/[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 75 by nigel, Sat Feb 24 21:40:37 2007 UTC revision 412 by ph10, Sat Apr 11 10:34:37 2009 UTC
# Line 1  Line 1 
1  .TH PCRE 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH "PCRE NATIVE API"  .SH "PCRE NATIVE API"
# Line 7  PCRE - Perl-compatible regular expressio Line 7  PCRE - Perl-compatible regular expressio
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
16  .br  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17    .ti +5n
18    .B int *\fIerrorcodeptr\fP,
19    .ti +5n
20    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21    .ti +5n
22    .B const unsigned char *\fItableptr\fP);
23    .PP
24  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .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
 .br  
28  .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,"
29  .ti +5n  .ti +5n
30  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31  .ti +5n  .ti +5n
32  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33  .PP  .PP
34  .br  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35    .ti +5n
36    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37    .ti +5n
38    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39    .ti +5n
40    .B int *\fIworkspace\fP, int \fIwscount\fP);
41    .PP
42  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
43  .ti +5n  .ti +5n
44  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 35  PCRE - Perl-compatible regular expressio Line 47  PCRE - Perl-compatible regular expressio
47  .ti +5n  .ti +5n
48  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49  .PP  .PP
 .br  
50  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
51  .ti +5n  .ti +5n
52  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53  .ti +5n  .ti +5n
54  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
55  .PP  .PP
 .br  
56  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_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 51  PCRE - Perl-compatible regular expressio Line 61  PCRE - Perl-compatible regular expressio
61  .ti +5n  .ti +5n
62  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
63  .PP  .PP
 .br  
64  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65  .ti +5n  .ti +5n
66  .B const char *\fIname\fP);  .B const char *\fIname\fP);
67  .PP  .PP
68  .br  .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
69    .ti +5n
70    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71    .PP
72  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
73  .ti +5n  .ti +5n
74  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
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_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
79  .ti +5n  .ti +5n
80  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81  .PP  .PP
 .br  
82  .B void pcre_free_substring(const char *\fIstringptr\fP);  .B void pcre_free_substring(const char *\fIstringptr\fP);
83  .PP  .PP
 .br  
84  .B void pcre_free_substring_list(const char **\fIstringptr\fP);  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
85  .PP  .PP
 .br  
86  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
87  .PP  .PP
 .br  
88  .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,"
89  .ti +5n  .ti +5n
90  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
91  .PP  .PP
 .br  
92  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93  .B *\fIfirstcharptr\fP);  .B *\fIfirstcharptr\fP);
94  .PP  .PP
95  .br  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
96    .PP
97  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
98  .PP  .PP
 .br  
99  .B char *pcre_version(void);  .B char *pcre_version(void);
100  .PP  .PP
 .br  
101  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
102  .PP  .PP
 .br  
103  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
104  .PP  .PP
 .br  
105  .B void *(*pcre_stack_malloc)(size_t);  .B void *(*pcre_stack_malloc)(size_t);
106  .PP  .PP
 .br  
107  .B void (*pcre_stack_free)(void *);  .B void (*pcre_stack_free)(void *);
108  .PP  .PP
 .br  
109  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
110  .  .
111  .  .
112  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
113  .rs  .rs
114  .sp  .sp
115  PCRE has its own native API, which is described in this document. There is also  PCRE has its own native API, which is described in this document. There are
116  a set of wrapper functions that correspond to the POSIX regular expression API.  also some wrapper functions that correspond to the POSIX regular expression
117  These are described in the  API. These are described in the
118  .\" HREF  .\" HREF
119  \fBpcreposix\fP  \fBpcreposix\fP
120  .\"  .\"
121  documentation.  documentation. Both of these APIs define a set of C function calls. A C++
122    wrapper is distributed with PCRE. It is documented in the
123    .\" HREF
124    \fBpcrecpp\fP
125    .\"
126    page.
127  .P  .P
128  The native API function prototypes are defined in the header file \fBpcre.h\fP,  The native API C function prototypes are defined in the header file
129  and on Unix systems the library itself is called \fBlibpcre\fP. It can  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130  normally be accessed by adding \fB-lpcre\fP to the command for linking an  It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131  application that uses PCRE. The header file defines the macros PCRE_MAJOR and  an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132  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.
133  Applications can use these to include support for different releases of PCRE.  Applications can use these to include support for different releases of PCRE.
134  .P  .P
135  The functions \fBpcre_compile()\fP, \fBpcre_study()\fP, and \fBpcre_exec()\fP  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136  are used for compiling and matching regular expressions. A sample program that  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137  demonstrates the simplest way of using them is provided in the file called  in a Perl-compatible manner. A sample program that demonstrates the simplest
138  \fIpcredemo.c\fP in the source distribution. The  way of using them is provided in the file called \fIpcredemo.c\fP in the source
139    distribution. The
140  .\" HREF  .\" HREF
141  \fBpcresample\fP  \fBpcresample\fP
142  .\"  .\"
143  documentation describes how to run it.  documentation describes how to compile and run it.
144    .P
145    A second matching function, \fBpcre_dfa_exec()\fP, which is not
146    Perl-compatible, is also provided. This uses a different algorithm for the
147    matching. The alternative algorithm finds all possible matches (at a given
148    point in the subject), and scans the subject just once. However, this algorithm
149    does not return captured substrings. A description of the two matching
150    algorithms and their advantages and disadvantages is given in the
151    .\" HREF
152    \fBpcrematching\fP
153    .\"
154    documentation.
155  .P  .P
156  In addition to the main compiling and matching functions, there are convenience  In addition to the main compiling and matching functions, there are convenience
157  functions for extracting captured substrings from a matched subject string.  functions for extracting captured substrings from a subject string that is
158  They are:  matched by \fBpcre_exec()\fP. They are:
159  .sp  .sp
160    \fBpcre_copy_substring()\fP    \fBpcre_copy_substring()\fP
161    \fBpcre_copy_named_substring()\fP    \fBpcre_copy_named_substring()\fP
# Line 145  They are: Line 163  They are:
163    \fBpcre_get_named_substring()\fP    \fBpcre_get_named_substring()\fP
164    \fBpcre_get_substring_list()\fP    \fBpcre_get_substring_list()\fP
165    \fBpcre_get_stringnumber()\fP    \fBpcre_get_stringnumber()\fP
166      \fBpcre_get_stringtable_entries()\fP
167  .sp  .sp
168  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
169  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
170  .P  .P
171  The function \fBpcre_maketables()\fP is used to build a set of character tables  The function \fBpcre_maketables()\fP is used to build a set of character tables
172  in the current locale for passing to \fBpcre_compile()\fP or \fBpcre_exec()\fP.  in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
173  This is an optional facility that is provided for specialist use. Most  or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
174  commonly, no special tables are passed, in which case internal tables that are  specialist use. Most commonly, no special tables are passed, in which case
175  generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
176  .P  .P
177  The function \fBpcre_fullinfo()\fP is used to find out information about a  The function \fBpcre_fullinfo()\fP is used to find out information about a
178  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
# Line 161  some of the available information, but i Line 180  some of the available information, but i
180  The function \fBpcre_version()\fP returns a pointer to a string containing the  The function \fBpcre_version()\fP returns a pointer to a string containing the
181  version of PCRE and its date of release.  version of PCRE and its date of release.
182  .P  .P
183    The function \fBpcre_refcount()\fP maintains a reference count in a data block
184    containing a compiled pattern. This is provided for the benefit of
185    object-oriented applications.
186    .P
187  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
188  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
189  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
# Line 170  should be done before calling any PCRE f Line 193  should be done before calling any PCRE f
193  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
194  indirections to memory management functions. These special functions are used  indirections to memory management functions. These special functions are used
195  only when PCRE is compiled to use the heap for remembering data, instead of  only when PCRE is compiled to use the heap for remembering data, instead of
196  recursive function calls. This is a non-standard way of building PCRE, for use  recursive function calls, when running the \fBpcre_exec()\fP function. See the
197  in environments that have limited stacks. Because of the greater use of memory  .\" HREF
198  management, it runs more slowly. Separate functions are provided so that  \fBpcrebuild\fP
199  special-purpose external code can be used for this case. When used, these  .\"
200  functions are always called in a stack-like manner (last obtained, first  documentation for details of how to do this. It is a non-standard way of
201  freed), and always for memory blocks of the same size.  building PCRE, for use in environments that have limited stacks. Because of the
202    greater use of memory management, it runs more slowly. Separate functions are
203    provided so that special-purpose external code can be used for this case. When
204    used, these functions are always called in a stack-like manner (last obtained,
205    first freed), and always for memory blocks of the same size. There is a
206    discussion about PCRE's stack usage in the
207    .\" HREF
208    \fBpcrestack\fP
209    .\"
210    documentation.
211  .P  .P
212  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
213  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
# Line 186  points during a matching operation. Deta Line 218  points during a matching operation. Deta
218  documentation.  documentation.
219  .  .
220  .  .
221    .\" HTML <a name="newlines"></a>
222    .SH NEWLINES
223    .rs
224    .sp
225    PCRE supports five different conventions for indicating line breaks in
226    strings: a single CR (carriage return) character, a single LF (linefeed)
227    character, the two-character sequence CRLF, any of the three preceding, or any
228    Unicode newline sequence. The Unicode newline sequences are the three just
229    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
230    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
231    (paragraph separator, U+2029).
232    .P
233    Each of the first three conventions is used by at least one operating system as
234    its standard newline sequence. When PCRE is built, a default can be specified.
235    The default default is LF, which is the Unix standard. When PCRE is run, the
236    default can be overridden, either when a pattern is compiled, or when it is
237    matched.
238    .P
239    At compile time, the newline convention can be specified by the \fIoptions\fP
240    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
241    start of the pattern itself; this overrides any other settings. See the
242    .\" HREF
243    \fBpcrepattern\fP
244    .\"
245    page for details of the special character sequences.
246    .P
247    In the PCRE documentation the word "newline" is used to mean "the character or
248    pair of characters that indicate a line break". The choice of newline
249    convention affects the handling of the dot, circumflex, and dollar
250    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
251    recognized line ending sequence, the match position advancement for a
252    non-anchored pattern. There is more detail about this in the
253    .\" HTML <a href="#execoptions">
254    .\" </a>
255    section on \fBpcre_exec()\fP options
256    .\"
257    below.
258    .P
259    The choice of newline convention does not affect the interpretation of
260    the \en or \er escape sequences, nor does it affect what \eR matches, which is
261    controlled in a similar way, but by separate options.
262    .
263    .
264  .SH MULTITHREADING  .SH MULTITHREADING
265  .rs  .rs
266  .sp  .sp
# Line 207  which it was compiled. Details are given Line 282  which it was compiled. Details are given
282  .\" HREF  .\" HREF
283  \fBpcreprecompile\fP  \fBpcreprecompile\fP
284  .\"  .\"
285  documentation.  documentation. However, compiling a regular expression with one version of PCRE
286    for use with a different version is not guaranteed to work and may cause
287    crashes.
288  .  .
289  .  .
290  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 238  properties is available; otherwise it is Line 315  properties is available; otherwise it is
315  .sp  .sp
316    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
317  .sp  .sp
318  The output is an integer that is set to the value of the code that is used for  The output is an integer whose value specifies the default character sequence
319  the newline character. It is either linefeed (10) or carriage return (13), and  that is recognized as meaning "newline". The four values that are supported
320  should normally be the standard character for your operating system.  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY.
321    Though they are derived from ASCII, the same values are returned in EBCDIC
322    environments. The default should normally correspond to the standard sequence
323    for your operating system.
324    .sp
325      PCRE_CONFIG_BSR
326    .sp
327    The output is an integer whose value indicates what character sequences the \eR
328    escape sequence matches by default. A value of 0 means that \eR matches any
329    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
330    or CRLF. The default can be overridden when a pattern is compiled or matched.
331  .sp  .sp
332    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
333  .sp  .sp
# Line 262  documentation. Line 349  documentation.
349  .sp  .sp
350    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
351  .sp  .sp
352  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
353  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
354  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
355  .sp  .sp
356      PCRE_CONFIG_MATCH_LIMIT_RECURSION
357    .sp
358    The output is a long integer that gives the default limit for the depth of
359    recursion when calling the internal matching function in a \fBpcre_exec()\fP
360    execution. Further details are given with \fBpcre_exec()\fP below.
361    .sp
362    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
363  .sp  .sp
364  The output is an integer that is set to one if internal recursion is  The output is an integer that is set to one if internal recursion when running
365  implemented by recursive function calls that use the stack to remember their  \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
366  state. This is the usual way that PCRE is compiled. The output is zero if PCRE  to remember their state. This is the usual way that PCRE is compiled. The
367  was compiled to use blocks of data on the heap instead of recursive function  output is zero if PCRE was compiled to use blocks of data on the heap instead
368  calls. In this case, \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are  of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
369  called to manage memory blocks on the heap, thus avoiding the use of the stack.  \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
370    avoiding the use of the stack.
371  .  .
372  .  .
373  .SH "COMPILING A PATTERN"  .SH "COMPILING A PATTERN"
# Line 284  called to manage memory blocks on the he Line 378  called to manage memory blocks on the he
378  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
379  .ti +5n  .ti +5n
380  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
381    .sp
382    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
383    .ti +5n
384    .B int *\fIerrorcodeptr\fP,
385    .ti +5n
386    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
387    .ti +5n
388    .B const unsigned char *\fItableptr\fP);
389  .P  .P
390  The function \fBpcre_compile()\fP is called to compile a pattern into an  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
391  internal form. The pattern is a C string terminated by a binary zero, and  called to compile a pattern into an internal form. The only difference between
392  is passed in the \fIpattern\fP argument. A pointer to a single block of memory  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
393  that is obtained via \fBpcre_malloc\fP is returned. This contains the compiled  \fIerrorcodeptr\fP, via which a numerical error code can be returned.
394  code and related data. The \fBpcre\fP type is defined for the returned block;  .P
395  this is a typedef for a structure whose contents are not externally defined. It  The pattern is a C string terminated by a binary zero, and is passed in the
396  is up to the caller to free the memory when it is no longer required.  \fIpattern\fP argument. A pointer to a single block of memory that is obtained
397    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
398    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
399    for a structure whose contents are not externally defined. It is up to the
400    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
401  .P  .P
402  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
403  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
404  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
405  argument, which is an address (see below).  argument, which is an address (see below).
406  .P  .P
407  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
408  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
409  options are described below. Some of them, in particular, those that are  options are described below. Some of them (in particular, those that are
410  compatible with Perl, can also be set and unset from within the pattern (see  compatible with Perl, but also some others) can also be set and unset from
411  the detailed description in the  within the pattern (see the detailed description in the
412  .\" HREF  .\" HREF
413  \fBpcrepattern\fP  \fBpcrepattern\fP
414  .\"  .\"
415  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For those options that can be different in different parts of
416  specifies their initial settings at the start of compilation and execution. The  the pattern, the contents of the \fIoptions\fP argument specifies their initial
417  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 and
418  time.  PCRE_NEWLINE_\fIxxx\fP options can be set at the time of matching as well as at
419    compile time.
420  .P  .P
421  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
422  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
423  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
424  error message. The offset from the start of the pattern to the character where  error message. This is a static string that is part of the library. You must
425  the error was discovered is placed in the variable pointed to by  not try to free it. The offset from the start of the pattern to the character
426    where the error was discovered is placed in the variable pointed to by
427  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
428  .P  .P
429    If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
430    \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
431    returned via this argument in the event of an error. This is in addition to the
432    textual error message. Error codes and messages are listed below.
433    .P
434  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
435  character tables that are built when PCRE is compiled, using the default C  character tables that are built when PCRE is compiled, using the default C
436  locale. Otherwise, \fItableptr\fP must be an address that is the result of a  locale. Otherwise, \fItableptr\fP must be an address that is the result of a
# Line 358  facility, see the Line 471  facility, see the
471  .\"  .\"
472  documentation.  documentation.
473  .sp  .sp
474      PCRE_BSR_ANYCRLF
475      PCRE_BSR_UNICODE
476    .sp
477    These options (which are mutually exclusive) control what the \eR escape
478    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
479    match any Unicode newline sequence. The default is specified when PCRE is
480    built. It can be overridden from within the pattern, or by setting an option
481    when a compiled pattern is matched.
482    .sp
483    PCRE_CASELESS    PCRE_CASELESS
484  .sp  .sp
485  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
486  letters. It is equivalent to Perl's /i option, and it can be changed within a  letters. It is equivalent to Perl's /i option, and it can be changed within a
487  pattern by a (?i) option setting. When running in UTF-8 mode, case support for  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
488  high-valued characters is available only when PCRE is built with Unicode  concept of case for characters whose values are less than 128, so caseless
489  character property support.  matching is always possible. For characters with higher values, the concept of
490    case is supported if PCRE is compiled with Unicode property support, but not
491    otherwise. If you want to use caseless matching for characters 128 and above,
492    you must ensure that PCRE is compiled with Unicode property support as well as
493    with UTF-8 support.
494  .sp  .sp
495    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
496  .sp  .sp
497  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
498  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
499  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
500  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.
501  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
502  a pattern.  pattern.
503  .sp  .sp
504    PCRE_DOTALL    PCRE_DOTALL
505  .sp  .sp
506  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharater in the pattern matches all characters,
507  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
508  equivalent to Perl's /s option, and it can be changed within a pattern by a  the current position is at a newline. This option is equivalent to Perl's /s
509  (?s) option setting. A negative class such as [^a] always matches a newline  option, and it can be changed within a pattern by a (?s) option setting. A
510  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
511    the setting of this option.
512    .sp
513      PCRE_DUPNAMES
514    .sp
515    If this bit is set, names used to identify capturing subpatterns need not be
516    unique. This can be helpful for certain types of pattern when it is known that
517    only one instance of the named subpattern can ever be matched. There are more
518    details of named subpatterns below; see also the
519    .\" HREF
520    \fBpcrepattern\fP
521    .\"
522    documentation.
523  .sp  .sp
524    PCRE_EXTENDED    PCRE_EXTENDED
525  .sp  .sp
526  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
527  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
528  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
529  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
530  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
531  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
532  .P  .P
533  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
534  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
# Line 404  that is incompatible with Perl, but it i Line 542  that is incompatible with Perl, but it i
542  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
543  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
544  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
545  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
546  controlled by this option. It can also be set by a (?X) option setting within a  give a warning for this.) There are at present no other features controlled by
547  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
548    .sp
549      PCRE_FIRSTLINE
550    .sp
551    If this option is set, an unanchored pattern is required to match before or at
552    the first newline in the subject string, though the matched text may continue
553    over the newline.
554    .sp
555      PCRE_JAVASCRIPT_COMPAT
556    .sp
557    If this option is set, PCRE's behaviour is changed in some ways so that it is
558    compatible with JavaScript rather than Perl. The changes are as follows:
559    .P
560    (1) A lone closing square bracket in a pattern causes a compile-time error,
561    because this is illegal in JavaScript (by default it is treated as a data
562    character). Thus, the pattern AB]CD becomes illegal when this option is set.
563    .P
564    (2) At run time, a back reference to an unset subpattern group matches an empty
565    string (by default this causes the current matching alternative to fail). A
566    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
567    an "a" in the subject), whereas it fails by default, for Perl compatibility.
568  .sp  .sp
569    PCRE_MULTILINE    PCRE_MULTILINE
570  .sp  .sp
# Line 418  terminating newline (unless PCRE_DOLLAR_ Line 576  terminating newline (unless PCRE_DOLLAR_
576  Perl.  Perl.
577  .P  .P
578  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
579  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
580  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
581  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
582  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
583  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
584  .sp  .sp
585      PCRE_NEWLINE_CR
586      PCRE_NEWLINE_LF
587      PCRE_NEWLINE_CRLF
588      PCRE_NEWLINE_ANYCRLF
589      PCRE_NEWLINE_ANY
590    .sp
591    These options override the default newline definition that was chosen when PCRE
592    was built. Setting the first or the second specifies that a newline is
593    indicated by a single character (CR or LF, respectively). Setting
594    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
595    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
596    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
597    that any Unicode newline sequence should be recognized. The Unicode newline
598    sequences are the three just mentioned, plus the single characters VT (vertical
599    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
600    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
601    recognized only in UTF-8 mode.
602    .P
603    The newline setting in the options word uses three bits that are treated
604    as a number, giving eight possibilities. Currently only six are used (default
605    plus the five values above). This means that if you set more than one newline
606    option, the combination may or may not be sensible. For example,
607    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
608    other combinations may yield unused numbers and cause an error.
609    .P
610    The only time that a line break is specially recognized when compiling a
611    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
612    class is encountered. This indicates a comment that lasts until after the next
613    line break sequence. In other circumstances, line break sequences are treated
614    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
615    as whitespace characters and are therefore ignored.
616    .P
617    The newline option that is set at compile time becomes the default that is used
618    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
619    .sp
620    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
621  .sp  .sp
622  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 458  page. Line 651  page.
651    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
652  .sp  .sp
653  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
654  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
655  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcre.html#utf8strings">
656  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
657  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
658  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
659  Note that this option can also be passed to \fBpcre_exec()\fP, to suppress the  in the main
660  UTF-8 validity checking of subject strings.  .\" HREF
661    \fBpcre\fP
662    .\"
663    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
664    returns an error. If you already know that your pattern is valid, and you want
665    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
666    option. When it is set, the effect of passing an invalid UTF-8 string as a
667    pattern is undefined. It may cause your program to crash. Note that this option
668    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
669    the UTF-8 validity checking of subject strings.
670    .
671    .
672    .SH "COMPILATION ERROR CODES"
673    .rs
674    .sp
675    The following table lists the error codes than may be returned by
676    \fBpcre_compile2()\fP, along with the error messages that may be returned by
677    both compiling functions. As PCRE has developed, some error codes have fallen
678    out of use. To avoid confusion, they have not been re-used.
679    .sp
680       0  no error
681       1  \e at end of pattern
682       2  \ec at end of pattern
683       3  unrecognized character follows \e
684       4  numbers out of order in {} quantifier
685       5  number too big in {} quantifier
686       6  missing terminating ] for character class
687       7  invalid escape sequence in character class
688       8  range out of order in character class
689       9  nothing to repeat
690      10  [this code is not in use]
691      11  internal error: unexpected repeat
692      12  unrecognized character after (? or (?-
693      13  POSIX named classes are supported only within a class
694      14  missing )
695      15  reference to non-existent subpattern
696      16  erroffset passed as NULL
697      17  unknown option bit(s) set
698      18  missing ) after comment
699      19  [this code is not in use]
700      20  regular expression is too large
701      21  failed to get memory
702      22  unmatched parentheses
703      23  internal error: code overflow
704      24  unrecognized character after (?<
705      25  lookbehind assertion is not fixed length
706      26  malformed number or name after (?(
707      27  conditional group contains more than two branches
708      28  assertion expected after (?(
709      29  (?R or (?[+-]digits must be followed by )
710      30  unknown POSIX class name
711      31  POSIX collating elements are not supported
712      32  this version of PCRE is not compiled with PCRE_UTF8 support
713      33  [this code is not in use]
714      34  character value in \ex{...} sequence is too large
715      35  invalid condition (?(0)
716      36  \eC not allowed in lookbehind assertion
717      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
718      38  number after (?C is > 255
719      39  closing ) for (?C expected
720      40  recursive call could loop indefinitely
721      41  unrecognized character after (?P
722      42  syntax error in subpattern name (missing terminator)
723      43  two named subpatterns have the same name
724      44  invalid UTF-8 string
725      45  support for \eP, \ep, and \eX has not been compiled
726      46  malformed \eP or \ep sequence
727      47  unknown property name after \eP or \ep
728      48  subpattern name is too long (maximum 32 characters)
729      49  too many named subpatterns (maximum 10000)
730      50  [this code is not in use]
731      51  octal value is greater than \e377 (not in UTF-8 mode)
732      52  internal error: overran compiling workspace
733      53  internal error: previously-checked referenced subpattern not found
734      54  DEFINE group contains more than one branch
735      55  repeating a DEFINE group is not allowed
736      56  inconsistent NEWLINE options
737      57  \eg is not followed by a braced, angle-bracketed, or quoted
738            name/number or by a plain number
739      58  a numbered reference must not be zero
740      59  (*VERB) with an argument is not supported
741      60  (*VERB) not recognized
742      61  number is too big
743      62  subpattern name expected
744      63  digit expected after (?+
745      64  ] is an invalid data character in JavaScript compatibility mode
746    .sp
747    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
748    be used if the limits were changed when PCRE was built.
749  .  .
750  .  .
751  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
752  .rs  .rs
753  .sp  .sp
754  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
755  .ti +5n  .ti +5n
756  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
757  .PP  .PP
# Line 492  below Line 773  below
773  .\"  .\"
774  in the section on matching a pattern.  in the section on matching a pattern.
775  .P  .P
776  If studying the pattern does not produce any additional information,  If studying the pattern does not produce any additional information
777  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
778  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, it must set up its
779  own \fBpcre_extra\fP block.  own \fBpcre_extra\fP block.
# Line 502  options are defined, and this argument s Line 783  options are defined, and this argument s
783  .P  .P
784  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
785  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
786  set to NULL. Otherwise it points to a textual error message. You should  set to NULL. Otherwise it is set to point to a textual error message. This is a
787  therefore test the error pointer for NULL after calling \fBpcre_study()\fP, to  static string that is part of the library. You must not try to free it. You
788  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
789    sure that it has run successfully.
790  .P  .P
791  This is a typical call to \fBpcre_study\fP():  This is a typical call to \fBpcre_study\fP():
792  .sp  .sp
# Line 525  bytes is created. Line 807  bytes is created.
807  .sp  .sp
808  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
809  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
810  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
811  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
812  can be tested with \ep if PCRE is built with Unicode character property  can be tested with \ep if PCRE is built with Unicode character property
813  support.)  support. The use of locales with Unicode is discouraged. If you are handling
814  .P  characters with codes greater than 128, you should either use UTF-8 and
815  An internal set of tables is created in the default C locale when PCRE is  Unicode, or use locales, but not try to mix the two.
816  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  .P
817  and is sufficient for many applications. An alternative set of tables can,  PCRE contains an internal set of tables that are used when the final argument
818  however, be supplied. These may be created in a different locale from the  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
819  default. As more and more applications change to using Unicode, the need for  Normally, the internal tables recognize only ASCII characters. However, when
820  this locale support is expected to die away.  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
821    default "C" locale of the local system, which may cause them to be different.
822    .P
823    The internal tables can always be overridden by tables supplied by the
824    application that calls PCRE. These may be created in a different locale from
825    the default. As more and more applications change to using Unicode, the need
826    for this locale support is expected to die away.
827  .P  .P
828  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
829  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 548  the following code could be used: Line 836  the following code could be used:
836    tables = pcre_maketables();    tables = pcre_maketables();
837    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
838  .sp  .sp
839    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
840    are using Windows, the name for the French locale is "french".
841    .P
842  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
843  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
844  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 594  check against passing an arbitrary memor Line 885  check against passing an arbitrary memor
885  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
886  .sp  .sp
887    int rc;    int rc;
888    unsigned long int length;    size_t length;
889    rc = pcre_fullinfo(    rc = pcre_fullinfo(
890      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
891      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
# Line 615  no back references. Line 906  no back references.
906  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
907  should point to an \fBint\fP variable.  should point to an \fBint\fP variable.
908  .sp  .sp
909    PCRE_INFO_DEFAULTTABLES    PCRE_INFO_DEFAULT_TABLES
910  .sp  .sp
911  Return a pointer to the internal default character tables within PCRE. The  Return a pointer to the internal default character tables within PCRE. The
912  fourth argument should point to an \fBunsigned char *\fP variable. This  fourth argument should point to an \fBunsigned char *\fP variable. This
# Line 626  a NULL table pointer. Line 917  a NULL table pointer.
917    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
918  .sp  .sp
919  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
920  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
921  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
922    still recognized for backwards compatibility.)
923  .P  .P
924  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
925  (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  
926  .sp  .sp
927  (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
928  starts with "^", or  starts with "^", or
# Line 650  table indicating a fixed set of bytes fo Line 941  table indicating a fixed set of bytes fo
941  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
942  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
943  .sp  .sp
944      PCRE_INFO_HASCRORLF
945    .sp
946    Return 1 if the pattern contains any explicit matches for CR or LF characters,
947    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
948    explicit match is either a literal CR or LF character, or \er or \en.
949    .sp
950      PCRE_INFO_JCHANGED
951    .sp
952    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
953    0. The fourth argument should point to an \fBint\fP variable. (?J) and
954    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
955    .sp
956    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
957  .sp  .sp
958  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 666  is -1. Line 969  is -1.
969  .sp  .sp
970  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
971  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
972  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
973  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
974  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
975  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
976  \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,
977  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
978    values.
979  .P  .P
980  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
981  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 680  length of the longest name. PCRE_INFO_NA Line 984  length of the longest name. PCRE_INFO_NA
984  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
985  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
986  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated. The names are in
987  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
988    their parentheses numbers. For example, consider the following pattern (assume
989  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
990  .sp  .sp
991  .\" JOIN  .\" JOIN
992    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
993    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
994  .sp  .sp
995  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
996  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 697  bytes shows in hexadecimal, and undefine Line 1002  bytes shows in hexadecimal, and undefine
1002    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
1003  .sp  .sp
1004  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
1005  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
1006  different for each compiled pattern.  different for each compiled pattern.
1007  .sp  .sp
1008      PCRE_INFO_OKPARTIAL
1009    .sp
1010    Return 1 if the pattern can be used for partial matching, otherwise 0. The
1011    fourth argument should point to an \fBint\fP variable. The
1012    .\" HREF
1013    \fBpcrepartial\fP
1014    .\"
1015    documentation lists the restrictions that apply to patterns when partial
1016    matching is used.
1017    .sp
1018    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1019  .sp  .sp
1020  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
1021  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
1022  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
1023  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1024    they are the options that will be in force when matching starts. For example,
1025    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1026    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1027  .P  .P
1028  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
1029  alternatives begin with one of the following:  alternatives begin with one of the following:
# Line 760  it is used to pass back information abou Line 1078  it is used to pass back information abou
1078  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1079  .  .
1080  .  .
1081  .SH "MATCHING A PATTERN"  .SH "REFERENCE COUNTS"
1082    .rs
1083    .sp
1084    .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1085    .PP
1086    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1087    data block that contains a compiled pattern. It is provided for the benefit of
1088    applications that operate in an object-oriented manner, where different parts
1089    of the application may be using the same compiled pattern, but you want to free
1090    the block when they are all done.
1091    .P
1092    When a pattern is compiled, the reference count field is initialized to zero.
1093    It is changed only by calling this function, whose action is to add the
1094    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1095    function is the new value. However, the value of the count is constrained to
1096    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1097    it is forced to the appropriate limit value.
1098    .P
1099    Except when it is zero, the reference count is not correctly preserved if a
1100    pattern is compiled on one host and then transferred to a host whose byte-order
1101    is different. (This seems a highly unlikely scenario.)
1102    .
1103    .
1104    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1105  .rs  .rs
1106  .sp  .sp
1107  .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,"
# Line 772  string (see PCRE_INFO_FIRSTBYTE above). Line 1113  string (see PCRE_INFO_FIRSTBYTE above).
1113  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
1114  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1115  pattern has been studied, the result of the study should be passed in the  pattern has been studied, the result of the study should be passed in the
1116  \fIextra\fP argument.  \fIextra\fP argument. This function is the main matching facility of the
1117    library, and it operates in a Perl-like manner. For specialist use there is
1118    also an alternative matching function, which is described
1119    .\" HTML <a href="#dfamatch">
1120    .\" </a>
1121    below
1122    .\"
1123    in the section about the \fBpcre_dfa_exec()\fP function.
1124  .P  .P
1125  In most applications, the pattern will have been compiled (and optionally  In most applications, the pattern will have been compiled (and optionally
1126  studied) in the same process that calls \fBpcre_exec()\fP. However, it is  studied) in the same process that calls \fBpcre_exec()\fP. However, it is
# Line 796  Here is an example of a simple call to \ Line 1144  Here is an example of a simple call to \
1144      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1145      0,              /* default options */      0,              /* default options */
1146      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1147      30);            /* number of elements in the vector (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1148  .  .
1149  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1150  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
# Line 805  Here is an example of a simple call to \ Line 1153  Here is an example of a simple call to \
1153  If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP  If the \fIextra\fP argument is not NULL, it must point to a \fBpcre_extra\fP
1154  data block. The \fBpcre_study()\fP function returns such a block (when it  data block. The \fBpcre_study()\fP function returns such a block (when it
1155  doesn't return NULL), but you can also create one for yourself, and pass  doesn't return NULL), but you can also create one for yourself, and pass
1156  additional information in it. The fields in a \fBpcre_extra\fP block are as  additional information in it. The \fBpcre_extra\fP block contains the following
1157  follows:  fields (not necessarily in this order):
1158  .sp  .sp
1159    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1160    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1161    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1162      unsigned long int \fImatch_limit_recursion\fP;
1163    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1164    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1165  .sp  .sp
# Line 819  are set. The flag bits are: Line 1168  are set. The flag bits are:
1168  .sp  .sp
1169    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1170    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1171      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1172    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1173    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1174  .sp  .sp
# Line 833  but which have a very large number of po Line 1183  but which have a very large number of po
1183  classic example is the use of nested unlimited repeats.  classic example is the use of nested unlimited repeats.
1184  .P  .P
1185  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1186  (sometimes recursively). The limit is imposed on the number of times this  (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the
1187  function is called during a match, which has the effect of limiting the amount  number of times this function is called during a match, which has the effect of
1188  of recursion and backtracking that can take place. For patterns that are not  limiting the amount of backtracking that can take place. For patterns that are
1189  anchored, the count starts from zero for each position in the subject string.  not anchored, the count restarts from zero for each position in the subject
1190    string.
1191  .P  .P
1192  The default limit for the library can be set when PCRE is built; the default  The default value for the limit can be set when PCRE is built; the default
1193  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
1194  reduce the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP block  override the default by suppling \fBpcre_exec()\fP with a \fBpcre_extra\fP
1195  in which \fImatch_limit\fP is set to a smaller value, and  block in which \fImatch_limit\fP is set, and PCRE_EXTRA_MATCH_LIMIT is set in
1196  PCRE_EXTRA_MATCH_LIMIT is set in the \fIflags\fP field. If the limit is  the \fIflags\fP field. If the limit is exceeded, \fBpcre_exec()\fP returns
1197  exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1198    .P
1199    The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1200    instead of limiting the total number of times that \fBmatch()\fP is called, it
1201    limits the depth of recursion. The recursion depth is a smaller number than the
1202    total number of calls, because not all calls to \fBmatch()\fP are recursive.
1203    This limit is of use only if it is set smaller than \fImatch_limit\fP.
1204    .P
1205    Limiting the recursion depth limits the amount of stack that can be used, or,
1206    when PCRE has been compiled to use memory on the heap instead of the stack, the
1207    amount of heap memory that can be used.
1208    .P
1209    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1210    built; the default default is the same value as the default for
1211    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1212    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1213    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1214    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1215  .P  .P
1216  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1217  which is described in the  which is described in the
# Line 866  called. See the Line 1234  called. See the
1234  .\"  .\"
1235  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1236  .  .
1237    .\" HTML <a name="execoptions"></a>
1238  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
1239  .rs  .rs
1240  .sp  .sp
1241  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
1242  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,
1243  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_START_OPTIMIZE,
1244    PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1245  .sp  .sp
1246    PCRE_ANCHORED    PCRE_ANCHORED
1247  .sp  .sp
# Line 880  matching position. If a pattern was comp Line 1250  matching position. If a pattern was comp
1250  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
1251  matching time.  matching time.
1252  .sp  .sp
1253      PCRE_BSR_ANYCRLF
1254      PCRE_BSR_UNICODE
1255    .sp
1256    These options (which are mutually exclusive) control what the \eR escape
1257    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1258    match any Unicode newline sequence. These options override the choice that was
1259    made or defaulted when the pattern was compiled.
1260    .sp
1261      PCRE_NEWLINE_CR
1262      PCRE_NEWLINE_LF
1263      PCRE_NEWLINE_CRLF
1264      PCRE_NEWLINE_ANYCRLF
1265      PCRE_NEWLINE_ANY
1266    .sp
1267    These options override the newline definition that was chosen or defaulted when
1268    the pattern was compiled. For details, see the description of
1269    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1270    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1271    the way the match position is advanced after a match failure for an unanchored
1272    pattern.
1273    .P
1274    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1275    match attempt for an unanchored pattern fails when the current position is at a
1276    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1277    characters, the match position is advanced by two characters instead of one, in
1278    other words, to after the CRLF.
1279    .P
1280    The above rule is a compromise that makes the most common cases work as
1281    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1282    set), it does not match the string "\er\enA" because, after failing at the
1283    start, it skips both the CR and the LF before retrying. However, the pattern
1284    [\er\en]A does match that string, because it contains an explicit CR or LF
1285    reference, and so advances only by one character after the first failure.
1286    .P
1287    An explicit match for CR of LF is either a literal appearance of one of those
1288    characters, or one of the \er or \en escape sequences. Implicit matches such as
1289    [^X] do not count, nor does \es (which includes CR and LF in the characters
1290    that it matches).
1291    .P
1292    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1293    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1294    .sp
1295    PCRE_NOTBOL    PCRE_NOTBOL
1296  .sp  .sp
1297  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 916  PCRE_NOTEMPTY and PCRE_ANCHORED, and the Line 1328  PCRE_NOTEMPTY and PCRE_ANCHORED, and the
1328  starting offset (see below) and trying an ordinary match again. There is some  starting offset (see below) and trying an ordinary match again. There is some
1329  code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.  code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.
1330  .sp  .sp
1331      PCRE_NO_START_OPTIMIZE
1332    .sp
1333    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1334    a match, in order to speed up the process. For example, if it is known that a
1335    match must start with a specific character, it searches the subject for that
1336    character, and fails immediately if it cannot find it, without actually running
1337    the main matching function. When callouts are in use, these optimizations can
1338    cause them to be skipped. This option disables the "start-up" optimizations,
1339    causing performance to suffer, but ensuring that the callouts do occur.
1340    .sp
1341    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1342  .sp  .sp
1343  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
1344  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1345  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
1346  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
1347  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  strings in the
1348  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HTML <a href="pcre.html#utf8strings">
1349    .\" </a>
1350    section on UTF-8 support
1351    .\"
1352    in the main
1353    .\" HREF
1354    \fBpcre\fP
1355    .\"
1356    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1357    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1358    PCRE_ERROR_BADUTF8_OFFSET is returned.
1359  .P  .P
1360  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
1361  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
# Line 953  documentation. Line 1385  documentation.
1385  .rs  .rs
1386  .sp  .sp
1387  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
1388  \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
1389  \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a  in \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of
1390  UTF-8 character. Unlike the pattern string, the subject may contain binary zero  a UTF-8 character. Unlike the pattern string, the subject may contain binary
1391  bytes. When the starting offset is zero, the search for a match starts at the  zero bytes. When the starting offset is zero, the search for a match starts at
1392  beginning of the subject, and this is by far the most common case.  the beginning of the subject, and this is by far the most common case.
1393  .P  .P
1394  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
1395  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 991  pattern. Following the usage in Jeffrey Line 1423  pattern. Following the usage in Jeffrey
1423  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
1424  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1425  .P  .P
1426  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
1427  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
1428  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
1429  this argument is NOT the size of \fIovector\fP in bytes.  argument is NOT the size of \fIovector\fP in bytes.
1430  .P  .P
1431  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,
1432  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
1433  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1434  and is not available for passing back information. The length passed in  and is not available for passing back information. The number passed in
1435  \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
1436  rounded down.  rounded down.
1437  .P  .P
1438  When a match is successful, information about captured substrings is returned  When a match is successful, information about captured substrings is returned
1439  in pairs of integers, starting at the beginning of \fIovector\fP, and  in pairs of integers, starting at the beginning of \fIovector\fP, and
1440  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
1441  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
1442  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
1443  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
1444  subject string matched by the entire pattern. The next pair is used for the  mode. They are not character counts.
1445  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  .P
1446  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
1447  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
1448  just the first pair of offsets has been set.  used for the first capturing subpattern, and so on. The value returned by
1449  .P  \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1450  Some convenience functions are provided for extracting the captured substrings  For example, if two substrings have been captured, the returned value is 3. If
1451  as separate strings. These are described in the following section.  there are no capturing subpatterns, the return value from a successful match is
1452  .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.  
1453  .P  .P
1454  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
1455  string that it matched that is returned.  string that it matched that is returned.
1456  .P  .P
1457  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
1458  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
1459  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,
1460  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
1461  \fIovecsize\fP as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1462  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
1463  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
1464  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1465  .P  .P
1466  Note that \fBpcre_info()\fP can be used to find out how many capturing  The \fBpcre_info()\fP function can be used to find out how many capturing
1467  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1468  \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
1469  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.
1470    .P
1471    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1472    the subject when subpattern \fIn\fP has not been used at all. For example, if
1473    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1474    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1475    happens, both values in the offset pairs corresponding to unused subpatterns
1476    are set to -1.
1477    .P
1478    Offset values that correspond to unused subpatterns at the end of the
1479    expression are also set to -1. For example, if the string "abc" is matched
1480    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1481    return from the function is 2, because the highest used capturing subpattern
1482    number is 1. However, you can refer to the offsets for the second and third
1483    capturing subpatterns if you wish (assuming the vector is large enough, of
1484    course).
1485    .P
1486    Some convenience functions are provided for extracting the captured substrings
1487    as separate strings. These are described below.
1488  .  .
1489  .SS "Return values from \fBpcre_exec()\fP"  .\" HTML <a name="errorlist"></a>
1490    .SS "Error return values from \fBpcre_exec()\fP"
1491  .rs  .rs
1492  .sp  .sp
1493  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 1068  compiled in an environment of one endian Line 1514  compiled in an environment of one endian
1514  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
1515  not present.  not present.
1516  .sp  .sp
1517    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1518  .sp  .sp
1519  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1520  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 1090  below). It is never returned by \fBpcre_ Line 1536  below). It is never returned by \fBpcre_
1536  .sp  .sp
1537    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1538  .sp  .sp
1539  The recursion and backtracking limit, as specified by the \fImatch_limit\fP  The backtracking limit, as specified by the \fImatch_limit\fP field in a
1540  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1541  description above.  above.
1542  .sp  .sp
1543    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1544  .sp  .sp
# Line 1112  A string that contains an invalid UTF-8 Line 1558  A string that contains an invalid UTF-8
1558  The UTF-8 byte sequence that was passed as a subject was valid, but the value  The UTF-8 byte sequence that was passed as a subject was valid, but the value
1559  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1560  .sp  .sp
1561    PCRE_ERROR_PARTIAL (-12)    PCRE_ERROR_PARTIAL        (-12)
1562  .sp  .sp
1563  The subject string did not match, but it did match partially. See the  The subject string did not match, but it did match partially. See the
1564  .\" HREF  .\" HREF
# Line 1120  The subject string did not match, but it Line 1566  The subject string did not match, but it
1566  .\"  .\"
1567  documentation for details of partial matching.  documentation for details of partial matching.
1568  .sp  .sp
1569    PCRE_ERROR_BAD_PARTIAL (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1570  .sp  .sp
1571  The PCRE_PARTIAL option was used with a compiled pattern containing items that  The PCRE_PARTIAL option was used with a compiled pattern containing items that
1572  are not supported for partial matching. See the  are not supported for partial matching. See the
# Line 1129  are not supported for partial matching. Line 1575  are not supported for partial matching.
1575  .\"  .\"
1576  documentation for details of partial matching.  documentation for details of partial matching.
1577  .sp  .sp
1578    PCRE_ERROR_INTERNAL (-14)    PCRE_ERROR_INTERNAL       (-14)
1579  .sp  .sp
1580  An unexpected internal error has occurred. This error could be caused by a bug  An unexpected internal error has occurred. This error could be caused by a bug
1581  in PCRE or by overwriting of the compiled pattern.  in PCRE or by overwriting of the compiled pattern.
1582  .sp  .sp
1583    PCRE_ERROR_BADCOUNT (-15)    PCRE_ERROR_BADCOUNT       (-15)
1584  .sp  .sp
1585  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.
1586    .sp
1587      PCRE_ERROR_RECURSIONLIMIT (-21)
1588    .sp
1589    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1590    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1591    description above.
1592    .sp
1593      PCRE_ERROR_BADNEWLINE     (-23)
1594    .sp
1595    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1596    .P
1597    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1598  .  .
1599  .  .
1600  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1148  This error is given if the value of the Line 1606  This error is given if the value of the
1606  .ti +5n  .ti +5n
1607  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
1608  .PP  .PP
 .br  
1609  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1610  .ti +5n  .ti +5n
1611  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1612  .ti +5n  .ti +5n
1613  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
1614  .PP  .PP
 .br  
1615  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
1616  .ti +5n  .ti +5n
1617  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
# Line 1166  Captured substrings can be accessed dire Line 1622  Captured substrings can be accessed dire
1622  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1623  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1624  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1625  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1626  has a further zero added on the end, but the result is not, of course,  .P
1627  a C string.  A substring that contains a binary zero is correctly extracted and has a
1628    further zero added on the end, but the result is not, of course, a C string.
1629    However, you can process such a string by referring to the length that is
1630    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1631    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1632    for handling strings containing binary zeros, because the end of the final
1633    string is not independently indicated.
1634  .P  .P
1635  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:
1636  \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 1188  the string is placed in \fIbuffer\fP, wh Line 1650  the string is placed in \fIbuffer\fP, wh
1650  \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
1651  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1652  \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
1653  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1654  .sp  .sp
1655    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1656  .sp  .sp
# Line 1204  and builds a list of pointers to them. A Line 1666  and builds a list of pointers to them. A
1666  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
1667  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
1668  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
1669  function is zero if all went well, or  function is zero if all went well, or the error code
1670  .sp  .sp
1671    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1672  .sp  .sp
# Line 1223  a previous call of \fBpcre_get_substring Line 1685  a previous call of \fBpcre_get_substring
1685  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1686  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
1687  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
1688  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1689  \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
1690  provided.  provided.
1691  .  .
# Line 1235  provided. Line 1697  provided.
1697  .ti +5n  .ti +5n
1698  .B const char *\fIname\fP);  .B const char *\fIname\fP);
1699  .PP  .PP
 .br  
1700  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1701  .ti +5n  .ti +5n
1702  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1244  provided. Line 1705  provided.
1705  .ti +5n  .ti +5n
1706  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1707  .PP  .PP
 .br  
1708  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1709  .ti +5n  .ti +5n
1710  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1258  For example, for this pattern Line 1718  For example, for this pattern
1718  .sp  .sp
1719    (a+)b(?<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
1720  .sp  .sp
1721  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
1722  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
1723  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
1724    pattern, and the second is the name. The yield of the function is the
1725  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
1726  that name.  that name.
1727  .P  .P
# Line 1268  Given the number, you can extract the su Line 1729  Given the number, you can extract the su
1729  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
1730  two functions that do the whole job.  two functions that do the whole job.
1731  .P  .P
1732  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
1733  \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
1734  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
1735  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
1736  .P  .P
# Line 1279  pattern. This is needed in order to gain Line 1740  pattern. This is needed in order to gain
1740  translation table.  translation table.
1741  .P  .P
1742  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1743  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1744  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1745    the behaviour may not be what you want (see the next section).
1746    .P
1747    \fBWarning:\fP If the pattern uses the "(?|" feature to set up multiple
1748    subpatterns with the same number, you cannot use names to distinguish them,
1749    because names are not included in the compiled code. The matching process uses
1750    only numbers.
1751    .
1752    .SH "DUPLICATE SUBPATTERN NAMES"
1753    .rs
1754    .sp
1755    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1756    .ti +5n
1757    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1758    .PP
1759    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1760    are not required to be unique. Normally, patterns with duplicate names are such
1761    that in any one match, only one of the named subpatterns participates. An
1762    example is shown in the
1763    .\" HREF
1764    \fBpcrepattern\fP
1765    .\"
1766    documentation.
1767  .P  .P
1768  .in 0  When duplicates are present, \fBpcre_copy_named_substring()\fP and
1769  Last updated: 09 September 2004  \fBpcre_get_named_substring()\fP return the first substring corresponding to
1770  .br  the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1771  Copyright (c) 1997-2004 University of Cambridge.  returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1772    returns one of the numbers that are associated with the name, but it is not
1773    defined which it is.
1774    .P
1775    If you want to get full details of all captured substrings for a given name,
1776    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1777    argument is the compiled pattern, and the second is the name. The third and
1778    fourth are pointers to variables which are updated by the function. After it
1779    has run, they point to the first and last entries in the name-to-number table
1780    for the given name. The function itself returns the length of each entry, or
1781    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1782    described above in the section entitled \fIInformation about a pattern\fP.
1783    Given all the relevant entries for the name, you can extract each of their
1784    numbers, and hence the captured data, if any.
1785    .
1786    .
1787    .SH "FINDING ALL POSSIBLE MATCHES"
1788    .rs
1789    .sp
1790    The traditional matching function uses a similar algorithm to Perl, which stops
1791    when it finds the first match, starting at a given point in the subject. If you
1792    want to find all possible matches, or the longest possible match, consider
1793    using the alternative matching function (see below) instead. If you cannot use
1794    the alternative function, but still need to find all possible matches, you
1795    can kludge it up by making use of the callout facility, which is described in
1796    the
1797    .\" HREF
1798    \fBpcrecallout\fP
1799    .\"
1800    documentation.
1801    .P
1802    What you have to do is to insert a callout right at the end of the pattern.
1803    When your callout function is called, extract and save the current matched
1804    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1805    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1806    will yield PCRE_ERROR_NOMATCH.
1807    .
1808    .
1809    .\" HTML <a name="dfamatch"></a>
1810    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1811    .rs
1812    .sp
1813    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1814    .ti +5n
1815    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1816    .ti +5n
1817    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1818    .ti +5n
1819    .B int *\fIworkspace\fP, int \fIwscount\fP);
1820    .P
1821    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1822    a compiled pattern, using a matching algorithm that scans the subject string
1823    just once, and does not backtrack. This has different characteristics to the
1824    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1825    patterns are not supported. Nevertheless, there are times when this kind of
1826    matching can be useful. For a discussion of the two matching algorithms, see
1827    the
1828    .\" HREF
1829    \fBpcrematching\fP
1830    .\"
1831    documentation.
1832    .P
1833    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1834    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1835    different way, and this is described below. The other common arguments are used
1836    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1837    here.
1838    .P
1839    The two additional arguments provide workspace for the function. The workspace
1840    vector should contain at least 20 elements. It is used for keeping track of
1841    multiple paths through the pattern tree. More workspace will be needed for
1842    patterns and subjects where there are a lot of potential matches.
1843    .P
1844    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1845    .sp
1846      int rc;
1847      int ovector[10];
1848      int wspace[20];
1849      rc = pcre_dfa_exec(
1850        re,             /* result of pcre_compile() */
1851        NULL,           /* we didn't study the pattern */
1852        "some string",  /* the subject string */
1853        11,             /* the length of the subject string */
1854        0,              /* start at offset 0 in the subject */
1855        0,              /* default options */
1856        ovector,        /* vector of integers for substring information */
1857        10,             /* number of elements (NOT size in bytes) */
1858        wspace,         /* working space vector */
1859        20);            /* number of elements (NOT size in bytes) */
1860    .
1861    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1862    .rs
1863    .sp
1864    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1865    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1866    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1867    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1868    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1869    .sp
1870      PCRE_PARTIAL
1871    .sp
1872    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1873    details are slightly different. When PCRE_PARTIAL is set for
1874    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1875    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1876    complete matches, but there is still at least one matching possibility. The
1877    portion of the string that provided the partial match is set as the first
1878    matching string.
1879    .sp
1880      PCRE_DFA_SHORTEST
1881    .sp
1882    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1883    soon as it has found one match. Because of the way the alternative algorithm
1884    works, this is necessarily the shortest possible match at the first possible
1885    matching point in the subject string.
1886    .sp
1887      PCRE_DFA_RESTART
1888    .sp
1889    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1890    a partial match, it is possible to call it again, with additional subject
1891    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1892    option requests this action; when it is set, the \fIworkspace\fP and
1893    \fIwscount\fP options must reference the same vector as before because data
1894    about the match so far is left in them after a partial match. There is more
1895    discussion of this facility in the
1896    .\" HREF
1897    \fBpcrepartial\fP
1898    .\"
1899    documentation.
1900    .
1901    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1902    .rs
1903    .sp
1904    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1905    substring in the subject. Note, however, that all the matches from one run of
1906    the function start at the same point in the subject. The shorter matches are
1907    all initial substrings of the longer matches. For example, if the pattern
1908    .sp
1909      <.*>
1910    .sp
1911    is matched against the string
1912    .sp
1913      This is <something> <something else> <something further> no more
1914    .sp
1915    the three matched strings are
1916    .sp
1917      <something>
1918      <something> <something else>
1919      <something> <something else> <something further>
1920    .sp
1921    On success, the yield of the function is a number greater than zero, which is
1922    the number of matched substrings. The substrings themselves are returned in
1923    \fIovector\fP. Each string uses two elements; the first is the offset to the
1924    start, and the second is the offset to the end. In fact, all the strings have
1925    the same start offset. (Space could have been saved by giving this only once,
1926    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1927    returns data, even though the meaning of the strings is different.)
1928    .P
1929    The strings are returned in reverse order of length; that is, the longest
1930    matching string is given first. If there were too many matches to fit into
1931    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1932    the longest matches.
1933    .
1934    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1935    .rs
1936    .sp
1937    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1938    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1939    described
1940    .\" HTML <a href="#errorlist">
1941    .\" </a>
1942    above.
1943    .\"
1944    There are in addition the following errors that are specific to
1945    \fBpcre_dfa_exec()\fP:
1946    .sp
1947      PCRE_ERROR_DFA_UITEM      (-16)
1948    .sp
1949    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1950    that it does not support, for instance, the use of \eC or a back reference.
1951    .sp
1952      PCRE_ERROR_DFA_UCOND      (-17)
1953    .sp
1954    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1955    uses a back reference for the condition, or a test for recursion in a specific
1956    group. These are not supported.
1957    .sp
1958      PCRE_ERROR_DFA_UMLIMIT    (-18)
1959    .sp
1960    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1961    block that contains a setting of the \fImatch_limit\fP field. This is not
1962    supported (it is meaningless).
1963    .sp
1964      PCRE_ERROR_DFA_WSSIZE     (-19)
1965    .sp
1966    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1967    \fIworkspace\fP vector.
1968    .sp
1969      PCRE_ERROR_DFA_RECURSE    (-20)
1970    .sp
1971    When a recursive subpattern is processed, the matching function calls itself
1972    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1973    error is given if the output vector is not large enough. This should be
1974    extremely rare, as a vector of size 1000 is used.
1975    .
1976    .
1977    .SH "SEE ALSO"
1978    .rs
1979    .sp
1980    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1981    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1982    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1983    .
1984    .
1985    .SH AUTHOR
1986    .rs
1987    .sp
1988    .nf
1989    Philip Hazel
1990    University Computing Service
1991    Cambridge CB2 3QH, England.
1992    .fi
1993    .
1994    .
1995    .SH REVISION
1996    .rs
1997    .sp
1998    .nf
1999    Last updated: 11 April 2009
2000    Copyright (c) 1997-2009 University of Cambridge.
2001    .fi

Legend:
Removed from v.75  
changed lines
  Added in v.412

  ViewVC Help
Powered by ViewVC 1.1.5