl.
441    .sp
442      PCRE_AUTO_CALLOUT
443    .sp
444    If this bit is set, \fBpcre_compile()\fP automatically inserts callout items,
445    all with number 255, before each pattern item. For discussion of the callout
446    facility, see the
447    .\" HREF
448    \fBpcrecallout\fP
449    .\"
450    documentation.
451    .sp
452    PCRE_CASELESS    PCRE_CASELESS
453    .sp
454  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
455  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
456  pattern by a (?i) option setting.  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
457    concept of case for characters whose values are less than 128, so caseless
458    matching is always possible. For characters with higher values, the concept of
459    case is supported if PCRE is compiled with Unicode property support, but not
460    otherwise. If you want to use caseless matching for characters 128 and above,
461    you must ensure that PCRE is compiled with Unicode property support as well as
462    with UTF-8 support.
463    .sp
464    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
465    .sp
466  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
467  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
468  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
469  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.
470  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
471  a pattern.  pattern.
472    .sp
473    PCRE_DOTALL    PCRE_DOTALL
474    .sp
475  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,
476  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
477  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
478  (?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
479  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
480    the setting of this option.
481    .sp
482      PCRE_DUPNAMES
483    .sp
484    If this bit is set, names used to identify capturing subpatterns need not be
485    unique. This can be helpful for certain types of pattern when it is known that
486    only one instance of the named subpattern can ever be matched. There are more
487    details of named subpatterns below; see also the
488    .\" HREF
489    \fBpcrepattern\fP
490    .\"
491    documentation.
492    .sp
493    PCRE_EXTENDED    PCRE_EXTENDED
494    .sp
495  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
496  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
497  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
498  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
499  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
500  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
501    .P
502  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
503  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
504  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
505  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( which introduces a conditional subpattern.
506    .sp
507    PCRE_EXTRA    PCRE_EXTRA
508    .sp
509  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
510  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
511  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
512  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
513  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
514  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
515  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
516  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
517    .sp
518      PCRE_FIRSTLINE
519    .sp
520    If this option is set, an unanchored pattern is required to match before or at
521    the first newline in the subject string, though the matched text may continue
522    over the newline.
523    .sp
524    PCRE_MULTILINE    PCRE_MULTILINE
525    .sp
526  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
527  characters (even if it actually contains several newlines). The "start of line"  characters (even if it actually contains newlines). The "start of line"
528  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
529  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
530  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
531  Perl.  Perl.
532    .P
533  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
534  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
535  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
536  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
537  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
538  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
539    .sp
540      PCRE_NEWLINE_CR
541      PCRE_NEWLINE_LF
542      PCRE_NEWLINE_CRLF
543      PCRE_NEWLINE_ANYCRLF
544      PCRE_NEWLINE_ANY
545    .sp
546    These options override the default newline definition that was chosen when PCRE
547    was built. Setting the first or the second specifies that a newline is
548    indicated by a single character (CR or LF, respectively). Setting
549    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
550    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
551    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
552    that any Unicode newline sequence should be recognized. The Unicode newline
553    sequences are the three just mentioned, plus the single characters VT (vertical
554    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
555    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
556    recognized only in UTF-8 mode.
557    .P
558    The newline setting in the options word uses three bits that are treated
559    as a number, giving eight possibilities. Currently only six are used (default
560    plus the five values above). This means that if you set more than one newline
561    option, the combination may or may not be sensible. For example,
562    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
563    other combinations may yield unused numbers and cause an error.
564    .P
565    The only time that a line break is specially recognized when compiling a
566    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
567    class is encountered. This indicates a comment that lasts until after the next
568    line break sequence. In other circumstances, line break sequences are treated
569    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
570    as whitespace characters and are therefore ignored.
571    .P
572    The newline option that is set at compile time becomes the default that is used
573    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
574    .sp
575    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
576    .sp
577  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
578  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
579  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
580  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
581  in Perl.  in Perl.
582    .sp
583    PCRE_UNGREEDY    PCRE_UNGREEDY
584    .sp
585  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
586  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
587  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.
588    .sp
589    PCRE_UTF8    PCRE_UTF8
590    .sp
591  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
592  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
593  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
594  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
595  behaviour of PCRE are given in the  behaviour of PCRE are given in the
596  .\" HTML <a href="pcre.html#utf8support">  .\" HTML <a href="pcre.html#utf8support">
# Line 367  section on UTF-8 support Line 599  section on UTF-8 support
599  .\"  .\"
600  in the main  in the main
601  .\" HREF  .\" HREF
602  \fBpcre\fR  \fBpcre\fP
603  .\"  .\"
604  page.  page.
605    .sp
606    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
607    .sp
608  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
609  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
610  \fBpcre_compile()\fR returns an error. If you already know that your pattern is  .\" HTML <a href="pcre.html#utf8strings">
611  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
612  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
613  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
614  Note that there is a similar option for suppressing the checking of subject  in the main
615  strings passed to \fBpcre_exec()\fR.  .\" HREF
616    \fBpcre\fP
617    .\"
618  .SH STUDYING A PATTERN  page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
619  .rs  returns an error. If you already know that your pattern is valid, and you want
620  .sp  to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
621  .B pcre_extra *pcre_study(const pcre *\fIcode\fR, int \fIoptions\fR,  option. When it is set, the effect of passing an invalid UTF-8 string as a
622  .ti +5n  pattern is undefined. It may cause your program to crash. Note that this option
623  .B const char **\fIerrptr\fR);  can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
624  .PP  the UTF-8 validity checking of subject strings.
625  When a pattern is going to be used several times, it is worth spending more  .
626  time analyzing it in order to speed up the time taken for matching. The  .
627  function \fBpcre_study()\fR takes a pointer to a compiled pattern as its first  .SH "COMPILATION ERROR CODES"
628  argument. If studing the pattern produces additional information that will help  .rs
629  speed up matching, \fBpcre_study()\fR returns a pointer to a \fBpcre_extra\fR  .sp
630  block, in which the \fIstudy_data\fR field points to the results of the study.  The following table lists the error codes than may be returned by
631    \fBpcre_compile2()\fP, along with the error messages that may be returned by
632  The returned value from a \fBpcre_study()\fR can be passed directly to  both compiling functions. As PCRE has developed, some error codes have fallen
633  \fBpcre_exec()\fR. However, the \fBpcre_extra\fR block also contains other  out of use. To avoid confusion, they have not been re-used.
634    .sp
635       0  no error
636       1  \e at end of pattern
637       2  \ec at end of pattern
638       3  unrecognized character follows \e
639       4  numbers out of order in {} quantifier
640       5  number too big in {} quantifier
641       6  missing terminating ] for character class
642       7  invalid escape sequence in character class
643       8  range out of order in character class
644       9  nothing to repeat
645      10  [this code is not in use]
646      11  internal error: unexpected repeat
647      12  unrecognized character after (?
648      13  POSIX named classes are supported only within a class
649      14  missing )
650      15  reference to non-existent subpattern
651      16  erroffset passed as NULL
652      17  unknown option bit(s) set
653      18  missing ) after comment
654      19  [this code is not in use]
655      20  regular expression too large
656      21  failed to get memory
657      22  unmatched parentheses
658      23  internal error: code overflow
659      24  unrecognized character after (?<
660      25  lookbehind assertion is not fixed length
661      26  malformed number or name after (?(
662      27  conditional group contains more than two branches
663      28  assertion expected after (?(
664      29  (?R or (?[+-]digits must be followed by )
665      30  unknown POSIX class name
666      31  POSIX collating elements are not supported
667      32  this version of PCRE is not compiled with PCRE_UTF8 support
668      33  [this code is not in use]
669      34  character value in \ex{...} sequence is too large
670      35  invalid condition (?(0)
671      36  \eC not allowed in lookbehind assertion
672      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
673      38  number after (?C is > 255
674      39  closing ) for (?C expected
675      40  recursive call could loop indefinitely
676      41  unrecognized character after (?P
677      42  syntax error in subpattern name (missing terminator)
678      43  two named subpatterns have the same name
679      44  invalid UTF-8 string
680      45  support for \eP, \ep, and \eX has not been compiled
681      46  malformed \eP or \ep sequence
682      47  unknown property name after \eP or \ep
683      48  subpattern name is too long (maximum 32 characters)
684      49  too many named subpatterns (maximum 10,000)
685      50  [this code is not in use]
686      51  octal value is greater than \e377 (not in UTF-8 mode)
687      52  internal error: overran compiling workspace
688      53  internal error: previously-checked referenced subpattern not found
689      54  DEFINE group contains more than one branch
690      55  repeating a DEFINE group is not allowed
691      56  inconsistent NEWLINE options"
692      57  \eg is not followed by a braced name or an optionally braced
693            non-zero number
694      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
695    .
696    .
697    .SH "STUDYING A PATTERN"
698    .rs
699    .sp
700    .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
701    .ti +5n
702    .B const char **\fIerrptr\fP);
703    .PP
704    If a compiled pattern is going to be used several times, it is worth spending
705    more time analyzing it in order to speed up the time taken for matching. The
706    function \fBpcre_study()\fP takes a pointer to a compiled pattern as its first
707    argument. If studying the pattern produces additional information that will
708    help speed up matching, \fBpcre_study()\fP returns a pointer to a
709    \fBpcre_extra\fP block, in which the \fIstudy_data\fP field points to the
710    results of the study.
711    .P
712    The returned value from \fBpcre_study()\fP can be passed directly to
713    \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other
714  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
715  described below. If studying the pattern does not produce any additional  described
716  information, \fBpcre_study()\fR returns NULL. In that circumstance, if the  .\" HTML <a href="#extradata">
717  calling program wants to pass some of the other fields to \fBpcre_exec()\fR, it  .\" </a>
718  must set up its own \fBpcre_extra\fR block.  below
719    .\"
720  The second argument contains option bits. At present, no options are defined  in the section on matching a pattern.
721  for \fBpcre_study()\fR, and this argument should always be zero.  .P
722    If studying the pattern does not produce any additional information
723  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
724    wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
725    own \fBpcre_extra\fP block.
726    .P
727    The second argument of \fBpcre_study()\fP contains option bits. At present, no
728    options are defined, and this argument should always be zero.
729    .P
730    The third argument for \fBpcre_study()\fP is a pointer for an error message. If
731  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
732  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
733  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
734  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
735    sure that it has run successfully.
736  This is a typical call to \fBpcre_study\fR():  .P
737    This is a typical call to \fBpcre_study\fP():
738    .sp
739    pcre_extra *pe;    pcre_extra *pe;
740    pe = pcre_study(    pe = pcre_study(
741      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
742      0,              /* no options exist */      0,              /* no options exist */
743      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
744    .sp
745  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
746  not have a single fixed starting character. A bitmap of possible starting  not have a single fixed starting character. A bitmap of possible starting
747  characters is created.  bytes is created.
748    .
749    .
750  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
751  .SH LOCALE SUPPORT  .SH "LOCALE SUPPORT"
752  .rs  .rs
753  .sp  .sp
754  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
755  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
756  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
757  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
758  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
759  is NULL, and is sufficient for many applications.  support. The use of locales with Unicode is discouraged. If you are handling
760    characters with codes greater than 128, you should either use UTF-8 and
761  An alternative set of tables can, however, be supplied. Such tables are built  Unicode, or use locales, but not try to mix the two.
762  by calling the \fBpcre_maketables()\fR function, which has no arguments, in the  .P
763  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
764  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.
765  French locale (where accented characters with codes greater than 128 are  Normally, the internal tables recognize only ASCII characters. However, when
766  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
767    default "C" locale of the local system, which may cause them to be different.
768    setlocale(LC_CTYPE, "fr");  .P
769    The internal tables can always be overridden by tables supplied by the
770    application that calls PCRE. These may be created in a different locale from
771    the default. As more and more applications change to using Unicode, the need
772    for this locale support is expected to die away.
773    .P
774    External tables are built by calling the \fBpcre_maketables()\fP function,
775    which has no arguments, in the relevant locale. The result can then be passed
776    to \fBpcre_compile()\fP or \fBpcre_exec()\fP as often as necessary. For
777    example, to build and use tables that are appropriate for the French locale
778    (where accented characters with values greater than 128 are treated as letters),
779    the following code could be used:
780    .sp
781      setlocale(LC_CTYPE, "fr_FR");
782    tables = pcre_maketables();    tables = pcre_maketables();
783    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
784    .sp
785  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
786  pointer that is passed to \fBpcre_compile\fR is saved with the compiled  are using Windows, the name for the French locale is "french".
787  pattern, and the same tables are used via this pointer by \fBpcre_study()\fR  .P
788  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
789  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
790  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
791  memory containing the tables remains available for as long as it is needed.  needed.
792    .P
793  .SH INFORMATION ABOUT A PATTERN  The pointer that is passed to \fBpcre_compile()\fP is saved with the compiled
794    pattern, and the same tables are used via this pointer by \fBpcre_study()\fP
795    and normally also by \fBpcre_exec()\fP. Thus, by default, for any single
796    pattern, compilation, studying and matching all happen in the same locale, but
797    different patterns can be compiled in different locales.
798    .P
799    It is possible to pass a table pointer or NULL (indicating the use of the
800    internal tables) to \fBpcre_exec()\fP. Although not intended for this purpose,
801    this facility could be used to match a pattern in a different locale from the
802    one in which it was compiled. Passing table pointers at run time is discussed
803    below in the section on matching a pattern.
804    .
805    .
806    .SH "INFORMATION ABOUT A PATTERN"
807  .rs  .rs
808  .sp  .sp
809  .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,"
810  .ti +5n  .ti +5n
811  .B int \fIwhat\fR, void *\fIwhere\fR);  .B int \fIwhat\fP, void *\fIwhere\fP);
812  .PP  .PP
813  The \fBpcre_fullinfo()\fR function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
814  pattern. It replaces the obsolete \fBpcre_info()\fR function, which is  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is
815  nevertheless retained for backwards compability (and is documented below).  nevertheless retained for backwards compability (and is documented below).
816    .P
817  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
818  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
819  the pattern was not studied. The third argument specifies which piece of  the pattern was not studied. The third argument specifies which piece of
820  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
821  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
822  the following negative numbers:  the following negative numbers:
823    .sp
824    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
825                          the argument \fIwhere\fR was NULL                          the argument \fIwhere\fP was NULL
826    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
827    PCRE_ERROR_BADOPTION  the value of \fIwhat\fR was invalid    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid
828    .sp
829  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
830  compiled pattern:  check against passing an arbitrary memory pointer. Here is a typical call of
831    \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
832    .sp
833    int rc;    int rc;
834    unsigned long int length;    size_t length;
835    rc = pcre_fullinfo(    rc = pcre_fullinfo(
836      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
837      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
838      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
839      &length);         /* where to put the data */      &length);         /* where to put the data */
840    .sp
841  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
842  as follows:  as follows:
843    .sp
844    PCRE_INFO_BACKREFMAX    PCRE_INFO_BACKREFMAX
845    .sp
846  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
847  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
848  no back references.  no back references.
849    .sp
850    PCRE_INFO_CAPTURECOUNT    PCRE_INFO_CAPTURECOUNT
851    .sp
852  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
853  should point to an \fbint\fR variable.  should point to an \fBint\fP variable.
854    .sp
855      PCRE_INFO_DEFAULT_TABLES
856    .sp
857    Return a pointer to the internal default character tables within PCRE. The
858    fourth argument should point to an \fBunsigned char *\fP variable. This
859    information call is provided for internal use by the \fBpcre_study()\fP
860    function. External callers can cause PCRE to use its internal tables by passing
861    a NULL table pointer.
862    .sp
863    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
864    .sp
865  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
866  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
867  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
868    still recognized for backwards compatibility.)
869  If there is a fixed first byte, e.g. from a pattern such as (cat|cow|coyote),  .P
870  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
871    (cat|cow|coyote), its value is returned. Otherwise, if either
872    .sp
873  (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
874  starts with "^", or  starts with "^", or
875    .sp
876  (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
877  (if it were set, the pattern would be anchored),  (if it were set, the pattern would be anchored),
878    .sp
879  -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
880  subject string or after any newline within the string. Otherwise -2 is  subject string or after any newline within the string. Otherwise -2 is
881  returned. For anchored patterns, -2 is returned.  returned. For anchored patterns, -2 is returned.
882    .sp
883    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
884    .sp
885  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
886  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
887  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
888  fourth argument should point to an \fBunsigned char *\fR variable.  fourth argument should point to an \fBunsigned char *\fP variable.
889    .sp
890      PCRE_INFO_HASCRORLF
891    .sp
892    Return 1 if the pattern contains any explicit matches for CR or LF characters,
893    otherwise 0. The fourth argument should point to an \fBint\fP variable.
894    .sp
895      PCRE_INFO_JCHANGED
896    .sp
897    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
898    fourth argument should point to an \fBint\fP variable. The (?J) internal option
899    setting changes the local PCRE_DUPNAMES option.
900    .sp
901    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
902    .sp
903  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
904  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
905  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
906  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
907  follows something of variable length. For example, for the pattern  follows something of variable length. For example, for the pattern
908  /^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
909  is -1.  is -1.
910    .sp
911    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
912    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
913    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
914    .sp
915  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
916  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
917  acquire a number. A caller that wants to extract data from a named subpattern  acquire numbers. Several convenience functions such as
918  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
919  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
920  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
921  table for the pattern.  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
922    you need to use the name-to-number map, which is described by these three
923    values.
924    .P
925  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
926  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
927  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
928  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
929  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
930  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
931  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
932  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
933    their parentheses numbers. For example, consider the following pattern (assume
934  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
935    .sp
936    (?P<date> (?P<year>(\\d\\d)?\\d\\d) -  .\" JOIN
937    (?P<month>\\d\\d) - (?P<day>\\d\\d) )    (?<date> (?<year>(\ed\ed)?\ed\ed) -
938      (?<month>\ed\ed) - (?<day>\ed\ed) )
939    .sp
940  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
941  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
942  bytes shows in hex, and undefined bytes shown as ??:  bytes shows in hexadecimal, and undefined bytes shown as ??:
943    .sp
944    00 01 d  a  t  e  00 ??    00 01 d  a  t  e  00 ??
945    00 05 d  a  y  00 ?? ??    00 05 d  a  y  00 ?? ??
946    00 04 m  o  n  t  h  00    00 04 m  o  n  t  h  00
947    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
948    .sp
949  When writing code to extract data from named subpatterns, remember that the  When writing code to extract data from named subpatterns using the
950  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
951    different for each compiled pattern.
952    .sp
953      PCRE_INFO_OKPARTIAL
954    .sp
955    Return 1 if the pattern can be used for partial matching, otherwise 0. The
956    fourth argument should point to an \fBint\fP variable. The
957    .\" HREF
958    \fBpcrepartial\fP
959    .\"
960    documentation lists the restrictions that apply to patterns when partial
961    matching is used.
962    .sp
963    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
964    .sp
965  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
966  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
967  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
968  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
969    they are the options that will be in force when matching starts. For example,
970    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
971    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
972    .P
973  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
974  alternatives begin with one of the following:  alternatives begin with one of the following:
975    .sp
976    ^     unless PCRE_MULTILINE is set    ^     unless PCRE_MULTILINE is set
977    \\A    always    \eA    always
978    \\G    always    \eG    always
979    .\" JOIN
980    .*    if PCRE_DOTALL is set and there are no back    .*    if PCRE_DOTALL is set and there are no back
981            references to the subpattern in which .* appears            references to the subpattern in which .* appears
982    .sp
983  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
984  \fBpcre_fullinfo()\fR.  \fBpcre_fullinfo()\fP.
985    .sp
986    PCRE_INFO_SIZE    PCRE_INFO_SIZE
987    .sp
988  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
989  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
990  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
991  variable.  variable.
992    .sp
993    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
994    .sp
995  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
996  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
997  \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
998  created by \fBpcre_study()\fR. The fourth argument should point to a  created by \fBpcre_study()\fP. The fourth argument should point to a
999  \fBsize_t\fR variable.  \fBsize_t\fP variable.
1000    .
1001  .SH OBSOLETE INFO FUNCTION  .
1002    .SH "OBSOLETE INFO FUNCTION"
1003  .rs  .rs
1004  .sp  .sp
1005  .B int pcre_info(const pcre *\fIcode\fR, int *\fIoptptr\fR, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
1006  .B *\fIfirstcharptr\fR);  .B *\fIfirstcharptr\fP);
1007  .PP  .PP
1008  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
1009  restrictive to return all the available data about a compiled pattern. New  restrictive to return all the available data about a compiled pattern. New
1010  programs should use \fBpcre_fullinfo()\fR instead. The yield of  programs should use \fBpcre_fullinfo()\fP instead. The yield of
1011  \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
1012  following negative numbers:  following negative numbers:
1013    .sp
1014    PCRE_ERROR_NULL       the argument \fIcode\fR was NULL    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL
1015    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC   the "magic number" was not found
1016    .sp
1017  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
1018  pattern was compiled is placed in the integer it points to (see  pattern was compiled is placed in the integer it points to (see
1019  PCRE_INFO_OPTIONS above).  PCRE_INFO_OPTIONS above).
1020    .P
1021  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,
1022  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
1023  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1024    .
1025  .SH MATCHING A PATTERN  .
1026    .SH "REFERENCE COUNTS"
1027  .rs  .rs
1028  .sp  .sp
1029  .B int pcre_exec(const pcre *\fIcode\fR, "const pcre_extra *\fIextra\fR,"  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1030    .PP
1031    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1032    data block that contains a compiled pattern. It is provided for the benefit of
1033    applications that operate in an object-oriented manner, where different parts
1034    of the application may be using the same compiled pattern, but you want to free
1035    the block when they are all done.
1036    .P
1037    When a pattern is compiled, the reference count field is initialized to zero.
1038    It is changed only by calling this function, whose action is to add the
1039    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1040    function is the new value. However, the value of the count is constrained to
1041    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1042    it is forced to the appropriate limit value.
1043    .P
1044    Except when it is zero, the reference count is not correctly preserved if a
1045    pattern is compiled on one host and then transferred to a host whose byte-order
1046    is different. (This seems a highly unlikely scenario.)
1047    .
1048    .
1049    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1050    .rs
1051    .sp
1052    .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1053  .ti +5n  .ti +5n
1054  .B "const char *\fIsubject\fR," int \fIlength\fR, int \fIstartoffset\fR,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1055  .ti +5n  .ti +5n
1056  .B int \fIoptions\fR, int *\fIovector\fR, int \fIovecsize\fR);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
1057  .PP  .P
1058  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
1059  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
1060  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
1061  \fIextra\fR argument.  \fIextra\fP argument. This function is the main matching facility of the
1062    library, and it operates in a Perl-like manner. For specialist use there is
1063  Here is an example of a simple call to \fBpcre_exec()\fR:  also an alternative matching function, which is described
1064    .\" HTML <a href="#dfamatch">
1065    .\" </a>
1066    below
1067    .\"
1068    in the section about the \fBpcre_dfa_exec()\fP function.
1069    .P
1070    In most applications, the pattern will have been compiled (and optionally
1071    studied) in the same process that calls \fBpcre_exec()\fP. However, it is
1072    possible to save compiled patterns and study data, and then use them later
1073    in different processes, possibly even on different hosts. For a discussion
1074    about this, see the
1075    .\" HREF
1076    \fBpcreprecompile\fP
1077    .\"
1078    documentation.
1079    .P
1080    Here is an example of a simple call to \fBpcre_exec()\fP:
1081    .sp
1082    int rc;    int rc;
1083    int ovector[30];    int ovector[30];
1084    rc = pcre_exec(    rc = pcre_exec(
# Line 659  Here is an example of a simple call to \ Line 1088  Here is an example of a simple call to \
1088      11,             /* the length of the subject string */      11,             /* the length of the subject string */
1089      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1090      0,              /* default options */      0,              /* default options */
1091      ovector,        /* vector for substring information */      ovector,        /* vector of integers for substring information */
1092      30);            /* number of elements in the vector */      30);            /* number of elements (NOT size in bytes) */
1093    .
1094  If the \fIextra\fR argument is not NULL, it must point to a \fBpcre_extra\fR  .\" HTML <a name="extradata"></a>
1095  data block. The \fBpcre_study()\fR function returns such a block (when it  .SS "Extra data for \fBpcre_exec()\fR"
1096    .rs
1097    .sp
1098    If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1099    data block. The \fBpcre_study()\fP function returns such a block (when it
1100  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
1101  additional information in it. The fields in the block are as follows:  additional information in it. The \fBpcre_extra\fP block contains the following
1102    fields (not necessarily in this order):
1103    unsigned long int \fIflags\fR;  .sp
1104    void *\fIstudy_data\fR;    unsigned long int \fIflags\fP;
1105    unsigned long int \fImatch_limit\fR;    void *\fIstudy_data\fP;
1106    void *\fIcallout_data\fR;    unsigned long int \fImatch_limit\fP;
1107      unsigned long int \fImatch_limit_recursion\fP;
1108  The \fIflags\fR field is a bitmap that specifies which of the other fields    void *\fIcallout_data\fP;
1109      const unsigned char *\fItables\fP;
1110    .sp
1111    The \fIflags\fP field is a bitmap that specifies which of the other fields
1112  are set. The flag bits are:  are set. The flag bits are:
1113    .sp
1114    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1115    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1116      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1117    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1118      PCRE_EXTRA_TABLES
1119  Other flag bits should be set to zero. The \fIstudy_data\fR field is set in the  .sp
1120  \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
1121  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
1122  the block by setting the other fields.  the appropriate flag bit. You should not set this yourself, but you may add to
1123    the block by setting the other fields and their corresponding flag bits.
1124  The \fImatch_limit\fR field provides a means of preventing PCRE from using up a  .P
1125    The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1126  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,
1127  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
1128  classic example is the use of nested unlimited repeats. Internally, PCRE uses a  classic example is the use of nested unlimited repeats.
1129  function called \fBmatch()\fR which it calls repeatedly (sometimes  .P
1130  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
1131  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
1132  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
1133  count starts from zero for each position in the subject string.  limiting the amount of backtracking that can take place. For patterns that are
1134    not anchored, the count restarts from zero for each position in the subject
1135  The default limit for the library can be set when PCRE is built; the default  string.
1136    .P
1137    The default value for the limit can be set when PCRE is built; the default
1138  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
1139  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
1140  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
1141  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
1142  exceeded, \fBpcre_exec()\fR returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1143    .P
1144  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
1145  which is described in the \fBpcrecallout\fR documentation.  instead of limiting the total number of times that \fBmatch()\fP is called, it
1146    limits the depth of recursion. The recursion depth is a smaller number than the
1147  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.
1148  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.
1149  first matching position. However, if a pattern was compiled with PCRE_ANCHORED,  .P
1150  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,
1151  unachored at matching time.  when PCRE has been compiled to use memory on the heap instead of the stack, the
1152    amount of heap memory that can be used.
1153  When PCRE_UTF8 was set at compile time, the validity of the subject as a UTF-8  .P
1154  string is automatically checked. If an invalid UTF-8 sequence of bytes is  The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1155  found, \fBpcre_exec()\fR returns the error PCRE_ERROR_BADUTF8. If you already  built; the default default is the same value as the default for
1156  know that your subject is valid, and you want to skip this check for  \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1157  performance reasons, you can set the PCRE_NO_UTF8_CHECK option when calling  with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1158  \fBpcre_exec()\fR. When this option is set, the effect of passing an invalid  PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1159  UTF-8 string as a subject is undefined. It may cause your program to crash.  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1160    .P
1161  There are also three further options that can be set only at matching time:  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1162    which is described in the
1163    .\" HREF
1164    \fBpcrecallout\fP
1165    .\"
1166    documentation.
1167    .P
1168    The \fItables\fP field is used to pass a character tables pointer to
1169    \fBpcre_exec()\fP; this overrides the value that is stored with the compiled
1170    pattern. A non-NULL value is stored with the compiled pattern only if custom
1171    tables were supplied to \fBpcre_compile()\fP via its \fItableptr\fP argument.
1172    If NULL is passed to \fBpcre_exec()\fP using this mechanism, it forces PCRE's
1173    internal tables to be used. This facility is helpful when re-using patterns
1174    that have been saved after compiling with an external set of tables, because
1175    the external tables might be at a different address when \fBpcre_exec()\fP is
1176    called. See the
1177    .\" HREF
1178    \fBpcreprecompile\fP
1179    .\"
1180    documentation for a discussion of saving compiled patterns for later use.
1181    .
1182    .\" HTML <a name="execoptions"></a>
1183    .SS "Option bits for \fBpcre_exec()\fP"
1184    .rs
1185    .sp
1186    The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1187    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1188    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1189    .sp
1190      PCRE_ANCHORED
1191    .sp
1192    The PCRE_ANCHORED option limits \fBpcre_exec()\fP to matching at the first
1193    matching position. If a pattern was compiled with PCRE_ANCHORED, or turned out
1194    to be anchored by virtue of its contents, it cannot be made unachored at
1195    matching time.
1196    .sp
1197      PCRE_NEWLINE_CR
1198      PCRE_NEWLINE_LF
1199      PCRE_NEWLINE_CRLF
1200      PCRE_NEWLINE_ANYCRLF
1201      PCRE_NEWLINE_ANY
1202    .sp
1203    These options override the newline definition that was chosen or defaulted when
1204    the pattern was compiled. For details, see the description of
1205    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1206    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1207    the way the match position is advanced after a match failure for an unanchored
1208    pattern.
1209    .P
1210    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1211    match attempt for an unanchored pattern fails when the current position is at a
1212    CRLF sequence, and the pattern contains no explicit matches for CR or NL
1213    characters, the match position is advanced by two characters instead of one, in
1214    other words, to after the CRLF.
1215    .P
1216    The above rule is a compromise that makes the most common cases work as
1217    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1218    set), it does not match the string "\er\enA" because, after failing at the
1219    start, it skips both the CR and the LF before retrying. However, the pattern
1220    [\er\en]A does match that string, because it contains an explicit CR or LF
1221    reference, and so advances only by one character after the first failure.
1222    Note than an explicit CR or LF reference occurs for negated character classes
1223    such as [^X] because they can match CR or LF characters.
1224    .P
1225    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1226    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1227    .sp
1228    PCRE_NOTBOL    PCRE_NOTBOL
1229    .sp
1230  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
1231  circumflex metacharacter should not match before it. Setting this without  beginning of a line, so the circumflex metacharacter should not match before
1232  PCRE_MULTILINE (at compile time) causes circumflex never to match.  it. Setting this without PCRE_MULTILINE (at compile time) causes circumflex
1233    never to match. This option affects only the behaviour of the circumflex
1234    metacharacter. It does not affect \eA.
1235    .sp
1236    PCRE_NOTEOL    PCRE_NOTEOL
1237    .sp
1238  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
1239  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
1240  it. Setting this without PCRE_MULTILINE (at compile time) causes dollar never  mode) a newline immediately before it. Setting this without PCRE_MULTILINE (at
1241  to match.  compile time) causes dollar never to match. This option affects only the
1242    behaviour of the dollar metacharacter. It does not affect \eZ or \ez.
1243    .sp
1244    PCRE_NOTEMPTY    PCRE_NOTEMPTY
1245    .sp
1246  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
1247  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
1248  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
1249    .sp
1250    a?b?    a?b?
1251    .sp
1252  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
1253  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
1254  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".
1255    .P
1256  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
1257  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
1258  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
1259  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
1260  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
1261  below) and trying an ordinary match again.  starting offset (see below) and trying an ordinary match again. There is some
1262    code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1263  The subject string is passed to \fBpcre_exec()\fR as a pointer in  .sp
1264  \fIsubject\fR, a length in \fIlength\fR, and a starting offset in    PCRE_NO_UTF8_CHECK
1265  \fIstartoffset\fR. Unlike the pattern string, the subject may contain binary  .sp
1266  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
1267  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.
1268    The value of \fIstartoffset\fP is also checked to ensure that it points to the
1269  If the pattern was compiled with the PCRE_UTF8 option, the subject must be a  start of a UTF-8 character. There is a discussion about the validity of UTF-8
1270  sequence of bytes that is a valid UTF-8 string. If an invalid UTF-8 string is  strings in the
1271  passed, PCRE's behaviour is not defined.  .\" HTML <a href="pcre.html#utf8strings">
1272    .\" </a>
1273    section on UTF-8 support
1274    .\"
1275    in the main
1276    .\" HREF
1277    \fBpcre\fP
1278    .\"
1279    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1280    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1281    PCRE_ERROR_BADUTF8_OFFSET is returned.
1282    .P
1283    If you already know that your subject is valid, and you want to skip these
1284    checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1285    calling \fBpcre_exec()\fP. You might want to do this for the second and
1286    subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1287    all the matches in a single subject string. However, you should be sure that
1288    the value of \fIstartoffset\fP points to the start of a UTF-8 character. When
1289    PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a
1290    subject, or a value of \fIstartoffset\fP that does not point to the start of a
1291    UTF-8 character, is undefined. Your program may crash.
1292    .sp
1293      PCRE_PARTIAL
1294    .sp
1295    This option turns on the partial matching feature. If the subject string fails
1296    to match the pattern, but at some point during the matching process the end of
1297    the subject was reached (that is, the subject partially matches the pattern and
1298    the failure to match occurred only because there were not enough subject
1299    characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of
1300    PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what
1301    may appear in the pattern. These are discussed in the
1302    .\" HREF
1303    \fBpcrepartial\fP
1304    .\"
1305    documentation.
1306    .
1307    .SS "The string to be matched by \fBpcre_exec()\fP"
1308    .rs
1309    .sp
1310    The subject string is passed to \fBpcre_exec()\fP as a pointer in
1311    \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in
1312    \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a
1313    UTF-8 character. Unlike the pattern string, the subject may contain binary zero
1314    bytes. When the starting offset is zero, the search for a match starts at the
1315    beginning of the subject, and this is by far the most common case.
1316    .P
1317  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
1318  same subject by calling \fBpcre_exec()\fR again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
1319  Setting \fIstartoffset\fR differs from just passing over a shortened string and  Setting \fIstartoffset\fP differs from just passing over a shortened string and
1320  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
1321  lookbehind. For example, consider the pattern  lookbehind. For example, consider the pattern
1322    .sp
1323    \\Biss\\B    \eBiss\eB
1324    .sp
1325  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
1326  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
1327  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
1328  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
1329  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
1330  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
1331  \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
1332  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
1333  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.
1334    .P
1335  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
1336  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
1337  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.
1338    .
1339    .SS "How \fBpcre_exec()\fP returns captured substrings"
1340    .rs
1341    .sp
1342  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
1343  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
1344  pattern. Following the usage in Jeffrey Friedl's book, this is called  pattern. Following the usage in Jeffrey Friedl's book, this is called
1345  "capturing" in what follows, and the phrase "capturing subpattern" is used for  "capturing" in what follows, and the phrase "capturing subpattern" is used for
1346  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
1347  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1348    .P
1349  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
1350  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
1351  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:
1352  back captured substrings, each substring using a pair of integers. The  this argument is NOT the size of \fIovector\fP in bytes.
1353  remaining third of the vector is used as workspace by \fBpcre_exec()\fR while  .P
1354  matching capturing subpatterns, and is not available for passing back  The first two-thirds of the vector is used to pass back captured substrings,
1355  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
1356  three. If it is not, it is rounded down.  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1357    and is not available for passing back information. The length passed in
1358  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
1359  returned in pairs of integers, starting at the beginning of \fIovector\fR, and  rounded down.
1360    .P
1361    When a match is successful, information about captured substrings is returned
1362    in pairs of integers, starting at the beginning of \fIovector\fP, and
1363  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
1364  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
1365  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
1366  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
1367  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
1368  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
1369  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
1370  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
1371  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1372    indicating that just the first pair of offsets has been set.
1373  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.  
   
1374  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
1375  string that it matched that gets returned.  string that it matched that is returned.
1376    .P
1377  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
1378  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
1379  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
1380  \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
1381  \fIovecsize\fR as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1382  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
1383  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
1384  to supply an \fIovector\fR.  advisable to supply an \fIovector\fP.
1385    .P
1386  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
1387  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1388  \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
1389  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.
1390    .P
1391  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
1392    the subject when subpattern \fIn\fP has not been used at all. For example, if
1393    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1394    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1395    happens, both values in the offset pairs corresponding to unused subpatterns
1396    are set to -1.
1397    .P
1398    Offset values that correspond to unused subpatterns at the end of the
1399    expression are also set to -1. For example, if the string "abc" is matched
1400    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1401    return from the function is 2, because the highest used capturing subpattern
1402    number is 1. However, you can refer to the offsets for the second and third
1403    capturing subpatterns if you wish (assuming the vector is large enough, of
1404    course).
1405    .P
1406    Some convenience functions are provided for extracting the captured substrings
1407    as separate strings. These are described below.
1408    .
1409    .\" HTML <a name="errorlist"></a>
1410    .SS "Error return values from \fBpcre_exec()\fP"
1411    .rs
1412    .sp
1413    If \fBpcre_exec()\fP fails, it returns a negative number. The following are
1414  defined in the header file:  defined in the header file:
1415    .sp
1416    PCRE_ERROR_NOMATCH        (-1)    PCRE_ERROR_NOMATCH        (-1)
1417    .sp
1418  The subject string did not match the pattern.  The subject string did not match the pattern.
1419    .sp
1420    PCRE_ERROR_NULL           (-2)    PCRE_ERROR_NULL           (-2)
1421    .sp
1422  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
1423  NULL and \fIovecsize\fR was not zero.  NULL and \fIovecsize\fP was not zero.
1424    .sp
1425    PCRE_ERROR_BADOPTION      (-3)    PCRE_ERROR_BADOPTION      (-3)
1426    .sp
1427  An unrecognized bit was set in the \fIoptions\fR argument.  An unrecognized bit was set in the \fIoptions\fP argument.
1428    .sp
1429    PCRE_ERROR_BADMAGIC       (-4)    PCRE_ERROR_BADMAGIC       (-4)
1430    .sp
1431  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
1432  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
1433  magic number isn't present.  compiled in an environment of one endianness is run in an environment with the
1434    other endianness. This is the error that PCRE gives when the magic number is
1435    PCRE_ERROR_UNKNOWN_NODE   (-5)  not present.
1436    .sp
1437      PCRE_ERROR_UNKNOWN_OPCODE (-5)
1438    .sp
1439  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1440  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
1441  of the compiled pattern.  of the compiled pattern.
1442    .sp
1443    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1444    .sp
1445  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
1446  \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
1447  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
1448  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
1449  the end of matching.  automatically freed at the end of matching.
1450    .sp
1451    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1452    .sp
1453  This error is used by the \fBpcre_copy_substring()\fR,  This error is used by the \fBpcre_copy_substring()\fP,
1454  \fBpcre_get_substring()\fR, and \fBpcre_get_substring_list()\fR functions (see  \fBpcre_get_substring()\fP, and \fBpcre_get_substring_list()\fP functions (see
1455  below). It is never returned by \fBpcre_exec()\fR.  below). It is never returned by \fBpcre_exec()\fP.
1456    .sp
1457    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1458    .sp
1459  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
1460  field in a \fBpcre_extra\fR structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1461  description above.  above.
1462    .sp
1463    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1464    .sp
1465  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
1466  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
1467  \fBpcrecallout\fR documentation for details.  .\" HREF
1468    \fBpcrecallout\fP
1469    PCRE_ERROR_BADUTF8       (-10)  .\"
1470    documentation for details.
1471    .sp
1472      PCRE_ERROR_BADUTF8        (-10)
1473    .sp
1474  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.
1475    .sp
1476  .SH EXTRACTING CAPTURED SUBSTRINGS BY NUMBER    PCRE_ERROR_BADUTF8_OFFSET (-11)
1477    .sp
1478    The UTF-8 byte sequence that was passed as a subject was valid, but the value
1479    of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1480    .sp
1481      PCRE_ERROR_PARTIAL        (-12)
1482    .sp
1483    The subject string did not match, but it did match partially. See the
1484    .\" HREF
1485    \fBpcrepartial\fP
1486    .\"
1487    documentation for details of partial matching.
1488    .sp
1489      PCRE_ERROR_BADPARTIAL     (-13)
1490    .sp
1491    The PCRE_PARTIAL option was used with a compiled pattern containing items that
1492    are not supported for partial matching. See the
1493    .\" HREF
1494    \fBpcrepartial\fP
1495    .\"
1496    documentation for details of partial matching.
1497    .sp
1498      PCRE_ERROR_INTERNAL       (-14)
1499    .sp
1500    An unexpected internal error has occurred. This error could be caused by a bug
1501    in PCRE or by overwriting of the compiled pattern.
1502    .sp
1503      PCRE_ERROR_BADCOUNT       (-15)
1504    .sp
1505    This error is given if the value of the \fIovecsize\fP argument is negative.
1506    .sp
1507      PCRE_ERROR_RECURSIONLIMIT (-21)
1508    .sp
1509    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1510    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1511    description above.
1512    .sp
1513      PCRE_ERROR_BADNEWLINE     (-23)
1514    .sp
1515    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1516    .P
1517    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1518    .
1519    .
1520    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
1521  .rs  .rs
1522  .sp  .sp
1523  .B int pcre_copy_substring(const char *\fIsubject\fR, int *\fIovector\fR,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1524  .ti +5n  .ti +5n
1525  .B int \fIstringcount\fR, int \fIstringnumber\fR, char *\fIbuffer\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
1526  .ti +5n  .ti +5n
1527  .B int \fIbuffersize\fR);  .B int \fIbuffersize\fP);
1528  .PP  .PP
1529  .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,  
1530  .ti +5n  .ti +5n
1531  .B int \fIstringcount\fR, int \fIstringnumber\fR,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1532  .ti +5n  .ti +5n
1533  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1534  .PP  .PP
1535  .br  .B int pcre_get_substring_list(const char *\fIsubject\fP,
 .B int pcre_get_substring_list(const char *\fIsubject\fR,  
1536  .ti +5n  .ti +5n
1537  .B int *\fIovector\fR, int \fIstringcount\fR, "const char ***\fIlistptr\fR);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
1538  .PP  .PP
1539  Captured substrings can be accessed directly by using the offsets returned by  Captured substrings can be accessed directly by using the offsets returned by
1540  \fBpcre_exec()\fR in \fIovector\fR. For convenience, the functions  \fBpcre_exec()\fP in \fIovector\fP. For convenience, the functions
1541  \fBpcre_copy_substring()\fR, \fBpcre_get_substring()\fR, and  \fBpcre_copy_substring()\fP, \fBpcre_get_substring()\fP, and
1542  \fBpcre_get_substring_list()\fR are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1543  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1544  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1545  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1546  has a further zero added on the end, but the result is not, of course,  .P
1547  a C string.  A substring that contains a binary zero is correctly extracted and has a
1548    further zero added on the end, but the result is not, of course, a C string.
1549    However, you can process such a string by referring to the length that is
1550    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1551    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1552    for handling strings containing binary zeros, because the end of the final
1553    string is not independently indicated.
1554    .P
1555  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:
1556  \fIsubject\fR is the subject string which has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
1557  \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
1558  \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
1559  captured by the match, including the substring that matched the entire regular  captured by the match, including the substring that matched the entire regular
1560  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
1561  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
1562  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
1563  the vector divided by three.  number of elements in the vector divided by three.
1564    .P
1565  The functions \fBpcre_copy_substring()\fR and \fBpcre_get_substring()\fR  The functions \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP
1566  extract a single substring, whose number is given as \fIstringnumber\fR. A  extract a single substring, whose number is given as \fIstringnumber\fP. A
1567  value of zero extracts the substring that matched the entire pattern, while  value of zero extracts the substring that matched the entire pattern, whereas
1568  higher values extract the captured substrings. For \fBpcre_copy_substring()\fR,  higher values extract the captured substrings. For \fBpcre_copy_substring()\fP,
1569  the string is placed in \fIbuffer\fR, whose length is given by  the string is placed in \fIbuffer\fP, whose length is given by
1570  \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
1571  obtained via \fBpcre_malloc\fR, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1572  \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
1573  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1574    .sp
1575    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1576    .sp
1577  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
1578  memory failed for \fBpcre_get_substring()\fR.  memory failed for \fBpcre_get_substring()\fP.
1579    .sp
1580    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1581    .sp
1582  There is no substring whose number is \fIstringnumber\fR.  There is no substring whose number is \fIstringnumber\fP.
1583    .P
1584  The \fBpcre_get_substring_list()\fR function extracts all available substrings  The \fBpcre_get_substring_list()\fP function extracts all available substrings
1585  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
1586  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
1587  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
1588  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
1589  function is zero if all went well, or  function is zero if all went well, or the error code
1590    .sp
1591    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1592    .sp
1593  if the attempt to get the memory block failed.  if the attempt to get the memory block failed.
1594    .P
1595  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
1596  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
1597  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
1598  string. This can be distinguished from a genuine zero-length substring by  string. This can be distinguished from a genuine zero-length substring by
1599  inspecting the appropriate offset in \fIovector\fR, which is negative for unset  inspecting the appropriate offset in \fIovector\fP, which is negative for unset
1600  substrings.  substrings.
1601    .P
1602  The two convenience functions \fBpcre_free_substring()\fR and  The two convenience functions \fBpcre_free_substring()\fP and
1603  \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
1604  a previous call of \fBpcre_get_substring()\fR or  a previous call of \fBpcre_get_substring()\fP or
1605  \fBpcre_get_substring_list()\fR, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1606  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
1607  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
1608  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1609  \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
1610  provided.  provided.
1611    .
1612  .SH EXTRACTING CAPTURED SUBSTRINGS BY NAME  .
1613    .SH "EXTRACTING CAPTURED SUBSTRINGS BY NAME"
1614  .rs  .rs
1615  .sp  .sp
1616  .B int pcre_copy_named_substring(const pcre *\fIcode\fR,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
1617  .ti +5n  .ti +5n
1618  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIname\fP);
1619    .PP
1620    .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1621  .ti +5n  .ti +5n
1622  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1623  .ti +5n  .ti +5n
1624  .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,  
1625  .ti +5n  .ti +5n
1626  .B const char *\fIname\fR);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1627  .PP  .PP
1628  .br  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
 .B int pcre_get_named_substring(const pcre *\fIcode\fR,  
1629  .ti +5n  .ti +5n
1630  .B const char *\fIsubject\fR, int *\fIovector\fR,  .B const char *\fIsubject\fP, int *\fIovector\fP,
1631  .ti +5n  .ti +5n
1632  .B int \fIstringcount\fR, const char *\fIstringname\fR,  .B int \fIstringcount\fP, const char *\fIstringname\fP,
1633  .ti +5n  .ti +5n
1634  .B const char **\fIstringptr\fR);  .B const char **\fIstringptr\fP);
1635  .PP  .PP
1636  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.
1637  can be done by calling \fBpcre_get_stringnumber()\fR. The first argument is the  For example, for this pattern
1638  compiled pattern, and the second is the name. For example, for this pattern  .sp
1639      (a+)b(?<xxx>\ed+)...
1640    ab(?<xxx>\\d+)...  .sp
1641    the number of the subpattern called "xxx" is 2. If the name is known to be
1642  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
1643  extract the substring directly, or use one of the functions described in the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
1644  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
1645  whole job.  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
1646    that name.
1647  Most of the arguments of \fIpcre_copy_named_substring()\fR and  .P
1648  \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
1649  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
1650  differences.  two functions that do the whole job.
1651    .P
1652    Most of the arguments of \fBpcre_copy_named_substring()\fP and
1653    \fBpcre_get_named_substring()\fP are the same as those for the similarly named
1654    functions that extract by number. As these are described in the previous
1655    section, they are not re-described here. There are just two differences:
1656    .P
1657  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
1658  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
1659  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
1660  translation table.  translation table.
1661    .P
1662  These functions call \fBpcre_get_stringnumber()\fR, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1663  then call \fIpcre_copy_substring()\fR or \fIpcre_get_substring()\fR, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1664  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1665    the behaviour may not be what you want (see the next section).
1666  .in 0  .
1667  Last updated: 20 August 2003  .
1668  .br  .SH "DUPLICATE SUBPATTERN NAMES"
1669  Copyright (c) 1997-2003 University of Cambridge.  .rs
1670    .sp
1671    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1672    .ti +5n
1673    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1674    .PP
1675    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1676    are not required to be unique. Normally, patterns with duplicate names are such
1677    that in any one match, only one of the named subpatterns participates. An
1678    example is shown in the
1679    .\" HREF
1680    \fBpcrepattern\fP
1681    .\"
1682    documentation.
1683    .P
1684    When duplicates are present, \fBpcre_copy_named_substring()\fP and
1685    \fBpcre_get_named_substring()\fP return the first substring corresponding to
1686    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1687    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1688    returns one of the numbers that are associated with the name, but it is not
1689    defined which it is.
1690    .P
1691    If you want to get full details of all captured substrings for a given name,
1692    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1693    argument is the compiled pattern, and the second is the name. The third and
1694    fourth are pointers to variables which are updated by the function. After it
1695    has run, they point to the first and last entries in the name-to-number table
1696    for the given name. The function itself returns the length of each entry, or
1697    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1698    described above in the section entitled \fIInformation about a pattern\fP.
1699    Given all the relevant entries for the name, you can extract each of their
1700    numbers, and hence the captured data, if any.
1701    .
1702    .
1703    .SH "FINDING ALL POSSIBLE MATCHES"
1704    .rs
1705    .sp
1706    The traditional matching function uses a similar algorithm to Perl, which stops
1707    when it finds the first match, starting at a given point in the subject. If you
1708    want to find all possible matches, or the longest possible match, consider
1709    using the alternative matching function (see below) instead. If you cannot use
1710    the alternative function, but still need to find all possible matches, you
1711    can kludge it up by making use of the callout facility, which is described in
1712    the
1713    .\" HREF
1714    \fBpcrecallout\fP
1715    .\"
1716    documentation.
1717    .P
1718    What you have to do is to insert a callout right at the end of the pattern.
1719    When your callout function is called, extract and save the current matched
1720    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1721    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1722    will yield PCRE_ERROR_NOMATCH.
1723    .
1724    .
1725    .\" HTML <a name="dfamatch"></a>
1726    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1727    .rs
1728    .sp
1729    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1730    .ti +5n
1731    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1732    .ti +5n
1733    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1734    .ti +5n
1735    .B int *\fIworkspace\fP, int \fIwscount\fP);
1736    .P
1737    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1738    a compiled pattern, using a matching algorithm that scans the subject string
1739    just once, and does not backtrack. This has different characteristics to the
1740    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1741    patterns are not supported. Nevertheless, there are times when this kind of
1742    matching can be useful. For a discussion of the two matching algorithms, see
1743    the
1744    .\" HREF
1745    \fBpcrematching\fP
1746    .\"
1747    documentation.
1748    .P
1749    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1750    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1751    different way, and this is described below. The other common arguments are used
1752    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1753    here.
1754    .P
1755    The two additional arguments provide workspace for the function. The workspace
1756    vector should contain at least 20 elements. It is used for keeping track of
1757    multiple paths through the pattern tree. More workspace will be needed for
1758    patterns and subjects where there are a lot of potential matches.
1759    .P
1760    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1761    .sp
1762      int rc;
1763      int ovector[10];
1764      int wspace[20];
1765      rc = pcre_dfa_exec(
1766        re,             /* result of pcre_compile() */
1767        NULL,           /* we didn't study the pattern */
1768        "some string",  /* the subject string */
1769        11,             /* the length of the subject string */
1770        0,              /* start at offset 0 in the subject */
1771        0,              /* default options */
1772        ovector,        /* vector of integers for substring information */
1773        10,             /* number of elements (NOT size in bytes) */
1774        wspace,         /* working space vector */
1775        20);            /* number of elements (NOT size in bytes) */
1776    .
1777    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1778    .rs
1779    .sp
1780    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1781    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1782    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1783    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1784    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1785    .sp
1786      PCRE_PARTIAL
1787    .sp
1788    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1789    details are slightly different. When PCRE_PARTIAL is set for
1790    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1791    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1792    complete matches, but there is still at least one matching possibility. The
1793    portion of the string that provided the partial match is set as the first
1794    matching string.
1795    .sp
1796      PCRE_DFA_SHORTEST
1797    .sp
1798    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1799    soon as it has found one match. Because of the way the alternative algorithm
1800    works, this is necessarily the shortest possible match at the first possible
1801    matching point in the subject string.
1802    .sp
1803      PCRE_DFA_RESTART
1804    .sp
1805    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1806    a partial match, it is possible to call it again, with additional subject
1807    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1808    option requests this action; when it is set, the \fIworkspace\fP and
1809    \fIwscount\fP options must reference the same vector as before because data
1810    about the match so far is left in them after a partial match. There is more
1811    discussion of this facility in the
1812    .\" HREF
1813    \fBpcrepartial\fP
1814    .\"
1815    documentation.
1816    .
1817    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1818    .rs
1819    .sp
1820    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1821    substring in the subject. Note, however, that all the matches from one run of
1822    the function start at the same point in the subject. The shorter matches are
1823    all initial substrings of the longer matches. For example, if the pattern
1824    .sp
1825      <.*>
1826    .sp
1827    is matched against the string
1828    .sp
1829      This is <something> <something else> <something further> no more
1830    .sp
1831    the three matched strings are
1832    .sp
1833      <something>
1834      <something> <something else>
1835      <something> <something else> <something further>
1836    .sp
1837    On success, the yield of the function is a number greater than zero, which is
1838    the number of matched substrings. The substrings themselves are returned in
1839    \fIovector\fP. Each string uses two elements; the first is the offset to the
1840    start, and the second is the offset to the end. In fact, all the strings have
1841    the same start offset. (Space could have been saved by giving this only once,
1842    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1843    returns data, even though the meaning of the strings is different.)
1844    .P
1845    The strings are returned in reverse order of length; that is, the longest
1846    matching string is given first. If there were too many matches to fit into
1847    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1848    the longest matches.
1849    .
1850    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1851    .rs
1852    .sp
1853    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1854    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1855    described
1856    .\" HTML <a href="#errorlist">
1857    .\" </a>
1858    above.
1859    .\"
1860    There are in addition the following errors that are specific to
1861    \fBpcre_dfa_exec()\fP:
1862    .sp
1863      PCRE_ERROR_DFA_UITEM      (-16)
1864    .sp
1865    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1866    that it does not support, for instance, the use of \eC or a back reference.
1867    .sp
1868      PCRE_ERROR_DFA_UCOND      (-17)
1869    .sp
1870    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1871    uses a back reference for the condition, or a test for recursion in a specific
1872    group. These are not supported.
1873    .sp
1874      PCRE_ERROR_DFA_UMLIMIT    (-18)
1875    .sp
1876    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1877    block that contains a setting of the \fImatch_limit\fP field. This is not
1878    supported (it is meaningless).
1879    .sp
1880      PCRE_ERROR_DFA_WSSIZE     (-19)
1881    .sp
1882    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1883    \fIworkspace\fP vector.
1884    .sp
1885      PCRE_ERROR_DFA_RECURSE    (-20)
1886    .sp
1887    When a recursive subpattern is processed, the matching function calls itself
1888    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1889    error is given if the output vector is not large enough. This should be
1890    extremely rare, as a vector of size 1000 is used.
1891    .
1892    .
1893    .SH "SEE ALSO"
1894    .rs
1895    .sp
1896    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1897    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1898    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1899    .
1900    .
1901    .SH AUTHOR
1902    .rs
1903    .sp
1904    .nf
1905    Philip Hazel
1906    University Computing Service
1907    Cambridge CB2 3QH, England.
1908    .fi
1909    .
1910    .
1911    .SH REVISION
1912    .rs
1913    .sp
1914    .nf
1915    Last updated: 21 August 2007
1916    Copyright (c) 1997-2007 University of Cambridge.
1917    .fi

Legend:
Removed from v.71  
changed lines
  Added in v.226

  ViewVC Help
Powered by ViewVC 1.1.5