/[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 87 by nigel, Sat Feb 24 21:41:21 2007 UTC revision 683 by ph10, Tue Sep 6 10:37:15 2011 UTC
# Line 1  Line 1 
1  .TH PCREAPI 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH "PCRE NATIVE API"  .SH "PCRE NATIVE API BASIC FUNCTIONS"
5  .rs  .rs
6  .sp  .sp
7  .B #include <pcre.h>  .B #include <pcre.h>
8  .PP  .PP
9  .SM  .SM
 .br  
10  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
11  .ti +5n  .ti +5n
12  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13  .ti +5n  .ti +5n
14  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
15  .PP  .PP
 .br  
16  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17  .ti +5n  .ti +5n
18  .B int *\fIerrorcodeptr\fP,  .B int *\fIerrorcodeptr\fP,
# Line 23  PCRE - Perl-compatible regular expressio Line 21  PCRE - Perl-compatible regular expressio
21  .ti +5n  .ti +5n
22  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
23  .PP  .PP
 .br  
24  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25  .ti +5n  .ti +5n
26  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
27  .PP  .PP
28  .br  .B void pcre_free_study(pcre_extra *\fIextra\fP);
29    .PP
30  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
31  .ti +5n  .ti +5n
32  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
33  .ti +5n  .ti +5n
34  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
35    .
36    .
37    .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
38    .rs
39    .sp
40    .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
41    .PP
42    .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
43    .PP
44    .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
45    .ti +5n
46    .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
47  .PP  .PP
 .br  
48  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
49  .ti +5n  .ti +5n
50  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
# Line 44  PCRE - Perl-compatible regular expressio Line 53  PCRE - Perl-compatible regular expressio
53  .ti +5n  .ti +5n
54  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
55  .PP  .PP
 .br  
56  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
57  .ti +5n  .ti +5n
58  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 53  PCRE - Perl-compatible regular expressio Line 61  PCRE - Perl-compatible regular expressio
61  .ti +5n  .ti +5n
62  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
63  .PP  .PP
 .br  
64  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
65  .ti +5n  .ti +5n
66  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
67  .ti +5n  .ti +5n
68  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
69  .PP  .PP
 .br  
70  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
71  .ti +5n  .ti +5n
72  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 69  PCRE - Perl-compatible regular expressio Line 75  PCRE - Perl-compatible regular expressio
75  .ti +5n  .ti +5n
76  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
77  .PP  .PP
 .br  
78  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
79  .ti +5n  .ti +5n
80  .B const char *\fIname\fP);  .B const char *\fIname\fP);
81  .PP  .PP
82  .br  .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
83    .ti +5n
84    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
85    .PP
86  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
87  .ti +5n  .ti +5n
88  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
89  .ti +5n  .ti +5n
90  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
91  .PP  .PP
 .br  
92  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
93  .ti +5n  .ti +5n
94  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
95  .PP  .PP
 .br  
96  .B void pcre_free_substring(const char *\fIstringptr\fP);  .B void pcre_free_substring(const char *\fIstringptr\fP);
97  .PP  .PP
 .br  
98  .B void pcre_free_substring_list(const char **\fIstringptr\fP);  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
99  .PP  .PP
 .br  
100  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
101  .PP  .PP
 .br  
102  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
103  .ti +5n  .ti +5n
104  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
105  .PP  .PP
 .br  
106  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
107  .B *\fIfirstcharptr\fP);  .B *\fIfirstcharptr\fP);
108  .PP  .PP
 .br  
109  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
110  .PP  .PP
 .br  
111  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
112  .PP  .PP
 .br  
113  .B char *pcre_version(void);  .B char *pcre_version(void);
114  .PP  .PP
 .br  
115  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
116  .PP  .PP
 .br  
117  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
118  .PP  .PP
 .br  
119  .B void *(*pcre_stack_malloc)(size_t);  .B void *(*pcre_stack_malloc)(size_t);
120  .PP  .PP
 .br  
121  .B void (*pcre_stack_free)(void *);  .B void (*pcre_stack_free)(void *);
122  .PP  .PP
 .br  
123  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
124  .  .
125  .  .
126  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
127  .rs  .rs
128  .sp  .sp
129  PCRE has its own native API, which is described in this document. There is  PCRE has its own native API, which is described in this document. There are
130  also a set of wrapper functions that correspond to the POSIX regular expression  also some wrapper functions that correspond to the POSIX regular expression
131  API. These are described in the  API, but they do not give access to all the functionality. They are described
132    in the
133  .\" HREF  .\" HREF
134  \fBpcreposix\fP  \fBpcreposix\fP
135  .\"  .\"
136  documentation. Both of these APIs define a set of C function calls. A C++  documentation. Both of these APIs define a set of C function calls. A C++
137  wrapper is distributed with PCRE. It is documented in the  wrapper is also distributed with PCRE. It is documented in the
138  .\" HREF  .\" HREF
139  \fBpcrecpp\fP  \fBpcrecpp\fP
140  .\"  .\"
# Line 152  an application that uses PCRE. The heade Line 147  an application that uses PCRE. The heade
147  and PCRE_MINOR to contain the major and minor release numbers for the library.  and PCRE_MINOR to contain the major and minor release numbers for the library.
148  Applications can use these to include support for different releases of PCRE.  Applications can use these to include support for different releases of PCRE.
149  .P  .P
150    In a Windows environment, if you want to statically link an application program
151    against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
152    including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
153    \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
154    \fB__declspec(dllimport)\fP, with unwanted results.
155    .P
156  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
157  and \fBpcre_exec()\fP are used for compiling and matching regular expressions  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
158  in a Perl-compatible manner. A sample program that demonstrates the simplest  in a Perl-compatible manner. A sample program that demonstrates the simplest
159  way of using them is provided in the file called \fIpcredemo.c\fP in the source  way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
160  distribution. The  source distribution. A listing of this program is given in the
161    .\" HREF
162    \fBpcredemo\fP
163    .\"
164    documentation, and the
165  .\" HREF  .\" HREF
166  \fBpcresample\fP  \fBpcresample\fP
167  .\"  .\"
168  documentation describes how to run it.  documentation describes how to compile and run it.
169    .P
170    Just-in-time compiler support is an optional feature of PCRE that can be built
171    in appropriate hardware environments. It greatly speeds up the matching
172    performance of many patterns. Simple programs can request its use if available.
173    More complicated programs might need to make use of the
174    \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
175    \fBpcre_assign_jit_stack()\fP functions in order to control its memory usage.
176    These functions are discussed in the
177    .\" HREF
178    \fBpcrejit\fP
179    .\"
180    documentation.
181  .P  .P
182  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
183  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
184  matching. This allows it to find all possible matches (at a given point in the  matching. The alternative algorithm finds all possible matches (at a given
185  subject), not just one. However, this algorithm does not return captured  point in the subject), and scans the subject just once (unless there are
186    lookbehind assertions). However, this algorithm does not return captured
187  substrings. A description of the two matching algorithms and their advantages  substrings. A description of the two matching algorithms and their advantages
188  and disadvantages is given in the  and disadvantages is given in the
189  .\" HREF  .\" HREF
# Line 183  matched by \fBpcre_exec()\fP. They are: Line 201  matched by \fBpcre_exec()\fP. They are:
201    \fBpcre_get_named_substring()\fP    \fBpcre_get_named_substring()\fP
202    \fBpcre_get_substring_list()\fP    \fBpcre_get_substring_list()\fP
203    \fBpcre_get_stringnumber()\fP    \fBpcre_get_stringnumber()\fP
204      \fBpcre_get_stringtable_entries()\fP
205  .sp  .sp
206  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
207  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
# Line 212  should be done before calling any PCRE f Line 231  should be done before calling any PCRE f
231  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
232  indirections to memory management functions. These special functions are used  indirections to memory management functions. These special functions are used
233  only when PCRE is compiled to use the heap for remembering data, instead of  only when PCRE is compiled to use the heap for remembering data, instead of
234  recursive function calls, when running the \fBpcre_exec()\fP function. This is  recursive function calls, when running the \fBpcre_exec()\fP function. See the
235  a non-standard way of building PCRE, for use in environments that have limited  .\" HREF
236  stacks. Because of the greater use of memory management, it runs more slowly.  \fBpcrebuild\fP
237  Separate functions are provided so that special-purpose external code can be  .\"
238  used for this case. When used, these functions are always called in a  documentation for details of how to do this. It is a non-standard way of
239  stack-like manner (last obtained, first freed), and always for memory blocks of  building PCRE, for use in environments that have limited stacks. Because of the
240  the same size.  greater use of memory management, it runs more slowly. Separate functions are
241    provided so that special-purpose external code can be used for this case. When
242    used, these functions are always called in a stack-like manner (last obtained,
243    first freed), and always for memory blocks of the same size. There is a
244    discussion about PCRE's stack usage in the
245    .\" HREF
246    \fBpcrestack\fP
247    .\"
248    documentation.
249  .P  .P
250  The global variable \fBpcre_callout\fP initially contains NULL. It can be set  The global variable \fBpcre_callout\fP initially contains NULL. It can be set
251  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
# Line 229  points during a matching operation. Deta Line 256  points during a matching operation. Deta
256  documentation.  documentation.
257  .  .
258  .  .
259    .\" HTML <a name="newlines"></a>
260    .SH NEWLINES
261    .rs
262    .sp
263    PCRE supports five different conventions for indicating line breaks in
264    strings: a single CR (carriage return) character, a single LF (linefeed)
265    character, the two-character sequence CRLF, any of the three preceding, or any
266    Unicode newline sequence. The Unicode newline sequences are the three just
267    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
268    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
269    (paragraph separator, U+2029).
270    .P
271    Each of the first three conventions is used by at least one operating system as
272    its standard newline sequence. When PCRE is built, a default can be specified.
273    The default default is LF, which is the Unix standard. When PCRE is run, the
274    default can be overridden, either when a pattern is compiled, or when it is
275    matched.
276    .P
277    At compile time, the newline convention can be specified by the \fIoptions\fP
278    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
279    start of the pattern itself; this overrides any other settings. See the
280    .\" HREF
281    \fBpcrepattern\fP
282    .\"
283    page for details of the special character sequences.
284    .P
285    In the PCRE documentation the word "newline" is used to mean "the character or
286    pair of characters that indicate a line break". The choice of newline
287    convention affects the handling of the dot, circumflex, and dollar
288    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
289    recognized line ending sequence, the match position advancement for a
290    non-anchored pattern. There is more detail about this in the
291    .\" HTML <a href="#execoptions">
292    .\" </a>
293    section on \fBpcre_exec()\fP options
294    .\"
295    below.
296    .P
297    The choice of newline convention does not affect the interpretation of
298    the \en or \er escape sequences, nor does it affect what \eR matches, which is
299    controlled in a similar way, but by separate options.
300    .
301    .
302  .SH MULTITHREADING  .SH MULTITHREADING
303  .rs  .rs
304  .sp  .sp
# Line 239  callout function pointed to by \fBpcre_c Line 309  callout function pointed to by \fBpcre_c
309  .P  .P
310  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
311  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.
312    .P
313    If the just-in-time optimization feature is being used, it needs separate
314    memory stack areas for each thread. See the
315    .\" HREF
316    \fBpcrejit\fP
317    .\"
318    documentation for more details.
319  .  .
320  .  .
321  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
# Line 250  which it was compiled. Details are given Line 327  which it was compiled. Details are given
327  .\" HREF  .\" HREF
328  \fBpcreprecompile\fP  \fBpcreprecompile\fP
329  .\"  .\"
330  documentation.  documentation. However, compiling a regular expression with one version of PCRE
331    for use with a different version is not guaranteed to work and may cause
332    crashes.
333  .  .
334  .  .
335  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 279  otherwise it is set to zero. Line 358  otherwise it is set to zero.
358  The output is an integer that is set to one if support for Unicode character  The output is an integer that is set to one if support for Unicode character
359  properties is available; otherwise it is set to zero.  properties is available; otherwise it is set to zero.
360  .sp  .sp
361      PCRE_CONFIG_JIT
362    .sp
363    The output is an integer that is set to one if support for just-in-time
364    compiling is available; otherwise it is set to zero.
365    .sp
366    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
367  .sp  .sp
368  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
369  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
370  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.
371    Though they are derived from ASCII, the same values are returned in EBCDIC
372    environments. The default should normally correspond to the standard sequence
373    for your operating system.
374    .sp
375      PCRE_CONFIG_BSR
376    .sp
377    The output is an integer whose value indicates what character sequences the \eR
378    escape sequence matches by default. A value of 0 means that \eR matches any
379    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
380    or CRLF. The default can be overridden when a pattern is compiled or matched.
381  .sp  .sp
382    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
383  .sp  .sp
# Line 305  documentation. Line 399  documentation.
399  .sp  .sp
400    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
401  .sp  .sp
402  The output is an integer that gives the default limit for the number of  The output is a long integer that gives the default limit for the number of
403  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
404  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
405  .sp  .sp
406    PCRE_CONFIG_MATCH_LIMIT_RECURSION    PCRE_CONFIG_MATCH_LIMIT_RECURSION
407  .sp  .sp
408  The output is an integer that gives the default limit for the depth of  The output is a long integer that gives the default limit for the depth of
409  recursion when calling the internal matching function in a \fBpcre_exec()\fP  recursion when calling the internal matching function in a \fBpcre_exec()\fP
410  execution. Further details are given with \fBpcre_exec()\fP below.  execution. Further details are given with \fBpcre_exec()\fP below.
411  .sp  .sp
# Line 346  avoiding the use of the stack. Line 440  avoiding the use of the stack.
440  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
441  called to compile a pattern into an internal form. The only difference between  called to compile a pattern into an internal form. The only difference between
442  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
443  \fIerrorcodeptr\fP, via which a numerical error code can be returned.  \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
444    too much repetition, we refer just to \fBpcre_compile()\fP below, but the
445    information applies equally to \fBpcre_compile2()\fP.
446  .P  .P
447  The pattern is a C string terminated by a binary zero, and is passed in the  The pattern is a C string terminated by a binary zero, and is passed in the
448  \fIpattern\fP argument. A pointer to a single block of memory that is obtained  \fIpattern\fP argument. A pointer to a single block of memory that is obtained
449  via \fBpcre_malloc\fP is returned. This contains the compiled code and related  via \fBpcre_malloc\fP is returned. This contains the compiled code and related
450  data. The \fBpcre\fP type is defined for the returned block; this is a typedef  data. The \fBpcre\fP type is defined for the returned block; this is a typedef
451  for a structure whose contents are not externally defined. It is up to the  for a structure whose contents are not externally defined. It is up to the
452  caller to free the memory when it is no longer required.  caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
453  .P  .P
454  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
455  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
456  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
457  argument, which is an address (see below).  argument, which is an address (see below).
458  .P  .P
459  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
460  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
461  options are described below. Some of them, in particular, those that are  options are described below. Some of them (in particular, those that are
462  compatible with Perl, can also be set and unset from within the pattern (see  compatible with Perl, but some others as well) can also be set and unset from
463  the detailed description in the  within the pattern (see the detailed description in the
464  .\" HREF  .\" HREF
465  \fBpcrepattern\fP  \fBpcrepattern\fP
466  .\"  .\"
467  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For those options that can be different in different parts of
468  specifies their initial settings at the start of compilation and execution. The  the pattern, the contents of the \fIoptions\fP argument specifies their
469  PCRE_ANCHORED option can be set at the time of matching as well as at compile  settings at the start of compilation and execution. The PCRE_ANCHORED,
470  time.  PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
471    PCRE_NO_START_OPT options can be set at the time of matching as well as at
472    compile time.
473  .P  .P
474  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
475  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
476  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual  NULL, and sets the variable pointed to by \fIerrptr\fP to point to a textual
477  error message. This is a static string that is part of the library. You must  error message. This is a static string that is part of the library. You must
478  not try to free it. The offset from the start of the pattern to the character  not try to free it. Normally, the offset from the start of the pattern to the
479  where the error was discovered is placed in the variable pointed to by  byte that was being processed when the error was discovered is placed in the
480  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.  variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
481    immediate error is given). However, for an invalid UTF-8 string, the offset is
482    that of the first byte of the failing character. Also, some errors are not
483    detected until checks are carried out when the whole pattern has been scanned;
484    in these cases the offset passed back is the length of the pattern.
485    .P
486    Note that the offset is in bytes, not characters, even in UTF-8 mode. It may
487    sometimes point into the middle of a UTF-8 character.
488  .P  .P
489  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
490  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
# Line 426  facility, see the Line 531  facility, see the
531  .\"  .\"
532  documentation.  documentation.
533  .sp  .sp
534      PCRE_BSR_ANYCRLF
535      PCRE_BSR_UNICODE
536    .sp
537    These options (which are mutually exclusive) control what the \eR escape
538    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
539    match any Unicode newline sequence. The default is specified when PCRE is
540    built. It can be overridden from within the pattern, or by setting an option
541    when a compiled pattern is matched.
542    .sp
543    PCRE_CASELESS    PCRE_CASELESS
544  .sp  .sp
545  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
# Line 442  with UTF-8 support. Line 556  with UTF-8 support.
556  .sp  .sp
557  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
558  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
559  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
560  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.
561  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
562  a pattern.  pattern.
563  .sp  .sp
564    PCRE_DOTALL    PCRE_DOTALL
565  .sp  .sp
566  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharacter in the pattern matches a character of
567  including newlines. Without it, newlines are excluded. This option is  any value, including one that indicates a newline. However, it only ever
568    matches one character, even if newlines are coded as CRLF. Without this option,
569    a dot does not match when the current position is at a newline. This option is
570  equivalent to Perl's /s option, and it can be changed within a pattern by a  equivalent to Perl's /s option, and it can be changed within a pattern by a
571  (?s) option setting. A negative class such as [^a] always matches a newline  (?s) option setting. A negative class such as [^a] always matches newline
572  character, independent of the setting of this option.  characters, independent of the setting of this option.
573    .sp
574      PCRE_DUPNAMES
575    .sp
576    If this bit is set, names used to identify capturing subpatterns need not be
577    unique. This can be helpful for certain types of pattern when it is known that
578    only one instance of the named subpattern can ever be matched. There are more
579    details of named subpatterns below; see also the
580    .\" HREF
581    \fBpcrepattern\fP
582    .\"
583    documentation.
584  .sp  .sp
585    PCRE_EXTENDED    PCRE_EXTENDED
586  .sp  .sp
587  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
588  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
589  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
590  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
591  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
592  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
593    .P
594    Which characters are interpreted as newlines is controlled by the options
595    passed to \fBpcre_compile()\fP or by a special sequence at the start of the
596    pattern, as described in the section entitled
597    .\" HTML <a href="pcrepattern.html#newlines">
598    .\" </a>
599    "Newline conventions"
600    .\"
601    in the \fBpcrepattern\fP documentation. Note that the end of this type of
602    comment is a literal newline sequence in the pattern; escape sequences that
603    happen to represent a newline do not count.
604  .P  .P
605  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
606  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
607  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
608  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( that introduces a conditional subpattern.
609  .sp  .sp
610    PCRE_EXTRA    PCRE_EXTRA
611  .sp  .sp
# Line 476  that is incompatible with Perl, but it i Line 614  that is incompatible with Perl, but it i
614  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
615  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
616  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
617  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
618  controlled by this option. It can also be set by a (?X) option setting within a  give an error for this, by running it with the -w option.) There are at present
619  pattern.  no other features controlled by this option. It can also be set by a (?X)
620    option setting within a pattern.
621  .sp  .sp
622    PCRE_FIRSTLINE    PCRE_FIRSTLINE
623  .sp  .sp
624  If this option is set, an unanchored pattern is required to match before or at  If this option is set, an unanchored pattern is required to match before or at
625  the first newline character in the subject string, though the matched text may  the first newline in the subject string, though the matched text may continue
626  continue over the newline.  over the newline.
627    .sp
628      PCRE_JAVASCRIPT_COMPAT
629    .sp
630    If this option is set, PCRE's behaviour is changed in some ways so that it is
631    compatible with JavaScript rather than Perl. The changes are as follows:
632    .P
633    (1) A lone closing square bracket in a pattern causes a compile-time error,
634    because this is illegal in JavaScript (by default it is treated as a data
635    character). Thus, the pattern AB]CD becomes illegal when this option is set.
636    .P
637    (2) At run time, a back reference to an unset subpattern group matches an empty
638    string (by default this causes the current matching alternative to fail). A
639    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
640    an "a" in the subject), whereas it fails by default, for Perl compatibility.
641  .sp  .sp
642    PCRE_MULTILINE    PCRE_MULTILINE
643  .sp  .sp
# Line 496  terminating newline (unless PCRE_DOLLAR_ Line 649  terminating newline (unless PCRE_DOLLAR_
649  Perl.  Perl.
650  .P  .P
651  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
652  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
653  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
654  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
655  setting. If there are no "\en" characters in a subject string, or no  (?m) option setting. If there are no newlines in a subject string, or no
656  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
657  .sp  .sp
658      PCRE_NEWLINE_CR
659      PCRE_NEWLINE_LF
660      PCRE_NEWLINE_CRLF
661      PCRE_NEWLINE_ANYCRLF
662      PCRE_NEWLINE_ANY
663    .sp
664    These options override the default newline definition that was chosen when PCRE
665    was built. Setting the first or the second specifies that a newline is
666    indicated by a single character (CR or LF, respectively). Setting
667    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
668    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
669    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
670    that any Unicode newline sequence should be recognized. The Unicode newline
671    sequences are the three just mentioned, plus the single characters VT (vertical
672    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
673    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
674    recognized only in UTF-8 mode.
675    .P
676    The newline setting in the options word uses three bits that are treated
677    as a number, giving eight possibilities. Currently only six are used (default
678    plus the five values above). This means that if you set more than one newline
679    option, the combination may or may not be sensible. For example,
680    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
681    other combinations may yield unused numbers and cause an error.
682    .P
683    The only time that a line break in a pattern is specially recognized when
684    compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
685    and so are ignored in this mode. Also, an unescaped # outside a character class
686    indicates a comment that lasts until after the next line break sequence. In
687    other circumstances, line break sequences in patterns are treated as literal
688    data.
689    .P
690    The newline option that is set at compile time becomes the default that is used
691    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
692    .sp
693    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
694  .sp  .sp
695  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
# Line 510  were followed by ?: but named parenthese Line 698  were followed by ?: but named parenthese
698  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
699  in Perl.  in Perl.
700  .sp  .sp
701      NO_START_OPTIMIZE
702    .sp
703    This is an option that acts at matching time; that is, it is really an option
704    for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
705    it is remembered with the compiled pattern and assumed at matching time. For
706    details see the discussion of PCRE_NO_START_OPTIMIZE
707    .\" HTML <a href="#execoptions">
708    .\" </a>
709    below.
710    .\"
711    .sp
712      PCRE_UCP
713    .sp
714    This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
715    \ew, and some of the POSIX character classes. By default, only ASCII characters
716    are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
717    classify characters. More details are given in the section on
718    .\" HTML <a href="pcre.html#genericchartypes">
719    .\" </a>
720    generic character types
721    .\"
722    in the
723    .\" HREF
724    \fBpcrepattern\fP
725    .\"
726    page. If you set PCRE_UCP, matching one of the items it affects takes much
727    longer. The option is available only if PCRE has been compiled with Unicode
728    property support.
729    .sp
730    PCRE_UNGREEDY    PCRE_UNGREEDY
731  .sp  .sp
732  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
# Line 523  of UTF-8 characters instead of single-by Line 740  of UTF-8 characters instead of single-by
740  available only when PCRE is 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
741  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
742  behaviour of PCRE are given in the  behaviour of PCRE are given in the
 .\" HTML <a href="pcre.html#utf8support">  
 .\" </a>  
 section on UTF-8 support  
 .\"  
 in the main  
743  .\" HREF  .\" HREF
744  \fBpcre\fP  \fBpcreunicode\fP
745  .\"  .\"
746  page.  page.
747  .sp  .sp
748    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
749  .sp  .sp
750  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
751  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
752  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcre.html#utf8strings">
753  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
754  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
755  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
756  Note that this option can also be passed to \fBpcre_exec()\fP and  in the main
757  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject  .\" HREF
758  strings.  \fBpcre\fP
759    .\"
760    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
761    returns an error. If you already know that your pattern is valid, and you want
762    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
763    option. When it is set, the effect of passing an invalid UTF-8 string as a
764    pattern is undefined. It may cause your program to crash. Note that this option
765    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
766    the UTF-8 validity checking of subject strings.
767  .  .
768  .  .
769  .SH "COMPILATION ERROR CODES"  .SH "COMPILATION ERROR CODES"
# Line 551  strings. Line 771  strings.
771  .sp  .sp
772  The following table lists the error codes than may be returned by  The following table lists the error codes than may be returned by
773  \fBpcre_compile2()\fP, along with the error messages that may be returned by  \fBpcre_compile2()\fP, along with the error messages that may be returned by
774  both compiling functions.  both compiling functions. As PCRE has developed, some error codes have fallen
775    out of use. To avoid confusion, they have not been re-used.
776  .sp  .sp
777     0  no error     0  no error
778     1  \e at end of pattern     1  \e at end of pattern
# Line 563  both compiling functions. Line 784  both compiling functions.
784     7  invalid escape sequence in character class     7  invalid escape sequence in character class
785     8  range out of order in character class     8  range out of order in character class
786     9  nothing to repeat     9  nothing to repeat
787    10  operand of unlimited repeat could match the empty string    10  [this code is not in use]
788    11  internal error: unexpected repeat    11  internal error: unexpected repeat
789    12  unrecognized character after (?    12  unrecognized character after (? or (?-
790    13  POSIX named classes are supported only within a class    13  POSIX named classes are supported only within a class
791    14  missing )    14  missing )
792    15  reference to non-existent subpattern    15  reference to non-existent subpattern
793    16  erroffset passed as NULL    16  erroffset passed as NULL
794    17  unknown option bit(s) set    17  unknown option bit(s) set
795    18  missing ) after comment    18  missing ) after comment
796    19  parentheses nested too deeply    19  [this code is not in use]
797    20  regular expression too large    20  regular expression is too large
798    21  failed to get memory    21  failed to get memory
799    22  unmatched parentheses    22  unmatched parentheses
800    23  internal error: code overflow    23  internal error: code overflow
801    24  unrecognized character after (?<    24  unrecognized character after (?<
802    25  lookbehind assertion is not fixed length    25  lookbehind assertion is not fixed length
803    26  malformed number after (?(    26  malformed number or name after (?(
804    27  conditional group contains more than two branches    27  conditional group contains more than two branches
805    28  assertion expected after (?(    28  assertion expected after (?(
806    29  (?R or (?digits must be followed by )    29  (?R or (?[+-]digits must be followed by )
807    30  unknown POSIX class name    30  unknown POSIX class name
808    31  POSIX collating elements are not supported    31  POSIX collating elements are not supported
809    32  this version of PCRE is not compiled with PCRE_UTF8 support    32  this version of PCRE is not compiled with PCRE_UTF8 support
810    33  spare error    33  [this code is not in use]
811    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
812    35  invalid condition (?(0)    35  invalid condition (?(0)
813    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
814    37  PCRE does not support \eL, \el, \eN, \eU, or \eu    37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
815    38  number after (?C is > 255    38  number after (?C is > 255
816    39  closing ) for (?C expected    39  closing ) for (?C expected
817    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
818    41  unrecognized character after (?P    41  unrecognized character after (?P
819    42  syntax error after (?P    42  syntax error in subpattern name (missing terminator)
820    43  two named groups have the same name    43  two named subpatterns have the same name
821    44  invalid UTF-8 string    44  invalid UTF-8 string
822    45  support for \eP, \ep, and \eX has not been compiled    45  support for \eP, \ep, and \eX has not been compiled
823    46  malformed \eP or \ep sequence    46  malformed \eP or \ep sequence
824    47  unknown property name after \eP or \ep    47  unknown property name after \eP or \ep
825      48  subpattern name is too long (maximum 32 characters)
826      49  too many named subpatterns (maximum 10000)
827      50  [this code is not in use]
828      51  octal value is greater than \e377 (not in UTF-8 mode)
829      52  internal error: overran compiling workspace
830      53  internal error: previously-checked referenced subpattern
831            not found
832      54  DEFINE group contains more than one branch
833      55  repeating a DEFINE group is not allowed
834      56  inconsistent NEWLINE options
835      57  \eg is not followed by a braced, angle-bracketed, or quoted
836            name/number or by a plain number
837      58  a numbered reference must not be zero
838      59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
839      60  (*VERB) not recognized
840      61  number is too big
841      62  subpattern name expected
842      63  digit expected after (?+
843      64  ] is an invalid data character in JavaScript compatibility mode
844      65  different names for subpatterns of the same number are
845            not allowed
846      66  (*MARK) must have an argument
847      67  this version of PCRE is not compiled with PCRE_UCP support
848      68  \ec must be followed by an ASCII character
849      69  \ek is not followed by a braced, angle-bracketed, or quoted name
850    .sp
851    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
852    be used if the limits were changed when PCRE was built.
853  .  .
854  .  .
855    .\" HTML <a name="studyingapattern"></a>
856  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
857  .rs  .rs
858  .sp  .sp
# Line 619  help speed up matching, \fBpcre_study()\ Line 869  help speed up matching, \fBpcre_study()\
869  results of the study.  results of the study.
870  .P  .P
871  The returned value from \fBpcre_study()\fP can be passed directly to  The returned value from \fBpcre_study()\fP can be passed directly to
872  \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other  \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
873  fields that can be set by the caller before the block is passed; these are  also contains other fields that can be set by the caller before the block is
874  described  passed; these are described
875  .\" HTML <a href="#extradata">  .\" HTML <a href="#extradata">
876  .\" </a>  .\" </a>
877  below  below
878  .\"  .\"
879  in the section on matching a pattern.  in the section on matching a pattern.
880  .P  .P
881  If studying the pattern does not produce any additional information  If studying the pattern does not produce any useful information,
882  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
883  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its  wants to pass any of the other fields to \fBpcre_exec()\fP or
884  own \fBpcre_extra\fP block.  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
885  .P  .P
886  The second argument of \fBpcre_study()\fP contains option bits. At present, no  The second argument of \fBpcre_study()\fP contains option bits. There is only
887  options are defined, and this argument should always be zero.  one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
888    compiler is available, the pattern is further compiled into machine code that
889    executes much faster than the \fBpcre_exec()\fP matching function. If
890    the just-in-time compiler is not available, this option is ignored. All other
891    bits in the \fIoptions\fP argument must be zero.
892    .P
893    JIT compilation is a heavyweight optimization. It can take some time for
894    patterns to be analyzed, and for one-off matches and simple patterns the
895    benefit of faster execution might be offset by a much slower study time.
896    Not all patterns can be optimized by the JIT compiler. For those that cannot be
897    handled, matching automatically falls back to the \fBpcre_exec()\fP
898    interpreter. For more details, see the
899    .\" HREF
900    \fBpcrejit\fP
901    .\"
902    documentation.
903  .P  .P
904  The third argument for \fBpcre_study()\fP is a pointer for an error message. If  The third argument for \fBpcre_study()\fP is a pointer for an error message. If
905  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
# Line 643  static string that is part of the librar Line 908  static string that is part of the librar
908  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
909  sure that it has run successfully.  sure that it has run successfully.
910  .P  .P
911  This is a typical call to \fBpcre_study\fP():  When you are finished with a pattern, you can free the memory used for the
912    study data by calling \fBpcre_free_study()\fP. This function was added to the
913    API for release 8.20. For earlier versions, the memory could be freed with
914    \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
915    where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
916    new function when convenient.
917    .P
918    This is a typical way in which \fBpcre_study\fP() is used (except that in a
919    real application there should be tests for errors):
920  .sp  .sp
921    pcre_extra *pe;    int rc;
922    pe = pcre_study(    pcre *re;
923      pcre_extra *sd;
924      re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
925      sd = pcre_study(
926      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
927      0,              /* no options exist */      0,              /* no options */
928      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
929  .sp    rc = pcre_exec(   /* see below for details of pcre_exec() options */
930  At present, studying a pattern is useful only for non-anchored patterns that do      re, sd, "subject", 7, 0, 0, ovector, 30);
931  not have a single fixed starting character. A bitmap of possible starting    ...
932  bytes is created.    pcre_free_study(sd);
933      pcre_free(re);
934    .sp
935    Studying a pattern does two things: first, a lower bound for the length of
936    subject string that is needed to match the pattern is computed. This does not
937    mean that there are any strings of that length that match, but it does
938    guarantee that no shorter strings match. The value is used by
939    \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
940    match strings that are shorter than the lower bound. You can find out the value
941    in a calling program via the \fBpcre_fullinfo()\fP function.
942    .P
943    Studying a pattern is also useful for non-anchored patterns that do not have a
944    single fixed starting character. A bitmap of possible starting bytes is
945    created. This speeds up finding a position in the subject at which to start
946    matching.
947    .P
948    These two optimizations apply to both \fBpcre_exec()\fP and
949    \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
950    \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
951    just-in-time compiling is successful. The optimizations can be disabled by
952    setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
953    \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
954    callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
955    to make use of these facilities in cases where matching fails. See the
956    discussion of PCRE_NO_START_OPTIMIZE
957    .\" HTML <a href="#execoptions">
958    .\" </a>
959    below.
960    .\"
961  .  .
962  .  .
963  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
964  .SH "LOCALE SUPPORT"  .SH "LOCALE SUPPORT"
965  .rs  .rs
966  .sp  .sp
967  PCRE handles caseless matching, and determines whether characters are letters  PCRE handles caseless matching, and determines whether characters are letters,
968  digits, or whatever, by reference to a set of tables, indexed by character  digits, or whatever, by reference to a set of tables, indexed by character
969  value. When running in UTF-8 mode, this applies only to characters with codes  value. When running in UTF-8 mode, this applies only to characters with codes
970  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  less than 128. By default, higher-valued codes never match escapes such as \ew
971  can be tested with \ep if PCRE is built with Unicode character property  or \ed, but they can be tested with \ep if PCRE is built with Unicode character
972  support. The use of locales with Unicode is discouraged.  property support. Alternatively, the PCRE_UCP option can be set at compile
973  .P  time; this causes \ew and friends to use Unicode property support instead of
974  An internal set of tables is created in the default C locale when PCRE is  built-in tables. The use of locales with Unicode is discouraged. If you are
975  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  handling characters with codes greater than 128, you should either use UTF-8
976  and is sufficient for many applications. An alternative set of tables can,  and Unicode, or use locales, but not try to mix the two.
977  however, be supplied. These may be created in a different locale from the  .P
978  default. As more and more applications change to using Unicode, the need for  PCRE contains an internal set of tables that are used when the final argument
979  this locale support is expected to die away.  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
980    Normally, the internal tables recognize only ASCII characters. However, when
981    PCRE is built, it is possible to cause the internal tables to be rebuilt in the
982    default "C" locale of the local system, which may cause them to be different.
983    .P
984    The internal tables can always be overridden by tables supplied by the
985    application that calls PCRE. These may be created in a different locale from
986    the default. As more and more applications change to using Unicode, the need
987    for this locale support is expected to die away.
988  .P  .P
989  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
990  which has no arguments, in the relevant locale. The result can then be passed  which has no arguments, in the relevant locale. The result can then be passed
# Line 685  the following code could be used: Line 997  the following code could be used:
997    tables = pcre_maketables();    tables = pcre_maketables();
998    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
999  .sp  .sp
1000    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1001    are using Windows, the name for the French locale is "french".
1002    .P
1003  When \fBpcre_maketables()\fP runs, the tables are built in memory that is  When \fBpcre_maketables()\fP runs, the tables are built in memory that is
1004  obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure  obtained via \fBpcre_malloc\fP. It is the caller's responsibility to ensure
1005  that the memory containing the tables remains available for as long as it is  that the memory containing the tables remains available for as long as it is
# Line 703  one in which it was compiled. Passing ta Line 1018  one in which it was compiled. Passing ta
1018  below in the section on matching a pattern.  below in the section on matching a pattern.
1019  .  .
1020  .  .
1021    .\" HTML <a name="infoaboutpattern"></a>
1022  .SH "INFORMATION ABOUT A PATTERN"  .SH "INFORMATION ABOUT A PATTERN"
1023  .rs  .rs
1024  .sp  .sp
# Line 731  check against passing an arbitrary memor Line 1047  check against passing an arbitrary memor
1047  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
1048  .sp  .sp
1049    int rc;    int rc;
1050    unsigned long int length;    size_t length;
1051    rc = pcre_fullinfo(    rc = pcre_fullinfo(
1052      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
1053      pe,               /* result of pcre_study(), or NULL */      sd,               /* result of pcre_study(), or NULL */
1054      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
1055      &length);         /* where to put the data */      &length);         /* where to put the data */
1056  .sp  .sp
# Line 763  a NULL table pointer. Line 1079  a NULL table pointer.
1079    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
1080  .sp  .sp
1081  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
1082  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
1083  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
1084    still recognized for backwards compatibility.)
1085  .P  .P
1086  If there is a fixed first byte, for example, from a pattern such as  If there is a fixed first byte, for example, from a pattern such as
1087  (cat|cow|coyote), it is returned in the integer pointed to by \fIwhere\fP.  (cat|cow|coyote), its value is returned. Otherwise, if either
 Otherwise, if either  
1088  .sp  .sp
1089  (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
1090  starts with "^", or  starts with "^", or
# Line 787  table indicating a fixed set of bytes fo Line 1103  table indicating a fixed set of bytes fo
1103  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
1104  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
1105  .sp  .sp
1106      PCRE_INFO_HASCRORLF
1107    .sp
1108    Return 1 if the pattern contains any explicit matches for CR or LF characters,
1109    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1110    explicit match is either a literal CR or LF character, or \er or \en.
1111    .sp
1112      PCRE_INFO_JCHANGED
1113    .sp
1114    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1115    0. The fourth argument should point to an \fBint\fP variable. (?J) and
1116    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1117    .sp
1118      PCRE_INFO_JIT
1119    .sp
1120    Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1121    just-in-time compiling was successful. The fourth argument should point to an
1122    \fBint\fP variable. A return value of 0 means that JIT support is not available
1123    in this version of PCRE, or that the pattern was not studied with the
1124    PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1125    particular pattern. See the
1126    .\" HREF
1127    \fBpcrejit\fP
1128    .\"
1129    documentation for details of what can and cannot be handled.
1130    .sp
1131    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1132  .sp  .sp
1133  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
# Line 797  follows something of variable length. Fo Line 1138  follows something of variable length. Fo
1138  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1139  is -1.  is -1.
1140  .sp  .sp
1141      PCRE_INFO_MINLENGTH
1142    .sp
1143    If the pattern was studied and a minimum length for matching subject strings
1144    was computed, its value is returned. Otherwise the returned value is -1. The
1145    value is a number of characters, not bytes (this may be relevant in UTF-8
1146    mode). The fourth argument should point to an \fBint\fP variable. A
1147    non-negative value is a lower bound to the length of any matching string. There
1148    may not be any strings of that length that do actually match, but every string
1149    that does match is at least that long.
1150    .sp
1151    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
1152    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
1153    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
1154  .sp  .sp
1155  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
1156  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
1157  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
1158  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
1159  possible to extract the data directly, by first converting the name to a number  substrings by name. It is also possible to extract the data directly, by first
1160  in order to access the correct pointers in the output vector (described with  converting the name to a number in order to access the correct pointers in the
1161  \fBpcre_exec()\fP below). To do the conversion, you need to use the  output vector (described with \fBpcre_exec()\fP below). To do the conversion,
1162  name-to-number map, which is described by these three values.  you need to use the name-to-number map, which is described by these three
1163    values.
1164  .P  .P
1165  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
1166  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
# Line 816  entry; both of these return an \fBint\fP Line 1168  entry; both of these return an \fBint\fP
1168  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
1169  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry
1170  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
1171  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated.
1172  alphabetical order. For example, consider the following pattern (assume  .P
1173  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  The names are in alphabetical order. Duplicate names may appear if (?| is used
1174    to create multiple groups with the same number, as described in the
1175    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1176    .\" </a>
1177    section on duplicate subpattern numbers
1178    .\"
1179    in the
1180    .\" HREF
1181    \fBpcrepattern\fP
1182    .\"
1183    page. Duplicate names for subpatterns with different numbers are permitted only
1184    if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1185    table in the order in which they were found in the pattern. In the absence of
1186    (?| this is the order of increasing number; when (?| is used this is not
1187    necessarily the case because later subpatterns may have lower numbers.
1188    .P
1189    As a simple example of the name/number table, consider the following pattern
1190    (assume PCRE_EXTENDED is set, so white space - including newlines - is
1191    ignored):
1192  .sp  .sp
1193  .\" JOIN  .\" JOIN
1194    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
1195    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
1196  .sp  .sp
1197  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
1198  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
# Line 834  bytes shows in hexadecimal, and undefine Line 1204  bytes shows in hexadecimal, and undefine
1204    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
1205  .sp  .sp
1206  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
1207  name-to-number map, remember that the length of each entry is likely to be  name-to-number map, remember that the length of the entries is likely to be
1208  different for each compiled pattern.  different for each compiled pattern.
1209  .sp  .sp
1210      PCRE_INFO_OKPARTIAL
1211    .sp
1212    Return 1 if the pattern can be used for partial matching with
1213    \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1214    \fBint\fP variable. From release 8.00, this always returns 1, because the
1215    restrictions that previously applied to partial matching have been lifted. The
1216    .\" HREF
1217    \fBpcrepartial\fP
1218    .\"
1219    documentation gives details of partial matching.
1220    .sp
1221    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1222  .sp  .sp
1223  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
1224  argument should point to an \fBunsigned long int\fP variable. These option bits  argument should point to an \fBunsigned long int\fP variable. These option bits
1225  are those specified in the call to \fBpcre_compile()\fP, modified by any  are those specified in the call to \fBpcre_compile()\fP, modified by any
1226  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1227    they are the options that will be in force when matching starts. For example,
1228    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1229    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1230  .P  .P
1231  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
1232  alternatives begin with one of the following:  alternatives begin with one of the following:
# Line 866  variable. Line 1250  variable.
1250  .sp  .sp
1251    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1252  .sp  .sp
1253  Return the size of the data block pointed to by the \fIstudy_data\fP field in  Return the size of the data block pointed to by the \fIstudy_data\fP field in a
1254  a \fBpcre_extra\fP block. That is, it is the value that was passed to  \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no study data,
1255  \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data  zero is returned. The fourth argument should point to a \fBsize_t\fP variable.
1256  created by \fBpcre_study()\fP. The fourth argument should point to a  The \fIstudy_data\fP field is set by \fBpcre_study()\fP to record information
1257  \fBsize_t\fP variable.  that will speed up matching (see the section entitled
1258    .\" HTML <a href="#studyingapattern">
1259    .\" </a>
1260    "Studying a pattern"
1261    .\"
1262    above). The format of the \fIstudy_data\fP block is private, but its length
1263    is made available via this option so that it can be saved and restored (see the
1264    .\" HREF
1265    \fBpcreprecompile\fP
1266    .\"
1267    documentation for details).
1268  .  .
1269  .  .
1270  .SH "OBSOLETE INFO FUNCTION"  .SH "OBSOLETE INFO FUNCTION"
# Line 931  is different. (This seems a highly unlik Line 1325  is different. (This seems a highly unlik
1325  .P  .P
1326  The function \fBpcre_exec()\fP is called to match a subject string against a  The function \fBpcre_exec()\fP is called to match a subject string against a
1327  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1328  pattern has been studied, the result of the study should be passed in the  pattern was studied, the result of the study should be passed in the
1329  \fIextra\fP argument. This function is the main matching facility of the  \fIextra\fP argument. This function is the main matching facility of the
1330  library, and it operates in a Perl-like manner. For specialist use there is  library, and it operates in a Perl-like manner. For specialist use there is
1331  also an alternative matching function, which is described  also an alternative matching function, which is described
# Line 965  Here is an example of a simple call to \ Line 1359  Here is an example of a simple call to \
1359      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1360      30);            /* number of elements (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1361  .  .
1362    .
1363  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1364  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
1365  .rs  .rs
# Line 977  fields (not necessarily in this order): Line 1372  fields (not necessarily in this order):
1372  .sp  .sp
1373    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1374    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1375      void *\fIexecutable_jit\fP;
1376    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1377    unsigned long int \fImatch_limit_recursion\fP;    unsigned long int \fImatch_limit_recursion\fP;
1378    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1379    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1380      unsigned char **\fImark\fP;
1381  .sp  .sp
1382  The \fIflags\fP field is a bitmap that specifies which of the other fields  The \fIflags\fP field is a bitmap that specifies which of the other fields
1383  are set. The flag bits are:  are set. The flag bits are:
1384  .sp  .sp
1385    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1386      PCRE_EXTRA_EXECUTABLE_JIT
1387    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1388    PCRE_EXTRA_MATCH_LIMIT_RECURSION    PCRE_EXTRA_MATCH_LIMIT_RECURSION
1389    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1390    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1391      PCRE_EXTRA_MARK
1392  .sp  .sp
1393  Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the  Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1394  \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with  the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1395  the appropriate flag bit. You should not set this yourself, but you may add to  returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1396  the block by setting the other fields and their corresponding flag bits.  should not set these yourself, but you may add to the block by setting the
1397    other fields and their corresponding flag bits.
1398  .P  .P
1399  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1400  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,
1401  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
1402  classic example is the use of nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1403  .P  .P
1404  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1405  (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the  calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1406  number of times this function is called during a match, which has the effect of  imposed on the number of times this function is called during a match, which
1407  limiting the amount of backtracking that can take place. For patterns that are  has the effect of limiting the amount of backtracking that can take place. For
1408  not anchored, the count restarts from zero for each position in the subject  patterns that are not anchored, the count restarts from zero for each position
1409  string.  in the subject string.
1410    .P
1411    When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1412    with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1413    is entirely different. However, there is still the possibility of runaway
1414    matching that goes on for a very long time, and so the \fImatch_limit\fP value
1415    is also used in this case (but in a different way) to limit how long the
1416    matching can continue.
1417  .P  .P
1418  The default value for the limit can be set when PCRE is built; the default  The default value for the limit can be set when PCRE is built; the default
1419  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
# Line 1019  The \fImatch_limit_recursion\fP field is Line 1426  The \fImatch_limit_recursion\fP field is
1426  instead of limiting the total number of times that \fBmatch()\fP is called, it  instead of limiting the total number of times that \fBmatch()\fP is called, it
1427  limits the depth of recursion. The recursion depth is a smaller number than the  limits the depth of recursion. The recursion depth is a smaller number than the
1428  total number of calls, because not all calls to \fBmatch()\fP are recursive.  total number of calls, because not all calls to \fBmatch()\fP are recursive.
1429  This limit is of use only if it is set smaller than \fImatch_limit\fP.  This limit is of use only if it is set smaller than \fImatch_limit\fP.
1430  .P  .P
1431  Limiting the recursion depth limits the amount of stack that can be used, or,  Limiting the recursion depth limits the amount of machine stack that can be
1432  when PCRE has been compiled to use memory on the heap instead of the stack, the  used, or, when PCRE has been compiled to use memory on the heap instead of the
1433  amount of heap memory that can be used.  stack, the amount of heap memory that can be used. This limit is relevant, and
1434    is ignored, when the pattern was successfully studied with
1435    PCRE_STUDY_JIT_COMPILE.
1436  .P  .P
1437  The default value for \fImatch_limit_recursion\fP can be set when PCRE is  The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1438  built; the default default is the same value as the default for  built; the default default is the same value as the default for
# Line 1032  with a \fBpcre_extra\fP block in which \ Line 1441  with a \fBpcre_extra\fP block in which \
1441  PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit  PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1442  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1443  .P  .P
1444  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,  The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1445  which is described in the  and is described in the
1446  .\" HREF  .\" HREF
1447  \fBpcrecallout\fP  \fBpcrecallout\fP
1448  .\"  .\"
# Line 1052  called. See the Line 1461  called. See the
1461  \fBpcreprecompile\fP  \fBpcreprecompile\fP
1462  .\"  .\"
1463  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1464    .P
1465    If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1466    be set to point to a \fBchar *\fP variable. If the pattern contains any
1467    backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1468    a name to pass back, a pointer to the name string (zero terminated) is placed
1469    in the variable pointed to by the \fImark\fP field. The names are within the
1470    compiled pattern; if you wish to retain such a name you must copy it before
1471    freeing the memory of a compiled pattern. If there is no name to pass back, the
1472    variable pointed to by the \fImark\fP field set to NULL. For details of the
1473    backtracking control verbs, see the section entitled
1474    .\" HTML <a href="pcrepattern#backtrackcontrol">
1475    .\" </a>
1476    "Backtracking control"
1477    .\"
1478    in the
1479    .\" HREF
1480    \fBpcrepattern\fP
1481    .\"
1482    documentation.
1483  .  .
1484    .
1485    .\" HTML <a name="execoptions"></a>
1486  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
1487  .rs  .rs
1488  .sp  .sp
1489  The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_exec()\fP must be
1490  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1491  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1492    PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1493    PCRE_PARTIAL_HARD.
1494  .sp  .sp
1495    PCRE_ANCHORED    PCRE_ANCHORED
1496  .sp  .sp
# Line 1067  matching position. If a pattern was comp Line 1499  matching position. If a pattern was comp
1499  to be anchored by virtue of its contents, it cannot be made unachored at  to be anchored by virtue of its contents, it cannot be made unachored at
1500  matching time.  matching time.
1501  .sp  .sp
1502      PCRE_BSR_ANYCRLF
1503      PCRE_BSR_UNICODE
1504    .sp
1505    These options (which are mutually exclusive) control what the \eR escape
1506    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1507    match any Unicode newline sequence. These options override the choice that was
1508    made or defaulted when the pattern was compiled.
1509    .sp
1510      PCRE_NEWLINE_CR
1511      PCRE_NEWLINE_LF
1512      PCRE_NEWLINE_CRLF
1513      PCRE_NEWLINE_ANYCRLF
1514      PCRE_NEWLINE_ANY
1515    .sp
1516    These options override the newline definition that was chosen or defaulted when
1517    the pattern was compiled. For details, see the description of
1518    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1519    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1520    the way the match position is advanced after a match failure for an unanchored
1521    pattern.
1522    .P
1523    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1524    match attempt for an unanchored pattern fails when the current position is at a
1525    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1526    characters, the match position is advanced by two characters instead of one, in
1527    other words, to after the CRLF.
1528    .P
1529    The above rule is a compromise that makes the most common cases work as
1530    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1531    set), it does not match the string "\er\enA" because, after failing at the
1532    start, it skips both the CR and the LF before retrying. However, the pattern
1533    [\er\en]A does match that string, because it contains an explicit CR or LF
1534    reference, and so advances only by one character after the first failure.
1535    .P
1536    An explicit match for CR of LF is either a literal appearance of one of those
1537    characters, or one of the \er or \en escape sequences. Implicit matches such as
1538    [^X] do not count, nor does \es (which includes CR and LF in the characters
1539    that it matches).
1540    .P
1541    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1542    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1543    .sp
1544    PCRE_NOTBOL    PCRE_NOTBOL
1545  .sp  .sp
1546  This option specifies that first character of the subject string is not the  This option specifies that first character of the subject string is not the
# Line 1091  match the empty string, the entire match Line 1565  match the empty string, the entire match
1565  .sp  .sp
1566    a?b?    a?b?
1567  .sp  .sp
1568  is applied to a string not beginning with "a" or "b", it matches the empty  is applied to a string not beginning with "a" or "b", it matches an empty
1569  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
1570  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".
1571    .sp
1572      PCRE_NOTEMPTY_ATSTART
1573    .sp
1574    This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1575    the start of the subject is permitted. If the pattern is anchored, such a match
1576    can occur only if the pattern contains \eK.
1577    .P
1578    Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1579    does make a special case of a pattern match of the empty string within its
1580    \fBsplit()\fP function, and when using the /g modifier. It is possible to
1581    emulate Perl's behaviour after matching a null string by first trying the match
1582    again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1583    if that fails, by advancing the starting offset (see below) and trying an
1584    ordinary match again. There is some code that demonstrates how to do this in
1585    the
1586    .\" HREF
1587    \fBpcredemo\fP
1588    .\"
1589    sample program. In the most general case, you have to check to see if the
1590    newline convention recognizes CRLF as a newline, and if so, and the current
1591    character is CR followed by LF, advance the starting offset by two characters
1592    instead of one.
1593    .sp
1594      PCRE_NO_START_OPTIMIZE
1595    .sp
1596    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1597    a match, in order to speed up the process. For example, if it is known that an
1598    unanchored match must start with a specific character, it searches the subject
1599    for that character, and fails immediately if it cannot find it, without
1600    actually running the main matching function. This means that a special item
1601    such as (*COMMIT) at the start of a pattern is not considered until after a
1602    suitable starting point for the match has been found. When callouts or (*MARK)
1603    items are in use, these "start-up" optimizations can cause them to be skipped
1604    if the pattern is never actually used. The start-up optimizations are in effect
1605    a pre-scan of the subject that takes place before the pattern is run.
1606    .P
1607    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1608    causing performance to suffer, but ensuring that in cases where the result is
1609    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1610    are considered at every possible starting position in the subject string. If
1611    PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1612    time.
1613  .P  .P
1614  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case  Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1615  of a pattern match of the empty string within its \fBsplit()\fP function, and  Consider the pattern
1616  when using the /g modifier. It is possible to emulate Perl's behaviour after  .sp
1617  matching a null string by first trying the match again at the same offset with    (*COMMIT)ABC
1618  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the  .sp
1619  starting offset (see below) and trying an ordinary match again. There is some  When this is compiled, PCRE records the fact that a match must start with the
1620  code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.  character "A". Suppose the subject string is "DEFABC". The start-up
1621    optimization scans along the subject, finds "A" and runs the first match
1622    attempt from there. The (*COMMIT) item means that the pattern must match the
1623    current starting position, which in this case, it does. However, if the same
1624    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1625    subject string does not happen. The first match attempt is run starting from
1626    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1627    the overall result is "no match". If the pattern is studied, more start-up
1628    optimizations may be used. For example, a minimum length for the subject may be
1629    recorded. Consider the pattern
1630    .sp
1631      (*MARK:A)(X|Y)
1632    .sp
1633    The minimum length for a match is one character. If the subject is "ABC", there
1634    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1635    If the pattern is studied, the final attempt does not take place, because PCRE
1636    knows that the subject is too short, and so the (*MARK) is never encountered.
1637    In this case, studying the pattern does not affect the overall match result,
1638    which is still "no match", but it does affect the auxiliary information that is
1639    returned.
1640  .sp  .sp
1641    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1642  .sp  .sp
1643  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1644  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1645  The value of \fIstartoffset\fP is also checked to ensure that it points to the  The value of \fIstartoffset\fP is also checked to ensure that it points to the
1646  start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,  start of a UTF-8 character. There is a discussion about the validity of UTF-8
1647  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  strings in the
1648  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HTML <a href="pcre.html#utf8strings">
1649    .\" </a>
1650    section on UTF-8 support
1651    .\"
1652    in the main
1653    .\" HREF
1654    \fBpcre\fP
1655    .\"
1656    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1657    the error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is
1658    a truncated UTF-8 character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1659    both cases, information about the precise nature of the error may also be
1660    returned (see the descriptions of these errors in the section entitled \fIError
1661    return values from\fP \fBpcre_exec()\fP
1662    .\" HTML <a href="#errorlist">
1663    .\" </a>
1664    below).
1665    .\"
1666    If \fIstartoffset\fP contains a value that does not point to the start of a
1667    UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1668    returned.
1669  .P  .P
1670  If you already know that your subject is valid, and you want to skip these  If you already know that your subject is valid, and you want to skip these
1671  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
1672  calling \fBpcre_exec()\fP. You might want to do this for the second and  calling \fBpcre_exec()\fP. You might want to do this for the second and
1673  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1674  all the matches in a single subject string. However, you should be sure that  all the matches in a single subject string. However, you should be sure that
1675  the value of \fIstartoffset\fP points to the start of a UTF-8 character. When  the value of \fIstartoffset\fP points to the start of a UTF-8 character (or the
1676  PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a  end of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1677  subject, or a value of \fIstartoffset\fP that does not point to the start of a  invalid UTF-8 string as a subject or an invalid value of \fIstartoffset\fP is
1678  UTF-8 character, is undefined. Your program may crash.  undefined. Your program may crash.
1679  .sp  .sp
1680    PCRE_PARTIAL    PCRE_PARTIAL_HARD
1681  .sp    PCRE_PARTIAL_SOFT
1682  This option turns on the partial matching feature. If the subject string fails  .sp
1683  to match the pattern, but at some point during the matching process the end of  These options turn on the partial matching feature. For backwards
1684  the subject was reached (that is, the subject partially matches the pattern and  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1685  the failure to match occurred only because there were not enough subject  occurs if the end of the subject string is reached successfully, but there are
1686  characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of  not enough subject characters to complete the match. If this happens when
1687  PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what  PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1688  may appear in the pattern. These are discussed in the  testing any remaining alternatives. Only if no complete match can be found is
1689    PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1690    PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1691    but only if no complete match can be found.
1692    .P
1693    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1694    partial match is found, \fBpcre_exec()\fP immediately returns
1695    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1696    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1697    important that an alternative complete match.
1698    .P
1699    In both cases, the portion of the string that was inspected when the partial
1700    match was found is set as the first matching string. There is a more detailed
1701    discussion of partial and multi-segment matching, with examples, in the
1702  .\" HREF  .\" HREF
1703  \fBpcrepartial\fP  \fBpcrepartial\fP
1704  .\"  .\"
1705  documentation.  documentation.
1706  .  .
1707    .
1708  .SS "The string to be matched by \fBpcre_exec()\fP"  .SS "The string to be matched by \fBpcre_exec()\fP"
1709  .rs  .rs
1710  .sp  .sp
1711  The subject string is passed to \fBpcre_exec()\fP as a pointer in  The subject string is passed to \fBpcre_exec()\fP as a pointer in
1712  \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in  \fIsubject\fP, a length (in bytes) in \fIlength\fP, and a starting byte offset
1713  \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a  in \fIstartoffset\fP. If this is negative or greater than the length of the
1714  UTF-8 character. Unlike the pattern string, the subject may contain binary zero  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1715  bytes. When the starting offset is zero, the search for a match starts at the  offset is zero, the search for a match starts at the beginning of the subject,
1716  beginning of the subject, and this is by far the most common case.  and this is by far the most common case. In UTF-8 mode, the byte offset must
1717    point to the start of a UTF-8 character (or the end of the subject). Unlike the
1718    pattern string, the subject may contain binary zero bytes.
1719  .P  .P
1720  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
1721  same subject by calling \fBpcre_exec()\fP again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
# Line 1164  start of the subject, which is deemed to Line 1735  start of the subject, which is deemed to
1735  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
1736  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.
1737  .P  .P
1738    Finding all the matches in a subject is tricky when the pattern can match an
1739    empty string. It is possible to emulate Perl's /g behaviour by first trying the
1740    match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1741    PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1742    and trying an ordinary match again. There is some code that demonstrates how to
1743    do this in the
1744    .\" HREF
1745    \fBpcredemo\fP
1746    .\"
1747    sample program. In the most general case, you have to check to see if the
1748    newline convention recognizes CRLF as a newline, and if so, and the current
1749    character is CR followed by LF, advance the starting offset by two characters
1750    instead of one.
1751    .P
1752  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
1753  attempt to match at the given offset is made. This can only succeed if the  attempt to match at the given offset is made. This can only succeed if the
1754  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.
1755  .  .
1756    .
1757  .SS "How \fBpcre_exec()\fP returns captured substrings"  .SS "How \fBpcre_exec()\fP returns captured substrings"
1758  .rs  .rs
1759  .sp  .sp
# Line 1178  pattern. Following the usage in Jeffrey Line 1764  pattern. Following the usage in Jeffrey
1764  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
1765  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1766  .P  .P
1767  Captured substrings are returned to the caller via a vector of integer offsets  Captured substrings are returned to the caller via a vector of integers whose
1768  whose address is passed in \fIovector\fP. The number of elements in the vector  address is passed in \fIovector\fP. The number of elements in the vector is
1769  is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:  passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1770  this argument is NOT the size of \fIovector\fP in bytes.  argument is NOT the size of \fIovector\fP in bytes.
1771  .P  .P
1772  The first two-thirds of the vector is used to pass back captured substrings,  The first two-thirds of the vector is used to pass back captured substrings,
1773  each substring using a pair of integers. The remaining third of the vector is  each substring using a pair of integers. The remaining third of the vector is
1774  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1775  and is not available for passing back information. The length passed in  and is not available for passing back information. The number passed in
1776  \fIovecsize\fP should always be a multiple of three. If it is not, it is  \fIovecsize\fP should always be a multiple of three. If it is not, it is
1777  rounded down.  rounded down.
1778  .P  .P
1779  When a match is successful, information about captured substrings is returned  When a match is successful, information about captured substrings is returned
1780  in pairs of integers, starting at the beginning of \fIovector\fP, and  in pairs of integers, starting at the beginning of \fIovector\fP, and
1781  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
1782  pair is set to the offset of the first character in a substring, and the second  each pair is set to the byte offset of the first character in a substring, and
1783  is set to the offset of the first character after the end of a substring. The  the second is set to the byte offset of the first character after the end of a
1784  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the  substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1785  subject string matched by the entire pattern. The next pair is used for the  mode. They are not character counts.
1786  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  .P
1787  is the number of pairs that have been set. If there are no capturing  The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1788  subpatterns, the return value from a successful match is 1, indicating that  portion of the subject string matched by the entire pattern. The next pair is
1789  just the first pair of offsets has been set.  used for the first capturing subpattern, and so on. The value returned by
1790  .P  \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1791  Some convenience functions are provided for extracting the captured substrings  For example, if two substrings have been captured, the returned value is 3. If
1792  as separate strings. These are described in the following section.  there are no capturing subpatterns, the return value from a successful match is
1793  .P  1, indicating that just the first pair of offsets has been set.
 It is possible for an capturing subpattern number \fIn+1\fP to match some  
 part of the subject when subpattern \fIn\fP 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.  
1794  .P  .P
1795  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
1796  string that it matched that is returned.  string that it matched that is returned.
1797  .P  .P
1798  If the vector is too small to hold all the captured substring offsets, it is  If the vector is too small to hold all the captured substring offsets, it is
1799  used as far as possible (up to two-thirds of its length), and the function  used as far as possible (up to two-thirds of its length), and the function
1800  returns a value of zero. In particular, if the substring offsets are not of  returns a value of zero. If the substring offsets are not of interest,
1801  interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and  \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and
1802  \fIovecsize\fP as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1803  the \fIovector\fP is not big enough to remember the related substrings, PCRE  the \fIovector\fP is not big enough to remember the related substrings, PCRE
1804  has to get additional memory for use during matching. Thus it is usually  has to get additional memory for use during matching. Thus it is usually
1805  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1806  .P  .P
1807  Note that \fBpcre_info()\fP can be used to find out how many capturing  The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1808  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1809  \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to  \fIovector\fP that will allow for \fIn\fP captured substrings, in addition to
1810  the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.  the offsets of the substring matched by the whole pattern, is (\fIn\fP+1)*3.
1811    .P
1812    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1813    the subject when subpattern \fIn\fP has not been used at all. For example, if
1814    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1815    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1816    happens, both values in the offset pairs corresponding to unused subpatterns
1817    are set to -1.
1818    .P
1819    Offset values that correspond to unused subpatterns at the end of the
1820    expression are also set to -1. For example, if the string "abc" is matched
1821    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1822    return from the function is 2, because the highest used capturing subpattern
1823    number is 1, and the offsets for for the second and third capturing subpatterns
1824    (assuming the vector is large enough, of course) are set to -1.
1825    .P
1826    \fBNote\fP: Elements in the first two-thids of \fIovector\fP that do not
1827    correspond to capturing parentheses in the pattern are never changed. That is,
1828    if a pattern contains \fIn\fP capturing parentheses, no more than
1829    \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1830    elements (in the first two-thirds) retain whatever values they previously had.
1831    .P
1832    Some convenience functions are provided for extracting the captured substrings
1833    as separate strings. These are described below.
1834    .
1835  .  .
1836  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
1837  .SS "Return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
1838  .rs  .rs
1839  .sp  .sp
1840  If \fBpcre_exec()\fP fails, it returns a negative number. The following are  If \fBpcre_exec()\fP fails, it returns a negative number. The following are
# Line 1256  compiled in an environment of one endian Line 1861  compiled in an environment of one endian
1861  other endianness. This is the error that PCRE gives when the magic number is  other endianness. This is the error that PCRE gives when the magic number is
1862  not present.  not present.
1863  .sp  .sp
1864    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1865  .sp  .sp
1866  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1867  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
# Line 1269  If a pattern contains back references, b Line 1874  If a pattern contains back references, b
1874  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
1875  call via \fBpcre_malloc()\fP fails, this error is given. The memory is  call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1876  automatically freed at the end of matching.  automatically freed at the end of matching.
1877    .P
1878    This error is also given if \fBpcre_stack_malloc()\fP fails in
1879    \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1880    \fB--disable-stack-for-recursion\fP.
1881  .sp  .sp
1882    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1883  .sp  .sp
# Line 1282  The backtracking limit, as specified by Line 1891  The backtracking limit, as specified by
1891  \fBpcre_extra\fP structure (or defaulted) was reached. See the description  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1892  above.  above.
1893  .sp  .sp
   PCRE_ERROR_RECURSIONLIMIT (-21)  
 .sp  
 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP  
 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  
 description above.  
 .sp  
1894    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1895  .sp  .sp
1896  This error is never generated by \fBpcre_exec()\fP itself. It is provided for  This error is never generated by \fBpcre_exec()\fP itself. It is provided for
# Line 1299  documentation for details. Line 1902  documentation for details.
1902  .sp  .sp
1903    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
1904  .sp  .sp
1905  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,
1906    and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
1907    (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
1908    UTF-8 character is placed in the first element, and a reason code is placed in
1909    the second element. The reason codes are listed in the
1910    .\" HTML <a href="#badutf8reasons">
1911    .\" </a>
1912    following section.
1913    .\"
1914    For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
1915    truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
1916    PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1917  .sp  .sp
1918    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
1919  .sp  .sp
1920  The UTF-8 byte sequence that was passed as a subject was valid, but the value  The UTF-8 byte sequence that was passed as a subject was checked and found to
1921  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.  be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
1922    \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
1923    end of the subject.
1924  .sp  .sp
1925    PCRE_ERROR_PARTIAL        (-12)    PCRE_ERROR_PARTIAL        (-12)
1926  .sp  .sp
# Line 1316  documentation for details of partial mat Line 1932  documentation for details of partial mat
1932  .sp  .sp
1933    PCRE_ERROR_BADPARTIAL     (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1934  .sp  .sp
1935  The PCRE_PARTIAL option was used with a compiled pattern containing items that  This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1936  are not supported for partial matching. See the  option was used with a compiled pattern containing items that were not
1937  .\" HREF  supported for partial matching. From release 8.00 onwards, there are no
1938  \fBpcrepartial\fP  restrictions on partial matching.
 .\"  
 documentation for details of partial matching.  
1939  .sp  .sp
1940    PCRE_ERROR_INTERNAL       (-14)    PCRE_ERROR_INTERNAL       (-14)
1941  .sp  .sp
# Line 1331  in PCRE or by overwriting of the compile Line 1945  in PCRE or by overwriting of the compile
1945    PCRE_ERROR_BADCOUNT       (-15)    PCRE_ERROR_BADCOUNT       (-15)
1946  .sp  .sp
1947  This error is given if the value of the \fIovecsize\fP argument is negative.  This error is given if the value of the \fIovecsize\fP argument is negative.
1948    .sp
1949      PCRE_ERROR_RECURSIONLIMIT (-21)
1950    .sp
1951    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1952    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1953    description above.
1954    .sp
1955      PCRE_ERROR_BADNEWLINE     (-23)
1956    .sp
1957    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1958    .sp
1959      PCRE_ERROR_BADOFFSET      (-24)
1960    .sp
1961    The value of \fIstartoffset\fP was negative or greater than the length of the
1962    subject, that is, the value in \fIlength\fP.
1963    .sp
1964      PCRE_ERROR_SHORTUTF8      (-25)
1965    .sp
1966    This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
1967    ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
1968    Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
1969    fact sufficient to detect this case, but this special error code for
1970    PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
1971    retained for backwards compatibility.
1972    .sp
1973      PCRE_ERROR_RECURSELOOP    (-26)
1974    .sp
1975    This error is returned when \fBpcre_exec()\fP detects a recursion loop within
1976    the pattern. Specifically, it means that either the whole pattern or a
1977    subpattern has been called recursively for the second time at the same position
1978    in the subject string. Some simple patterns that might do this are detected and
1979    faulted at compile time, but more complicated cases, in particular mutual
1980    recursions between two different subpatterns, cannot be detected until run
1981    time.
1982    .sp
1983      PCRE_ERROR_JIT_STACKLIMIT (-27)
1984    .sp
1985    This error is returned when a pattern that was successfully studied using the
1986    PCRE_STUDY_JIT_COMPILE option is matched, but the memory available for the
1987    just-in-time processing stack is not large enough. See the
1988    .\" HREF
1989    \fBpcrejit\fP
1990    .\"
1991    documentation for more details.
1992    .P
1993    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1994    .
1995    .
1996    .\" HTML <a name="badutf8reasons"></a>
1997    .SS "Reason codes for invalid UTF-8 strings"
1998    .rs
1999    .sp
2000    When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2001    PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2002    least 2, the offset of the start of the invalid UTF-8 character is placed in
2003    the first output vector element (\fIovector[0]\fP) and a reason code is placed
2004    in the second element (\fIovector[1]\fP). The reason codes are given names in
2005    the \fBpcre.h\fP header file:
2006    .sp
2007      PCRE_UTF8_ERR1
2008      PCRE_UTF8_ERR2
2009      PCRE_UTF8_ERR3
2010      PCRE_UTF8_ERR4
2011      PCRE_UTF8_ERR5
2012    .sp
2013    The string ends with a truncated UTF-8 character; the code specifies how many
2014    bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2015    no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2016    allows for up to 6 bytes, and this is checked first; hence the possibility of
2017    4 or 5 missing bytes.
2018    .sp
2019      PCRE_UTF8_ERR6
2020      PCRE_UTF8_ERR7
2021      PCRE_UTF8_ERR8
2022      PCRE_UTF8_ERR9
2023      PCRE_UTF8_ERR10
2024    .sp
2025    The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2026    character do not have the binary value 0b10 (that is, either the most
2027    significant bit is 0, or the next bit is 1).
2028    .sp
2029      PCRE_UTF8_ERR11
2030      PCRE_UTF8_ERR12
2031    .sp
2032    A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2033    these code points are excluded by RFC 3629.
2034    .sp
2035      PCRE_UTF8_ERR13
2036    .sp
2037    A 4-byte character has a value greater than 0x10fff; these code points are
2038    excluded by RFC 3629.
2039    .sp
2040      PCRE_UTF8_ERR14
2041    .sp
2042    A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2043    code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2044    from UTF-8.
2045    .sp
2046      PCRE_UTF8_ERR15
2047      PCRE_UTF8_ERR16
2048      PCRE_UTF8_ERR17
2049      PCRE_UTF8_ERR18
2050      PCRE_UTF8_ERR19
2051    .sp
2052    A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2053    value that can be represented by fewer bytes, which is invalid. For example,
2054    the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2055    one byte.
2056    .sp
2057      PCRE_UTF8_ERR20
2058    .sp
2059    The two most significant bits of the first byte of a character have the binary
2060    value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2061    byte can only validly occur as the second or subsequent byte of a multi-byte
2062    character.
2063    .sp
2064      PCRE_UTF8_ERR21
2065    .sp
2066    The first byte of a character has the value 0xfe or 0xff. These values can
2067    never occur in a valid UTF-8 string.
2068  .  .
2069  .  .
2070  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1342  This error is given if the value of the Line 2076  This error is given if the value of the
2076  .ti +5n  .ti +5n
2077  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
2078  .PP  .PP
 .br  
2079  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2080  .ti +5n  .ti +5n
2081  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
2082  .ti +5n  .ti +5n
2083  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
2084  .PP  .PP
 .br  
2085  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
2086  .ti +5n  .ti +5n
2087  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
# Line 1360  Captured substrings can be accessed dire Line 2092  Captured substrings can be accessed dire
2092  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2093  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
2094  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
2095  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
2096  has a further zero added on the end, but the result is not, of course,  .P
2097  a C string.  A substring that contains a binary zero is correctly extracted and has a
2098    further zero added on the end, but the result is not, of course, a C string.
2099    However, you can process such a string by referring to the length that is
2100    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2101    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2102    for handling strings containing binary zeros, because the end of the final
2103    string is not independently indicated.
2104  .P  .P
2105  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:
2106  \fIsubject\fP is the subject string that has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
# Line 1382  the string is placed in \fIbuffer\fP, wh Line 2120  the string is placed in \fIbuffer\fP, wh
2120  \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is  \fIbuffersize\fP, while for \fBpcre_get_substring()\fP a new block of memory is
2121  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
2122  \fIstringptr\fP. 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
2123  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
2124  .sp  .sp
2125    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
2126  .sp  .sp
# Line 1398  and builds a list of pointers to them. A Line 2136  and builds a list of pointers to them. A
2136  memory that is obtained via \fBpcre_malloc\fP. The address of the memory block  memory that is obtained via \fBpcre_malloc\fP. The address of the memory block
2137  is returned via \fIlistptr\fP, 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
2138  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
2139  function is zero if all went well, or  function is zero if all went well, or the error code
2140  .sp  .sp
2141    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
2142  .sp  .sp
# Line 1417  a previous call of \fBpcre_get_substring Line 2155  a previous call of \fBpcre_get_substring
2155  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2156  the function pointed to by \fBpcre_free\fP, which of course could be called  the function pointed to by \fBpcre_free\fP, which of course could be called
2157  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
2158  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
2159  \fBpcre_free\fP directly; it is for these cases that the functions are  \fBpcre_free\fP directly; it is for these cases that the functions are
2160  provided.  provided.
2161  .  .
# Line 1429  provided. Line 2167  provided.
2167  .ti +5n  .ti +5n
2168  .B const char *\fIname\fP);  .B const char *\fIname\fP);
2169  .PP  .PP
 .br  
2170  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2171  .ti +5n  .ti +5n
2172  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1438  provided. Line 2175  provided.
2175  .ti +5n  .ti +5n
2176  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2177  .PP  .PP
 .br  
2178  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2179  .ti +5n  .ti +5n
2180  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1450  provided. Line 2186  provided.
2186  To extract a substring by name, you first have to find associated number.  To extract a substring by name, you first have to find associated number.
2187  For example, for this pattern  For example, for this pattern
2188  .sp  .sp
2189    (a+)b(?P<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
2190  .sp  .sp
2191  the number of the subpattern called "xxx" is 2. You can find the number from  the number of the subpattern called "xxx" is 2. If the name is known to be
2192  the name by calling \fBpcre_get_stringnumber()\fP. The first argument is the  unique (PCRE_DUPNAMES was not set), you can find the number from the name by
2193  compiled pattern, and the second is the name. The yield of the function is the  calling \fBpcre_get_stringnumber()\fP. The first argument is the compiled
2194    pattern, and the second is the name. The yield of the function is the
2195  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of  subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no subpattern of
2196  that name.  that name.
2197  .P  .P
# Line 1462  Given the number, you can extract the su Line 2199  Given the number, you can extract the su
2199  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
2200  two functions that do the whole job.  two functions that do the whole job.
2201  .P  .P
2202  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
2203  \fIpcre_get_named_substring()\fP are the same as those for the similarly named  \fBpcre_get_named_substring()\fP are the same as those for the similarly named
2204  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
2205  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
2206  .P  .P
# Line 1473  pattern. This is needed in order to gain Line 2210  pattern. This is needed in order to gain
2210  translation table.  translation table.
2211  .P  .P
2212  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2213  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2214  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2215    the behaviour may not be what you want (see the next section).
2216    .P
2217    \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2218    subpatterns with the same number, as described in the
2219    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2220    .\" </a>
2221    section on duplicate subpattern numbers
2222    .\"
2223    in the
2224    .\" HREF
2225    \fBpcrepattern\fP
2226    .\"
2227    page, you cannot use names to distinguish the different subpatterns, because
2228    names are not included in the compiled code. The matching process uses only
2229    numbers. For this reason, the use of different names for subpatterns of the
2230    same number causes an error at compile time.
2231    .
2232    .SH "DUPLICATE SUBPATTERN NAMES"
2233    .rs
2234    .sp
2235    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2236    .ti +5n
2237    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2238    .PP
2239    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2240    are not required to be unique. (Duplicate names are always allowed for
2241    subpatterns with the same number, created by using the (?| feature. Indeed, if
2242    such subpatterns are named, they are required to use the same names.)
2243    .P
2244    Normally, patterns with duplicate names are such that in any one match, only
2245    one of the named subpatterns participates. An example is shown in the
2246    .\" HREF
2247    \fBpcrepattern\fP
2248    .\"
2249    documentation.
2250    .P
2251    When duplicates are present, \fBpcre_copy_named_substring()\fP and
2252    \fBpcre_get_named_substring()\fP return the first substring corresponding to
2253    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2254    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2255    returns one of the numbers that are associated with the name, but it is not
2256    defined which it is.
2257    .P
2258    If you want to get full details of all captured substrings for a given name,
2259    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2260    argument is the compiled pattern, and the second is the name. The third and
2261    fourth are pointers to variables which are updated by the function. After it
2262    has run, they point to the first and last entries in the name-to-number table
2263    for the given name. The function itself returns the length of each entry, or
2264    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2265    described above in the section entitled \fIInformation about a pattern\fP
2266    .\" HTML <a href="#infoaboutpattern">
2267    .\" </a>
2268    above.
2269    .\"
2270    Given all the relevant entries for the name, you can extract each of their
2271    numbers, and hence the captured data, if any.
2272  .  .
2273  .  .
2274  .SH "FINDING ALL POSSIBLE MATCHES"  .SH "FINDING ALL POSSIBLE MATCHES"
# Line 1512  will yield PCRE_ERROR_NOMATCH. Line 2306  will yield PCRE_ERROR_NOMATCH.
2306  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
2307  .P  .P
2308  The function \fBpcre_dfa_exec()\fP is called to match a subject string against  The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2309  a compiled pattern, using a "DFA" matching algorithm. This has different  a compiled pattern, using a matching algorithm that scans the subject string
2310  characteristics to the normal algorithm, and is not compatible with Perl. Some  just once, and does not backtrack. This has different characteristics to the
2311  of the features of PCRE patterns are not supported. Nevertheless, there are  normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2312  times when this kind of matching can be useful. For a discussion of the two  patterns are not supported. Nevertheless, there are times when this kind of
2313  matching algorithms, see the  matching can be useful. For a discussion of the two matching algorithms, and a
2314    list of features that \fBpcre_dfa_exec()\fP does not support, see the
2315  .\" HREF  .\" HREF
2316  \fBpcrematching\fP  \fBpcrematching\fP
2317  .\"  .\"
# Line 1531  here. Line 2326  here.
2326  The two additional arguments provide workspace for the function. The workspace  The two additional arguments provide workspace for the function. The workspace
2327  vector should contain at least 20 elements. It is used for keeping track of  vector should contain at least 20 elements. It is used for keeping track of
2328  multiple paths through the pattern tree. More workspace will be needed for  multiple paths through the pattern tree. More workspace will be needed for
2329  patterns and subjects where there are a lot of possible matches.  patterns and subjects where there are a lot of potential matches.
2330  .P  .P
2331  Here is an example of a simple call to \fBpcre_dfa_exec()\fP:  Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2332  .sp  .sp
# Line 1554  Here is an example of a simple call to \ Line 2349  Here is an example of a simple call to \
2349  .rs  .rs
2350  .sp  .sp
2351  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2352  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NOTBOL,  zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
2353  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2354  PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are  PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2355  the same as for \fBpcre_exec()\fP, so their description is not repeated here.  PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2356  .sp  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2357    PCRE_PARTIAL  so their description is not repeated here.
2358  .sp  .sp
2359  This has the same general effect as it does for \fBpcre_exec()\fP, but the    PCRE_PARTIAL_HARD
2360  details are slightly different. When PCRE_PARTIAL is set for    PCRE_PARTIAL_SOFT
2361  \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into  .sp
2362  PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no  These have the same general effect as they do for \fBpcre_exec()\fP, but the
2363  complete matches, but there is still at least one matching possibility. The  details are slightly different. When PCRE_PARTIAL_HARD is set for
2364  portion of the string that provided the partial match is set as the first  \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2365  matching string.  is reached and there is still at least one matching possibility that requires
2366    additional characters. This happens even if some complete matches have also
2367    been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2368    is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2369    there have been no complete matches, but there is still at least one matching
2370    possibility. The portion of the string that was inspected when the longest
2371    partial match was found is set as the first matching string in both cases.
2372    There is a more detailed discussion of partial and multi-segment matching, with
2373    examples, in the
2374    .\" HREF
2375    \fBpcrepartial\fP
2376    .\"
2377    documentation.
2378  .sp  .sp
2379    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2380  .sp  .sp
2381  Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as  Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2382  soon as it has found one match. Because of the way the DFA algorithm works,  soon as it has found one match. Because of the way the alternative algorithm
2383  this is necessarily the shortest possible match at the first possible matching  works, this is necessarily the shortest possible match at the first possible
2384  point in the subject string.  matching point in the subject string.
2385  .sp  .sp
2386    PCRE_DFA_RESTART    PCRE_DFA_RESTART
2387  .sp  .sp
2388  When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns  When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2389  a partial match, it is possible to call it again, with additional subject  again, with additional subject characters, and have it continue with the same
2390  characters, and have it continue with the same match. The PCRE_DFA_RESTART  match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2391  option requests this action; when it is set, the \fIworkspace\fP and  \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2392  \fIwscount\fP options must reference the same vector as before because data  before because data about the match so far is left in them after a partial
2393  about the match so far is left in them after a partial match. There is more  match. There is more discussion of this facility in the
 discussion of this facility in the  
2394  .\" HREF  .\" HREF
2395  \fBpcrepartial\fP  \fBpcrepartial\fP
2396  .\"  .\"
2397  documentation.  documentation.
2398  .  .
2399    .
2400  .SS "Successful returns from \fBpcre_dfa_exec()\fP"  .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2401  .rs  .rs
2402  .sp  .sp
# Line 1613  the three matched strings are Line 2420  the three matched strings are
2420  On success, the yield of the function is a number greater than zero, which is  On success, the yield of the function is a number greater than zero, which is
2421  the number of matched substrings. The substrings themselves are returned in  the number of matched substrings. The substrings themselves are returned in
2422  \fIovector\fP. Each string uses two elements; the first is the offset to the  \fIovector\fP. Each string uses two elements; the first is the offset to the
2423  start, and the second is the offset to the end. All the strings have the same  start, and the second is the offset to the end. In fact, all the strings have
2424  start offset. (Space could have been saved by giving this only once, but it was  the same start offset. (Space could have been saved by giving this only once,
2425  decided to retain some compatibility with the way \fBpcre_exec()\fP returns  but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2426  data, even though the meaning of the strings is different.)  returns data, even though the meaning of the strings is different.)
2427  .P  .P
2428  The strings are returned in reverse order of length; that is, the longest  The strings are returned in reverse order of length; that is, the longest
2429  matching string is given first. If there were too many matches to fit into  matching string is given first. If there were too many matches to fit into
2430  \fIovector\fP, the yield of the function is zero, and the vector is filled with  \fIovector\fP, the yield of the function is zero, and the vector is filled with
2431  the longest matches.  the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2432    the entire \fIovector\fP for returning matched strings.
2433    .
2434  .  .
2435  .SS "Error returns from \fBpcre_dfa_exec()\fP"  .SS "Error returns from \fBpcre_dfa_exec()\fP"
2436  .rs  .rs
# Line 1643  that it does not support, for instance, Line 2452  that it does not support, for instance,
2452  .sp  .sp
2453    PCRE_ERROR_DFA_UCOND      (-17)    PCRE_ERROR_DFA_UCOND      (-17)
2454  .sp  .sp
2455  This return is given if \fBpcre_dfa_exec()\fP encounters a condition item in a  This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2456  pattern that uses a back reference for the condition. This is not supported.  uses a back reference for the condition, or a test for recursion in a specific
2457    group. These are not supported.
2458  .sp  .sp
2459    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2460  .sp  .sp
2461  This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP  This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2462  block that contains a setting of the \fImatch_limit\fP field. This is not  block that contains a setting of the \fImatch_limit\fP or
2463  supported (it is meaningless).  \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2464    meaningless for DFA matching).
2465  .sp  .sp
2466    PCRE_ERROR_DFA_WSSIZE     (-19)    PCRE_ERROR_DFA_WSSIZE     (-19)
2467  .sp  .sp
# Line 1663  When a recursive subpattern is processed Line 2474  When a recursive subpattern is processed
2474  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2475  error is given if the output vector is not large enough. This should be  error is given if the output vector is not large enough. This should be
2476  extremely rare, as a vector of size 1000 is used.  extremely rare, as a vector of size 1000 is used.
2477  .P  .
2478  .in 0  .
2479  Last updated: 18 January 2006  .SH "SEE ALSO"
2480  .br  .rs
2481  Copyright (c) 1997-2006 University of Cambridge.  .sp
2482    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2483    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2484    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2485    .
2486    .
2487    .SH AUTHOR
2488    .rs
2489    .sp
2490    .nf
2491    Philip Hazel
2492    University Computing Service
2493    Cambridge CB2 3QH, England.
2494    .fi
2495    .
2496    .
2497    .SH REVISION
2498    .rs
2499    .sp
2500    .nf
2501    Last updated: 29 August 2011
2502    Copyright (c) 1997-2011 University of Cambridge.
2503    .fi

Legend:
Removed from v.87  
changed lines
  Added in v.683

  ViewVC Help
Powered by ViewVC 1.1.5