/[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 77 by nigel, Sat Feb 24 21:40:45 2007 UTC revision 566 by ph10, Wed Nov 3 18:32:55 2010 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
 .br  
16  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17  .ti +5n  .ti +5n
18  .B int *\fIerrorcodeptr\fP,  .B int *\fIerrorcodeptr\fP,
# Line 23  PCRE - Perl-compatible regular expressio Line 21  PCRE - Perl-compatible regular expressio
21  .ti +5n  .ti +5n
22  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
23  .PP  .PP
 .br  
24  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25  .ti +5n  .ti +5n
26  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
27  .PP  .PP
 .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
 .br  
34  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35  .ti +5n  .ti +5n
36  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
# Line 44  PCRE - Perl-compatible regular expressio Line 39  PCRE - Perl-compatible regular expressio
39  .ti +5n  .ti +5n
40  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
41  .PP  .PP
 .br  
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 53  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 69  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
 .br  
95  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
96  .PP  .PP
 .br  
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  PCRE has its own native API, which is described in this document. There are
116  also a set of wrapper functions that correspond to the POSIX regular expression  also some wrapper functions that correspond to the POSIX regular expression
117  API. These are described in the  API. These are described in the
118  .\" HREF  .\" HREF
119  \fBpcreposix\fP  \fBpcreposix\fP
# Line 152  an application that uses PCRE. The heade Line 132  an application that uses PCRE. The heade
132  and PCRE_MINOR to contain the major and minor release numbers for the library.  and PCRE_MINOR to contain the major and minor release numbers for the library.
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    In a Windows environment, if you want to statically link an application program
136    against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
137    including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
138    \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
139    \fB__declspec(dllimport)\fP, with unwanted results.
140    .P
141  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
142  and \fBpcre_exec()\fP are used for compiling and matching regular expressions  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
143  in a Perl-compatible manner. A sample program that demonstrates the simplest  in a Perl-compatible manner. A sample program that demonstrates the simplest
144  way of using them is provided in the file called \fIpcredemo.c\fP in the source  way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
145  distribution. The  source distribution. A listing of this program is given in the
146    .\" HREF
147    \fBpcredemo\fP
148    .\"
149    documentation, and the
150  .\" HREF  .\" HREF
151  \fBpcresample\fP  \fBpcresample\fP
152  .\"  .\"
153  documentation describes how to run it.  documentation describes how to compile and run it.
154  .P  .P
155  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
156  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
157  matching. This allows it to find all possible matches (at a given point in the  matching. The alternative algorithm finds all possible matches (at a given
158  subject), not just one. However, this algorithm does not return captured  point in the subject), and scans the subject just once (unless there are
159    lookbehind assertions). However, this algorithm does not return captured
160  substrings. A description of the two matching algorithms and their advantages  substrings. A description of the two matching algorithms and their advantages
161  and disadvantages is given in the  and disadvantages is given in the
162  .\" HREF  .\" HREF
# Line 183  matched by \fBpcre_exec()\fP. They are: Line 174  matched by \fBpcre_exec()\fP. They are:
174    \fBpcre_get_named_substring()\fP    \fBpcre_get_named_substring()\fP
175    \fBpcre_get_substring_list()\fP    \fBpcre_get_substring_list()\fP
176    \fBpcre_get_stringnumber()\fP    \fBpcre_get_stringnumber()\fP
177      \fBpcre_get_stringtable_entries()\fP
178  .sp  .sp
179  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
180  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
# Line 212  should be done before calling any PCRE f Line 204  should be done before calling any PCRE f
204  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
205  indirections to memory management functions. These special functions are used  indirections to memory management functions. These special functions are used
206  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
207  recursive function calls, when running the \fBpcre_exec()\fP function. This is  recursive function calls, when running the \fBpcre_exec()\fP function. See the
208  a non-standard way of building PCRE, for use in environments that have limited  .\" HREF
209  stacks. Because of the greater use of memory management, it runs more slowly.  \fBpcrebuild\fP
210  Separate functions are provided so that special-purpose external code can be  .\"
211  used for this case. When used, these functions are always called in a  documentation for details of how to do this. It is a non-standard way of
212  stack-like manner (last obtained, first freed), and always for memory blocks of  building PCRE, for use in environments that have limited stacks. Because of the
213  the same size.  greater use of memory management, it runs more slowly. Separate functions are
214    provided so that special-purpose external code can be used for this case. When
215    used, these functions are always called in a stack-like manner (last obtained,
216    first freed), and always for memory blocks of the same size. There is a
217    discussion about PCRE's stack usage in the
218    .\" HREF
219    \fBpcrestack\fP
220    .\"
221    documentation.
222  .P  .P
223  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
224  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
# Line 229  points during a matching operation. Deta Line 229  points during a matching operation. Deta
229  documentation.  documentation.
230  .  .
231  .  .
232    .\" HTML <a name="newlines"></a>
233    .SH NEWLINES
234    .rs
235    .sp
236    PCRE supports five different conventions for indicating line breaks in
237    strings: a single CR (carriage return) character, a single LF (linefeed)
238    character, the two-character sequence CRLF, any of the three preceding, or any
239    Unicode newline sequence. The Unicode newline sequences are the three just
240    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
241    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
242    (paragraph separator, U+2029).
243    .P
244    Each of the first three conventions is used by at least one operating system as
245    its standard newline sequence. When PCRE is built, a default can be specified.
246    The default default is LF, which is the Unix standard. When PCRE is run, the
247    default can be overridden, either when a pattern is compiled, or when it is
248    matched.
249    .P
250    At compile time, the newline convention can be specified by the \fIoptions\fP
251    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
252    start of the pattern itself; this overrides any other settings. See the
253    .\" HREF
254    \fBpcrepattern\fP
255    .\"
256    page for details of the special character sequences.
257    .P
258    In the PCRE documentation the word "newline" is used to mean "the character or
259    pair of characters that indicate a line break". The choice of newline
260    convention affects the handling of the dot, circumflex, and dollar
261    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
262    recognized line ending sequence, the match position advancement for a
263    non-anchored pattern. There is more detail about this in the
264    .\" HTML <a href="#execoptions">
265    .\" </a>
266    section on \fBpcre_exec()\fP options
267    .\"
268    below.
269    .P
270    The choice of newline convention does not affect the interpretation of
271    the \en or \er escape sequences, nor does it affect what \eR matches, which is
272    controlled in a similar way, but by separate options.
273    .
274    .
275  .SH MULTITHREADING  .SH MULTITHREADING
276  .rs  .rs
277  .sp  .sp
# Line 250  which it was compiled. Details are given Line 293  which it was compiled. Details are given
293  .\" HREF  .\" HREF
294  \fBpcreprecompile\fP  \fBpcreprecompile\fP
295  .\"  .\"
296  documentation.  documentation. However, compiling a regular expression with one version of PCRE
297    for use with a different version is not guaranteed to work and may cause
298    crashes.
299  .  .
300  .  .
301  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 281  properties is available; otherwise it is Line 326  properties is available; otherwise it is
326  .sp  .sp
327    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
328  .sp  .sp
329  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
330  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
331  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.
332    Though they are derived from ASCII, the same values are returned in EBCDIC
333    environments. The default should normally correspond to the standard sequence
334    for your operating system.
335    .sp
336      PCRE_CONFIG_BSR
337    .sp
338    The output is an integer whose value indicates what character sequences the \eR
339    escape sequence matches by default. A value of 0 means that \eR matches any
340    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
341    or CRLF. The default can be overridden when a pattern is compiled or matched.
342  .sp  .sp
343    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
344  .sp  .sp
# Line 305  documentation. Line 360  documentation.
360  .sp  .sp
361    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
362  .sp  .sp
363  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
364  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
365  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
366  .sp  .sp
367      PCRE_CONFIG_MATCH_LIMIT_RECURSION
368    .sp
369    The output is a long integer that gives the default limit for the depth of
370    recursion when calling the internal matching function in a \fBpcre_exec()\fP
371    execution. Further details are given with \fBpcre_exec()\fP below.
372    .sp
373    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
374  .sp  .sp
375  The output is an integer that is set to one if internal recursion when running  The output is an integer that is set to one if internal recursion when running
# Line 340  avoiding the use of the stack. Line 401  avoiding the use of the stack.
401  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
402  called to compile a pattern into an internal form. The only difference between  called to compile a pattern into an internal form. The only difference between
403  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
404  \fIerrorcodeptr\fP, via which a numerical error code can be returned.  \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
405    too much repetition, we refer just to \fBpcre_compile()\fP below, but the
406    information applies equally to \fBpcre_compile2()\fP.
407  .P  .P
408  The pattern is a C string terminated by a binary zero, and is passed in the  The pattern is a C string terminated by a binary zero, and is passed in the
409  \fIpattern\fP argument. A pointer to a single block of memory that is obtained  \fIpattern\fP argument. A pointer to a single block of memory that is obtained
410  via \fBpcre_malloc\fP is returned. This contains the compiled code and related  via \fBpcre_malloc\fP is returned. This contains the compiled code and related
411  data. The \fBpcre\fP type is defined for the returned block; this is a typedef  data. The \fBpcre\fP type is defined for the returned block; this is a typedef
412  for a structure whose contents are not externally defined. It is up to the  for a structure whose contents are not externally defined. It is up to the
413  caller to free the memory when it is no longer required.  caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
414  .P  .P
415  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
416  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
417  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
418  argument, which is an address (see below).  argument, which is an address (see below).
419  .P  .P
420  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
421  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
422  options are described below. Some of them, in particular, those that are  options are described below. Some of them (in particular, those that are
423  compatible with Perl, can also be set and unset from within the pattern (see  compatible with Perl, but some others as well) can also be set and unset from
424  the detailed description in the  within the pattern (see the detailed description in the
425  .\" HREF  .\" HREF
426  \fBpcrepattern\fP  \fBpcrepattern\fP
427  .\"  .\"
428  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For those options that can be different in different parts of
429  specifies their initial settings at the start of compilation and execution. The  the pattern, the contents of the \fIoptions\fP argument specifies their
430  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,
431  time.  PCRE_BSR_\fIxxx\fP, and PCRE_NEWLINE_\fIxxx\fP options can be set at the time
432    of matching as well as at compile time.
433  .P  .P
434  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
435  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
436  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
437  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
438  the error was discovered is placed in the variable pointed to by  not try to free it. The byte offset from the start of the pattern to the
439  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.  character that was being processed when the error was discovered is placed in
440    the variable pointed to by \fIerroffset\fP, which must not be NULL. If it is,
441    an immediate error is given. Some errors are not detected until checks are
442    carried out when the whole pattern has been scanned; in this case the offset is
443    set to the end of the pattern.
444  .P  .P
445  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
446  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
# Line 419  facility, see the Line 487  facility, see the
487  .\"  .\"
488  documentation.  documentation.
489  .sp  .sp
490      PCRE_BSR_ANYCRLF
491      PCRE_BSR_UNICODE
492    .sp
493    These options (which are mutually exclusive) control what the \eR escape
494    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
495    match any Unicode newline sequence. The default is specified when PCRE is
496    built. It can be overridden from within the pattern, or by setting an option
497    when a compiled pattern is matched.
498    .sp
499    PCRE_CASELESS    PCRE_CASELESS
500  .sp  .sp
501  If this bit is set, letters in the pattern match both upper and lower case  If this bit is set, letters in the pattern match both upper and lower case
# Line 435  with UTF-8 support. Line 512  with UTF-8 support.
512  .sp  .sp
513  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
514  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
515  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
516  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.
517  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
518  a pattern.  pattern.
519  .sp  .sp
520    PCRE_DOTALL    PCRE_DOTALL
521  .sp  .sp
522  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,
523  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
524  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
525  (?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
526  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
527    the setting of this option.
528    .sp
529      PCRE_DUPNAMES
530    .sp
531    If this bit is set, names used to identify capturing subpatterns need not be
532    unique. This can be helpful for certain types of pattern when it is known that
533    only one instance of the named subpattern can ever be matched. There are more
534    details of named subpatterns below; see also the
535    .\" HREF
536    \fBpcrepattern\fP
537    .\"
538    documentation.
539  .sp  .sp
540    PCRE_EXTENDED    PCRE_EXTENDED
541  .sp  .sp
542  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
543  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
544  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
545  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
546  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
547  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
548  .P  .P
549  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
550  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
# Line 469  that is incompatible with Perl, but it i Line 558  that is incompatible with Perl, but it i
558  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
559  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
560  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
561  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
562  controlled by this option. It can also be set by a (?X) option setting within a  give an error for this, by running it with the -w option.) There are at present
563  pattern.  no other features controlled by this option. It can also be set by a (?X)
564    option setting within a pattern.
565  .sp  .sp
566    PCRE_FIRSTLINE    PCRE_FIRSTLINE
567  .sp  .sp
568  If this option is set, an unanchored pattern is required to match before or at  If this option is set, an unanchored pattern is required to match before or at
569  the first newline character in the subject string, though the matched text may  the first newline in the subject string, though the matched text may continue
570  continue over the newline.  over the newline.
571    .sp
572      PCRE_JAVASCRIPT_COMPAT
573    .sp
574    If this option is set, PCRE's behaviour is changed in some ways so that it is
575    compatible with JavaScript rather than Perl. The changes are as follows:
576    .P
577    (1) A lone closing square bracket in a pattern causes a compile-time error,
578    because this is illegal in JavaScript (by default it is treated as a data
579    character). Thus, the pattern AB]CD becomes illegal when this option is set.
580    .P
581    (2) At run time, a back reference to an unset subpattern group matches an empty
582    string (by default this causes the current matching alternative to fail). A
583    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
584    an "a" in the subject), whereas it fails by default, for Perl compatibility.
585  .sp  .sp
586    PCRE_MULTILINE    PCRE_MULTILINE
587  .sp  .sp
# Line 489  terminating newline (unless PCRE_DOLLAR_ Line 593  terminating newline (unless PCRE_DOLLAR_
593  Perl.  Perl.
594  .P  .P
595  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
596  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
597  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
598  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
599  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
600  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
601  .sp  .sp
602      PCRE_NEWLINE_CR
603      PCRE_NEWLINE_LF
604      PCRE_NEWLINE_CRLF
605      PCRE_NEWLINE_ANYCRLF
606      PCRE_NEWLINE_ANY
607    .sp
608    These options override the default newline definition that was chosen when PCRE
609    was built. Setting the first or the second specifies that a newline is
610    indicated by a single character (CR or LF, respectively). Setting
611    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
612    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
613    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
614    that any Unicode newline sequence should be recognized. The Unicode newline
615    sequences are the three just mentioned, plus the single characters VT (vertical
616    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
617    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
618    recognized only in UTF-8 mode.
619    .P
620    The newline setting in the options word uses three bits that are treated
621    as a number, giving eight possibilities. Currently only six are used (default
622    plus the five values above). This means that if you set more than one newline
623    option, the combination may or may not be sensible. For example,
624    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
625    other combinations may yield unused numbers and cause an error.
626    .P
627    The only time that a line break is specially recognized when compiling a
628    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
629    class is encountered. This indicates a comment that lasts until after the next
630    line break sequence. In other circumstances, line break sequences are treated
631    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
632    as whitespace characters and are therefore ignored.
633    .P
634    The newline option that is set at compile time becomes the default that is used
635    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
636    .sp
637    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
638  .sp  .sp
639  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 503  were followed by ?: but named parenthese Line 642  were followed by ?: but named parenthese
642  they acquire numbers in the usual way). There is no equivalent of this option  they acquire numbers in the usual way). There is no equivalent of this option
643  in Perl.  in Perl.
644  .sp  .sp
645      PCRE_UCP
646    .sp
647    This option changes the way PCRE processes \eb, \ed, \es, \ew, and some of the
648    POSIX character classes. By default, only ASCII characters are recognized, but
649    if PCRE_UCP is set, Unicode properties are used instead to classify characters.
650    More details are given in the section on
651    .\" HTML <a href="pcre.html#genericchartypes">
652    .\" </a>
653    generic character types
654    .\"
655    in the
656    .\" HREF
657    \fBpcrepattern\fP
658    .\"
659    page. If you set PCRE_UCP, matching one of the items it affects takes much
660    longer. The option is available only if PCRE has been compiled with Unicode
661    property support.
662    .sp
663    PCRE_UNGREEDY    PCRE_UNGREEDY
664  .sp  .sp
665  This option inverts the "greediness" of the quantifiers so that they are not  This option inverts the "greediness" of the quantifiers so that they are not
# Line 529  page. Line 686  page.
686    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
687  .sp  .sp
688  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
689  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
690  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcre.html#utf8strings">
691  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
692  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
693  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
694  Note that this option can also be passed to \fBpcre_exec()\fP and  in the main
695  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject  .\" HREF
696  strings.  \fBpcre\fP
697    .\"
698    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
699    returns an error. If you already know that your pattern is valid, and you want
700    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
701    option. When it is set, the effect of passing an invalid UTF-8 string as a
702    pattern is undefined. It may cause your program to crash. Note that this option
703    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
704    the UTF-8 validity checking of subject strings.
705  .  .
706  .  .
707  .SH "COMPILATION ERROR CODES"  .SH "COMPILATION ERROR CODES"
# Line 544  strings. Line 709  strings.
709  .sp  .sp
710  The following table lists the error codes than may be returned by  The following table lists the error codes than may be returned by
711  \fBpcre_compile2()\fP, along with the error messages that may be returned by  \fBpcre_compile2()\fP, along with the error messages that may be returned by
712  both compiling functions.  both compiling functions. As PCRE has developed, some error codes have fallen
713    out of use. To avoid confusion, they have not been re-used.
714  .sp  .sp
715     0  no error     0  no error
716     1  \e at end of pattern     1  \e at end of pattern
# Line 556  both compiling functions. Line 722  both compiling functions.
722     7  invalid escape sequence in character class     7  invalid escape sequence in character class
723     8  range out of order in character class     8  range out of order in character class
724     9  nothing to repeat     9  nothing to repeat
725    10  operand of unlimited repeat could match the empty string    10  [this code is not in use]
726    11  internal error: unexpected repeat    11  internal error: unexpected repeat
727    12  unrecognized character after (?    12  unrecognized character after (? or (?-
728    13  POSIX named classes are supported only within a class    13  POSIX named classes are supported only within a class
729    14  missing )    14  missing )
730    15  reference to non-existent subpattern    15  reference to non-existent subpattern
731    16  erroffset passed as NULL    16  erroffset passed as NULL
732    17  unknown option bit(s) set    17  unknown option bit(s) set
733    18  missing ) after comment    18  missing ) after comment
734    19  parentheses nested too deeply    19  [this code is not in use]
735    20  regular expression too large    20  regular expression is too large
736    21  failed to get memory    21  failed to get memory
737    22  unmatched parentheses    22  unmatched parentheses
738    23  internal error: code overflow    23  internal error: code overflow
739    24  unrecognized character after (?<    24  unrecognized character after (?<
740    25  lookbehind assertion is not fixed length    25  lookbehind assertion is not fixed length
741    26  malformed number after (?(    26  malformed number or name after (?(
742    27  conditional group contains more than two branches    27  conditional group contains more than two branches
743    28  assertion expected after (?(    28  assertion expected after (?(
744    29  (?R or (?digits must be followed by )    29  (?R or (?[+-]digits must be followed by )
745    30  unknown POSIX class name    30  unknown POSIX class name
746    31  POSIX collating elements are not supported    31  POSIX collating elements are not supported
747    32  this version of PCRE is not compiled with PCRE_UTF8 support    32  this version of PCRE is not compiled with PCRE_UTF8 support
748    33  spare error    33  [this code is not in use]
749    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
750    35  invalid condition (?(0)    35  invalid condition (?(0)
751    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
# Line 588  both compiling functions. Line 754  both compiling functions.
754    39  closing ) for (?C expected    39  closing ) for (?C expected
755    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
756    41  unrecognized character after (?P    41  unrecognized character after (?P
757    42  syntax error after (?P    42  syntax error in subpattern name (missing terminator)
758    43  two named groups have the same name    43  two named subpatterns have the same name
759    44  invalid UTF-8 string    44  invalid UTF-8 string
760    45  support for \eP, \ep, and \eX has not been compiled    45  support for \eP, \ep, and \eX has not been compiled
761    46  malformed \eP or \ep sequence    46  malformed \eP or \ep sequence
762    47  unknown property name after \eP or \ep    47  unknown property name after \eP or \ep
763      48  subpattern name is too long (maximum 32 characters)
764      49  too many named subpatterns (maximum 10000)
765      50  [this code is not in use]
766      51  octal value is greater than \e377 (not in UTF-8 mode)
767      52  internal error: overran compiling workspace
768      53  internal error: previously-checked referenced subpattern
769            not found
770      54  DEFINE group contains more than one branch
771      55  repeating a DEFINE group is not allowed
772      56  inconsistent NEWLINE options
773      57  \eg is not followed by a braced, angle-bracketed, or quoted
774            name/number or by a plain number
775      58  a numbered reference must not be zero
776      59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
777      60  (*VERB) not recognized
778      61  number is too big
779      62  subpattern name expected
780      63  digit expected after (?+
781      64  ] is an invalid data character in JavaScript compatibility mode
782      65  different names for subpatterns of the same number are
783            not allowed
784      66  (*MARK) must have an argument
785      67  this version of PCRE is not compiled with PCRE_UCP support
786    .sp
787    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
788    be used if the limits were changed when PCRE was built.
789  .  .
790  .  .
791  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
# Line 612  help speed up matching, \fBpcre_study()\ Line 804  help speed up matching, \fBpcre_study()\
804  results of the study.  results of the study.
805  .P  .P
806  The returned value from \fBpcre_study()\fP can be passed directly to  The returned value from \fBpcre_study()\fP can be passed directly to
807  \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other  \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
808  fields that can be set by the caller before the block is passed; these are  also contains other fields that can be set by the caller before the block is
809  described  passed; these are described
810  .\" HTML <a href="#extradata">  .\" HTML <a href="#extradata">
811  .\" </a>  .\" </a>
812  below  below
813  .\"  .\"
814  in the section on matching a pattern.  in the section on matching a pattern.
815  .P  .P
816  If studying the pattern does not produce any additional information  If studying the pattern does not produce any useful information,
817  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
818  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its  wants to pass any of the other fields to \fBpcre_exec()\fP or
819  own \fBpcre_extra\fP block.  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
820  .P  .P
821  The second argument of \fBpcre_study()\fP contains option bits. At present, no  The second argument of \fBpcre_study()\fP contains option bits. At present, no
822  options are defined, and this argument should always be zero.  options are defined, and this argument should always be zero.
823  .P  .P
824  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
825  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
826  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
827  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
828  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
829    sure that it has run successfully.
830  .P  .P
831  This is a typical call to \fBpcre_study\fP():  This is a typical call to \fBpcre_study\fP():
832  .sp  .sp
# Line 643  This is a typical call to \fBpcre_study\ Line 836  This is a typical call to \fBpcre_study\
836      0,              /* no options exist */      0,              /* no options exist */
837      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
838  .sp  .sp
839  At present, studying a pattern is useful only for non-anchored patterns that do  Studying a pattern does two things: first, a lower bound for the length of
840  not have a single fixed starting character. A bitmap of possible starting  subject string that is needed to match the pattern is computed. This does not
841  bytes is created.  mean that there are any strings of that length that match, but it does
842    guarantee that no shorter strings match. The value is used by
843    \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
844    match strings that are shorter than the lower bound. You can find out the value
845    in a calling program via the \fBpcre_fullinfo()\fP function.
846    .P
847    Studying a pattern is also useful for non-anchored patterns that do not have a
848    single fixed starting character. A bitmap of possible starting bytes is
849    created. This speeds up finding a position in the subject at which to start
850    matching.
851    .P
852    The two optimizations just described can be disabled by setting the
853    PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
854    \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
855    callouts, or make use of (*MARK), and you make use of these in cases where
856    matching fails. See the discussion of PCRE_NO_START_OPTIMIZE
857    .\" HTML <a href="#execoptions">
858    .\" </a>
859    below.
860    .\"
861  .  .
862  .  .
863  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
864  .SH "LOCALE SUPPORT"  .SH "LOCALE SUPPORT"
865  .rs  .rs
866  .sp  .sp
867  PCRE handles caseless matching, and determines whether characters are letters  PCRE handles caseless matching, and determines whether characters are letters,
868  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
869  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
870  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  less than 128. By default, higher-valued codes never match escapes such as \ew
871  can be tested with \ep if PCRE is built with Unicode character property  or \ed, but they can be tested with \ep if PCRE is built with Unicode character
872  support.  property support. Alternatively, the PCRE_UCP option can be set at compile
873  .P  time; this causes \ew and friends to use Unicode property support instead of
874  An internal set of tables is created in the default C locale when PCRE is  built-in tables. The use of locales with Unicode is discouraged. If you are
875  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  handling characters with codes greater than 128, you should either use UTF-8
876  and is sufficient for many applications. An alternative set of tables can,  and Unicode, or use locales, but not try to mix the two.
877  however, be supplied. These may be created in a different locale from the  .P
878  default. As more and more applications change to using Unicode, the need for  PCRE contains an internal set of tables that are used when the final argument
879  this locale support is expected to die away.  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
880    Normally, the internal tables recognize only ASCII characters. However, when
881    PCRE is built, it is possible to cause the internal tables to be rebuilt in the
882    default "C" locale of the local system, which may cause them to be different.
883    .P
884    The internal tables can always be overridden by tables supplied by the
885    application that calls PCRE. These may be created in a different locale from
886    the default. As more and more applications change to using Unicode, the need
887    for this locale support is expected to die away.
888  .P  .P
889  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
890  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 677  the following code could be used: Line 897  the following code could be used:
897    tables = pcre_maketables();    tables = pcre_maketables();
898    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
899  .sp  .sp
900    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
901    are using Windows, the name for the French locale is "french".
902    .P
903  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
904  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
905  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 723  check against passing an arbitrary memor Line 946  check against passing an arbitrary memor
946  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
947  .sp  .sp
948    int rc;    int rc;
949    unsigned long int length;    size_t length;
950    rc = pcre_fullinfo(    rc = pcre_fullinfo(
951      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
952      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
# Line 755  a NULL table pointer. Line 978  a NULL table pointer.
978    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
979  .sp  .sp
980  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
981  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
982  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
983    still recognized for backwards compatibility.)
984  .P  .P
985  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
986  (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  
987  .sp  .sp
988  (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
989  starts with "^", or  starts with "^", or
# Line 779  table indicating a fixed set of bytes fo Line 1002  table indicating a fixed set of bytes fo
1002  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
1003  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
1004  .sp  .sp
1005      PCRE_INFO_HASCRORLF
1006    .sp
1007    Return 1 if the pattern contains any explicit matches for CR or LF characters,
1008    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1009    explicit match is either a literal CR or LF character, or \er or \en.
1010    .sp
1011      PCRE_INFO_JCHANGED
1012    .sp
1013    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1014    0. The fourth argument should point to an \fBint\fP variable. (?J) and
1015    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1016    .sp
1017    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1018  .sp  .sp
1019  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 789  follows something of variable length. Fo Line 1024  follows something of variable length. Fo
1024  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1025  is -1.  is -1.
1026  .sp  .sp
1027      PCRE_INFO_MINLENGTH
1028    .sp
1029    If the pattern was studied and a minimum length for matching subject strings
1030    was computed, its value is returned. Otherwise the returned value is -1. The
1031    value is a number of characters, not bytes (this may be relevant in UTF-8
1032    mode). The fourth argument should point to an \fBint\fP variable. A
1033    non-negative value is a lower bound to the length of any matching string. There
1034    may not be any strings of that length that do actually match, but every string
1035    that does match is at least that long.
1036    .sp
1037    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
1038    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
1039    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
1040  .sp  .sp
1041  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
1042  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
1043  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
1044  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
1045  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
1046  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
1047  \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,
1048  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
1049    values.
1050  .P  .P
1051  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
1052  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 808  entry; both of these return an \fBint\fP Line 1054  entry; both of these return an \fBint\fP
1054  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1055  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
1056  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
1057  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated.
1058  alphabetical order. For example, consider the following pattern (assume  .P
1059  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  The names are in alphabetical order. Duplicate names may appear if (?| is used
1060    to create multiple groups with the same number, as described in the
1061    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1062    .\" </a>
1063    section on duplicate subpattern numbers
1064    .\"
1065    in the
1066    .\" HREF
1067    \fBpcrepattern\fP
1068    .\"
1069    page. Duplicate names for subpatterns with different numbers are permitted only
1070    if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1071    table in the order in which they were found in the pattern. In the absence of
1072    (?| this is the order of increasing number; when (?| is used this is not
1073    necessarily the case because later subpatterns may have lower numbers.
1074    .P
1075    As a simple example of the name/number table, consider the following pattern
1076    (assume PCRE_EXTENDED is set, so white space - including newlines - is
1077    ignored):
1078  .sp  .sp
1079  .\" JOIN  .\" JOIN
1080    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
1081    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
1082  .sp  .sp
1083  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
1084  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 826  bytes shows in hexadecimal, and undefine Line 1090  bytes shows in hexadecimal, and undefine
1090    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
1091  .sp  .sp
1092  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
1093  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
1094  different for each compiled pattern.  different for each compiled pattern.
1095  .sp  .sp
1096      PCRE_INFO_OKPARTIAL
1097    .sp
1098    Return 1 if the pattern can be used for partial matching with
1099    \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1100    \fBint\fP variable. From release 8.00, this always returns 1, because the
1101    restrictions that previously applied to partial matching have been lifted. The
1102    .\" HREF
1103    \fBpcrepartial\fP
1104    .\"
1105    documentation gives details of partial matching.
1106    .sp
1107    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1108  .sp  .sp
1109  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
1110  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
1111  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
1112  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1113    they are the options that will be in force when matching starts. For example,
1114    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1115    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1116  .P  .P
1117  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
1118  alternatives begin with one of the following:  alternatives begin with one of the following:
# Line 861  variable. Line 1139  variable.
1139  Return the size of the data block pointed to by the \fIstudy_data\fP field in  Return the size of the data block pointed to by the \fIstudy_data\fP field in
1140  a \fBpcre_extra\fP block. That is, it is the value that was passed to  a \fBpcre_extra\fP block. That is, it is the value that was passed to
1141  \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data  \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data
1142  created by \fBpcre_study()\fP. The fourth argument should point to a  created by \fBpcre_study()\fP. If \fBpcre_extra\fP is NULL, or there is no
1143    study data, zero is returned. The fourth argument should point to a
1144  \fBsize_t\fP variable.  \fBsize_t\fP variable.
1145  .  .
1146  .  .
# Line 923  is different. (This seems a highly unlik Line 1202  is different. (This seems a highly unlik
1202  .P  .P
1203  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
1204  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1205  pattern has been studied, the result of the study should be passed in the  pattern was studied, the result of the study should be passed in the
1206  \fIextra\fP argument. This function is the main matching facility of the  \fIextra\fP argument. This function is the main matching facility of the
1207  library, and it operates in a Perl-like manner. For specialist use there is  library, and it operates in a Perl-like manner. For specialist use there is
1208  also an alternative matching function, which is described  also an alternative matching function, which is described
# Line 964  Here is an example of a simple call to \ Line 1243  Here is an example of a simple call to \
1243  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
1244  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
1245  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
1246  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
1247  follows:  fields (not necessarily in this order):
1248  .sp  .sp
1249    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1250    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1251    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1252      unsigned long int \fImatch_limit_recursion\fP;
1253    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1254    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1255      unsigned char **\fImark\fP;
1256  .sp  .sp
1257  The \fIflags\fP field is a bitmap that specifies which of the other fields  The \fIflags\fP field is a bitmap that specifies which of the other fields
1258  are set. The flag bits are:  are set. The flag bits are:
1259  .sp  .sp
1260    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1261    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1262      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1263    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1264    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1265      PCRE_EXTRA_MARK
1266  .sp  .sp
1267  Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the  Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the
1268  \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with  \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with
# Line 989  the block by setting the other fields an Line 1272  the block by setting the other fields an
1272  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1273  vast amount of resources when running patterns that are not going to match,  vast amount of resources when running patterns that are not going to match,
1274  but which have a very large number of possibilities in their search trees. The  but which have a very large number of possibilities in their search trees. The
1275  classic example is the use of nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1276  .P  .P
1277  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1278  (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
1279  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
1280  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
1281  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
1282    string.
1283  .P  .P
1284  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
1285  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
1286  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
1287  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
1288  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
1289  exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1290    .P
1291    The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1292    instead of limiting the total number of times that \fBmatch()\fP is called, it
1293    limits the depth of recursion. The recursion depth is a smaller number than the
1294    total number of calls, because not all calls to \fBmatch()\fP are recursive.
1295    This limit is of use only if it is set smaller than \fImatch_limit\fP.
1296    .P
1297    Limiting the recursion depth limits the amount of stack that can be used, or,
1298    when PCRE has been compiled to use memory on the heap instead of the stack, the
1299    amount of heap memory that can be used.
1300    .P
1301    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1302    built; the default default is the same value as the default for
1303    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1304    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1305    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1306    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1307  .P  .P
1308  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,  The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1309  which is described in the  and is described in the
1310  .\" HREF  .\" HREF
1311  \fBpcrecallout\fP  \fBpcrecallout\fP
1312  .\"  .\"
# Line 1024  called. See the Line 1325  called. See the
1325  \fBpcreprecompile\fP  \fBpcreprecompile\fP
1326  .\"  .\"
1327  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1328    .P
1329    If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1330    be set to point to a \fBchar *\fP variable. If the pattern contains any
1331    backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1332    a name to pass back, a pointer to the name string (zero terminated) is placed
1333    in the variable pointed to by the \fImark\fP field. The names are within the
1334    compiled pattern; if you wish to retain such a name you must copy it before
1335    freeing the memory of a compiled pattern. If there is no name to pass back, the
1336    variable pointed to by the \fImark\fP field set to NULL. For details of the
1337    backtracking control verbs, see the section entitled
1338    .\" HTML <a href="pcrepattern#backtrackcontrol">
1339    .\" </a>
1340    "Backtracking control"
1341    .\"
1342    in the
1343    .\" HREF
1344    \fBpcrepattern\fP
1345    .\"
1346    documentation.
1347    .
1348  .  .
1349    .\" HTML <a name="execoptions"></a>
1350  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
1351  .rs  .rs
1352  .sp  .sp
1353  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
1354  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,
1355  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1356    PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1357    PCRE_PARTIAL_HARD.
1358  .sp  .sp
1359    PCRE_ANCHORED    PCRE_ANCHORED
1360  .sp  .sp
# Line 1039  matching position. If a pattern was comp Line 1363  matching position. If a pattern was comp
1363  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
1364  matching time.  matching time.
1365  .sp  .sp
1366      PCRE_BSR_ANYCRLF
1367      PCRE_BSR_UNICODE
1368    .sp
1369    These options (which are mutually exclusive) control what the \eR escape
1370    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1371    match any Unicode newline sequence. These options override the choice that was
1372    made or defaulted when the pattern was compiled.
1373    .sp
1374      PCRE_NEWLINE_CR
1375      PCRE_NEWLINE_LF
1376      PCRE_NEWLINE_CRLF
1377      PCRE_NEWLINE_ANYCRLF
1378      PCRE_NEWLINE_ANY
1379    .sp
1380    These options override the newline definition that was chosen or defaulted when
1381    the pattern was compiled. For details, see the description of
1382    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1383    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1384    the way the match position is advanced after a match failure for an unanchored
1385    pattern.
1386    .P
1387    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1388    match attempt for an unanchored pattern fails when the current position is at a
1389    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1390    characters, the match position is advanced by two characters instead of one, in
1391    other words, to after the CRLF.
1392    .P
1393    The above rule is a compromise that makes the most common cases work as
1394    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1395    set), it does not match the string "\er\enA" because, after failing at the
1396    start, it skips both the CR and the LF before retrying. However, the pattern
1397    [\er\en]A does match that string, because it contains an explicit CR or LF
1398    reference, and so advances only by one character after the first failure.
1399    .P
1400    An explicit match for CR of LF is either a literal appearance of one of those
1401    characters, or one of the \er or \en escape sequences. Implicit matches such as
1402    [^X] do not count, nor does \es (which includes CR and LF in the characters
1403    that it matches).
1404    .P
1405    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1406    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1407    .sp
1408    PCRE_NOTBOL    PCRE_NOTBOL
1409  .sp  .sp
1410  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 1063  match the empty string, the entire match Line 1429  match the empty string, the entire match
1429  .sp  .sp
1430    a?b?    a?b?
1431  .sp  .sp
1432  is applied to a string not beginning with "a" or "b", it matches the empty  is applied to a string not beginning with "a" or "b", it matches an empty
1433  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1434  valid, so PCRE searches further into the string for occurrences of "a" or "b".  valid, so PCRE searches further into the string for occurrences of "a" or "b".
1435  .P  .sp
1436  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case    PCRE_NOTEMPTY_ATSTART
1437  of a pattern match of the empty string within its \fBsplit()\fP function, and  .sp
1438  when using the /g modifier. It is possible to emulate Perl's behaviour after  This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1439  matching a null string by first trying the match again at the same offset with  the start of the subject is permitted. If the pattern is anchored, such a match
1440  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the  can occur only if the pattern contains \eK.
1441  starting offset (see below) and trying an ordinary match again. There is some  .P
1442  code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.  Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1443    does make a special case of a pattern match of the empty string within its
1444    \fBsplit()\fP function, and when using the /g modifier. It is possible to
1445    emulate Perl's behaviour after matching a null string by first trying the match
1446    again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1447    if that fails, by advancing the starting offset (see below) and trying an
1448    ordinary match again. There is some code that demonstrates how to do this in
1449    the
1450    .\" HREF
1451    \fBpcredemo\fP
1452    .\"
1453    sample program. In the most general case, you have to check to see if the
1454    newline convention recognizes CRLF as a newline, and if so, and the current
1455    character is CR followed by LF, advance the starting offset by two characters
1456    instead of one.
1457    .sp
1458      PCRE_NO_START_OPTIMIZE
1459    .sp
1460    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1461    a match, in order to speed up the process. For example, if it is known that an
1462    unanchored match must start with a specific character, it searches the subject
1463    for that character, and fails immediately if it cannot find it, without
1464    actually running the main matching function. This means that a special item
1465    such as (*COMMIT) at the start of a pattern is not considered until after a
1466    suitable starting point for the match has been found. When callouts or (*MARK)
1467    items are in use, these "start-up" optimizations can cause them to be skipped
1468    if the pattern is never actually used. The start-up optimizations are in effect
1469    a pre-scan of the subject that takes place before the pattern is run.
1470    .P
1471    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1472    causing performance to suffer, but ensuring that in cases where the result is
1473    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1474    are considered at every possible starting position in the subject string.
1475    Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1476    Consider the pattern
1477    .sp
1478      (*COMMIT)ABC
1479    .sp
1480    When this is compiled, PCRE records the fact that a match must start with the
1481    character "A". Suppose the subject string is "DEFABC". The start-up
1482    optimization scans along the subject, finds "A" and runs the first match
1483    attempt from there. The (*COMMIT) item means that the pattern must match the
1484    current starting position, which in this case, it does. However, if the same
1485    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1486    subject string does not happen. The first match attempt is run starting from
1487    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1488    the overall result is "no match". If the pattern is studied, more start-up
1489    optimizations may be used. For example, a minimum length for the subject may be
1490    recorded. Consider the pattern
1491    .sp
1492      (*MARK:A)(X|Y)
1493    .sp
1494    The minimum length for a match is one character. If the subject is "ABC", there
1495    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1496    If the pattern is studied, the final attempt does not take place, because PCRE
1497    knows that the subject is too short, and so the (*MARK) is never encountered.
1498    In this case, studying the pattern does not affect the overall match result,
1499    which is still "no match", but it does affect the auxiliary information that is
1500    returned.
1501  .sp  .sp
1502    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1503  .sp  .sp
1504  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
1505  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1506  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
1507  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
1508  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  strings in the
1509  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HTML <a href="pcre.html#utf8strings">
1510    .\" </a>
1511    section on UTF-8 support
1512    .\"
1513    in the main
1514    .\" HREF
1515    \fBpcre\fP
1516    .\"
1517    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1518    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1519    PCRE_ERROR_BADUTF8_OFFSET is returned.
1520  .P  .P
1521  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
1522  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 1094  PCRE_NO_UTF8_CHECK is set, the effect of Line 1528  PCRE_NO_UTF8_CHECK is set, the effect of
1528  subject, or a value of \fIstartoffset\fP that does not point to the start of a  subject, or a value of \fIstartoffset\fP that does not point to the start of a
1529  UTF-8 character, is undefined. Your program may crash.  UTF-8 character, is undefined. Your program may crash.
1530  .sp  .sp
1531    PCRE_PARTIAL    PCRE_PARTIAL_HARD
1532      PCRE_PARTIAL_SOFT
1533  .sp  .sp
1534  This option turns on the partial matching feature. If the subject string fails  These options turn on the partial matching feature. For backwards
1535  to match the pattern, but at some point during the matching process the end of  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1536  the subject was reached (that is, the subject partially matches the pattern and  occurs if the end of the subject string is reached successfully, but there are
1537  the failure to match occurred only because there were not enough subject  not enough subject characters to complete the match. If this happens when
1538  characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of  PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1539  PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what  testing any remaining alternatives. Only if no complete match can be found is
1540  may appear in the pattern. These are discussed in the  PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1541    PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1542    but only if no complete match can be found.
1543    .P
1544    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1545    partial match is found, \fBpcre_exec()\fP immediately returns
1546    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1547    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1548    important that an alternative complete match.
1549    .P
1550    In both cases, the portion of the string that was inspected when the partial
1551    match was found is set as the first matching string. There is a more detailed
1552    discussion of partial and multi-segment matching, with examples, in the
1553  .\" HREF  .\" HREF
1554  \fBpcrepartial\fP  \fBpcrepartial\fP
1555  .\"  .\"
# Line 1112  documentation. Line 1559  documentation.
1559  .rs  .rs
1560  .sp  .sp
1561  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
1562  \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
1563  \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
1564  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
1565  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
1566  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.
1567  .P  .P
1568  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
1569  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 1150  pattern. Following the usage in Jeffrey Line 1597  pattern. Following the usage in Jeffrey
1597  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
1598  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1599  .P  .P
1600  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
1601  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
1602  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
1603  this argument is NOT the size of \fIovector\fP in bytes.  argument is NOT the size of \fIovector\fP in bytes.
1604  .P  .P
1605  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,
1606  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
1607  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1608  and is not available for passing back information. The length passed in  and is not available for passing back information. The number passed in
1609  \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
1610  rounded down.  rounded down.
1611  .P  .P
1612  When a match is successful, information about captured substrings is returned  When a match is successful, information about captured substrings is returned
1613  in pairs of integers, starting at the beginning of \fIovector\fP, and  in pairs of integers, starting at the beginning of \fIovector\fP, and
1614  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
1615  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
1616  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
1617  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
1618  subject string matched by the entire pattern. The next pair is used for the  mode. They are not character counts.
1619  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  .P
1620  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
1621  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
1622  just the first pair of offsets has been set.  used for the first capturing subpattern, and so on. The value returned by
1623  .P  \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1624  Some convenience functions are provided for extracting the captured substrings  For example, if two substrings have been captured, the returned value is 3. If
1625  as separate strings. These are described in the following section.  there are no capturing subpatterns, the return value from a successful match is
1626  .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.  
1627  .P  .P
1628  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
1629  string that it matched that is returned.  string that it matched that is returned.
1630  .P  .P
1631  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
1632  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
1633  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,
1634  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
1635  \fIovecsize\fP as zero. However, if the pattern contains back references and  \fIovecsize\fP as zero. However, if the pattern contains back references and
1636  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
1637  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
1638  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1639  .P  .P
1640  Note that \fBpcre_info()\fP can be used to find out how many capturing  The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1641  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1642  \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
1643  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.
1644    .P
1645    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1646    the subject when subpattern \fIn\fP has not been used at all. For example, if
1647    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1648    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1649    happens, both values in the offset pairs corresponding to unused subpatterns
1650    are set to -1.
1651    .P
1652    Offset values that correspond to unused subpatterns at the end of the
1653    expression are also set to -1. For example, if the string "abc" is matched
1654    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1655    return from the function is 2, because the highest used capturing subpattern
1656    number is 1. However, you can refer to the offsets for the second and third
1657    capturing subpatterns if you wish (assuming the vector is large enough, of
1658    course).
1659    .P
1660    Some convenience functions are provided for extracting the captured substrings
1661    as separate strings. These are described below.
1662  .  .
1663  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
1664  .SS "Return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
1665  .rs  .rs
1666  .sp  .sp
1667  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 1228  compiled in an environment of one endian Line 1688  compiled in an environment of one endian
1688  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
1689  not present.  not present.
1690  .sp  .sp
1691    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1692  .sp  .sp
1693  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1694  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 1241  If a pattern contains back references, b Line 1701  If a pattern contains back references, b
1701  gets a block of memory at the start of matching to use for this purpose. If the  gets a block of memory at the start of matching to use for this purpose. If the
1702  call via \fBpcre_malloc()\fP fails, this error is given. The memory is  call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1703  automatically freed at the end of matching.  automatically freed at the end of matching.
1704    .P
1705    This error is also given if \fBpcre_stack_malloc()\fP fails in
1706    \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1707    \fB--disable-stack-for-recursion\fP.
1708  .sp  .sp
1709    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1710  .sp  .sp
# Line 1250  below). It is never returned by \fBpcre_ Line 1714  below). It is never returned by \fBpcre_
1714  .sp  .sp
1715    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1716  .sp  .sp
1717  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
1718  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1719  description above.  above.
1720  .sp  .sp
1721    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1722  .sp  .sp
# Line 1282  documentation for details of partial mat Line 1746  documentation for details of partial mat
1746  .sp  .sp
1747    PCRE_ERROR_BADPARTIAL     (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1748  .sp  .sp
1749  The PCRE_PARTIAL option was used with a compiled pattern containing items that  This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
1750  are not supported for partial matching. See the  option was used with a compiled pattern containing items that were not
1751  .\" HREF  supported for partial matching. From release 8.00 onwards, there are no
1752  \fBpcrepartial\fP  restrictions on partial matching.
 .\"  
 documentation for details of partial matching.  
1753  .sp  .sp
1754    PCRE_ERROR_INTERNAL       (-14)    PCRE_ERROR_INTERNAL       (-14)
1755  .sp  .sp
# Line 1297  in PCRE or by overwriting of the compile Line 1759  in PCRE or by overwriting of the compile
1759    PCRE_ERROR_BADCOUNT       (-15)    PCRE_ERROR_BADCOUNT       (-15)
1760  .sp  .sp
1761  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.
1762    .sp
1763      PCRE_ERROR_RECURSIONLIMIT (-21)
1764    .sp
1765    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1766    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1767    description above.
1768    .sp
1769      PCRE_ERROR_BADNEWLINE     (-23)
1770    .sp
1771    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1772    .P
1773    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1774  .  .
1775  .  .
1776  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1308  This error is given if the value of the Line 1782  This error is given if the value of the
1782  .ti +5n  .ti +5n
1783  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
1784  .PP  .PP
 .br  
1785  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1786  .ti +5n  .ti +5n
1787  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1788  .ti +5n  .ti +5n
1789  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
1790  .PP  .PP
 .br  
1791  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
1792  .ti +5n  .ti +5n
1793  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
# Line 1326  Captured substrings can be accessed dire Line 1798  Captured substrings can be accessed dire
1798  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1799  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1800  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1801  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1802  has a further zero added on the end, but the result is not, of course,  .P
1803  a C string.  A substring that contains a binary zero is correctly extracted and has a
1804    further zero added on the end, but the result is not, of course, a C string.
1805    However, you can process such a string by referring to the length that is
1806    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1807    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1808    for handling strings containing binary zeros, because the end of the final
1809    string is not independently indicated.
1810  .P  .P
1811  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:
1812  \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 1348  the string is placed in \fIbuffer\fP, wh Line 1826  the string is placed in \fIbuffer\fP, wh
1826  \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
1827  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1828  \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
1829  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1830  .sp  .sp
1831    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1832  .sp  .sp
# Line 1364  and builds a list of pointers to them. A Line 1842  and builds a list of pointers to them. A
1842  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
1843  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
1844  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
1845  function is zero if all went well, or  function is zero if all went well, or the error code
1846  .sp  .sp
1847    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1848  .sp  .sp
# Line 1383  a previous call of \fBpcre_get_substring Line 1861  a previous call of \fBpcre_get_substring
1861  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1862  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
1863  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
1864  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1865  \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
1866  provided.  provided.
1867  .  .
# Line 1395  provided. Line 1873  provided.
1873  .ti +5n  .ti +5n
1874  .B const char *\fIname\fP);  .B const char *\fIname\fP);
1875  .PP  .PP
 .br  
1876  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1877  .ti +5n  .ti +5n
1878  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1404  provided. Line 1881  provided.
1881  .ti +5n  .ti +5n
1882  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1883  .PP  .PP
 .br  
1884  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1885  .ti +5n  .ti +5n
1886  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1418  For example, for this pattern Line 1894  For example, for this pattern
1894  .sp  .sp
1895    (a+)b(?<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
1896  .sp  .sp
1897  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
1898  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
1899  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
1900    pattern, and the second is the name. The yield of the function is the
1901  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
1902  that name.  that name.
1903  .P  .P
# Line 1428  Given the number, you can extract the su Line 1905  Given the number, you can extract the su
1905  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
1906  two functions that do the whole job.  two functions that do the whole job.
1907  .P  .P
1908  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
1909  \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
1910  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
1911  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
1912  .P  .P
# Line 1439  pattern. This is needed in order to gain Line 1916  pattern. This is needed in order to gain
1916  translation table.  translation table.
1917  .P  .P
1918  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1919  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1920  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1921    the behaviour may not be what you want (see the next section).
1922    .P
1923    \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
1924    subpatterns with the same number, as described in the
1925    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1926    .\" </a>
1927    section on duplicate subpattern numbers
1928    .\"
1929    in the
1930    .\" HREF
1931    \fBpcrepattern\fP
1932    .\"
1933    page, you cannot use names to distinguish the different subpatterns, because
1934    names are not included in the compiled code. The matching process uses only
1935    numbers. For this reason, the use of different names for subpatterns of the
1936    same number causes an error at compile time.
1937    .
1938    .SH "DUPLICATE SUBPATTERN NAMES"
1939    .rs
1940    .sp
1941    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1942    .ti +5n
1943    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1944    .PP
1945    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1946    are not required to be unique. (Duplicate names are always allowed for
1947    subpatterns with the same number, created by using the (?| feature. Indeed, if
1948    such subpatterns are named, they are required to use the same names.)
1949    .P
1950    Normally, patterns with duplicate names are such that in any one match, only
1951    one of the named subpatterns participates. An example is shown in the
1952    .\" HREF
1953    \fBpcrepattern\fP
1954    .\"
1955    documentation.
1956    .P
1957    When duplicates are present, \fBpcre_copy_named_substring()\fP and
1958    \fBpcre_get_named_substring()\fP return the first substring corresponding to
1959    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1960    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1961    returns one of the numbers that are associated with the name, but it is not
1962    defined which it is.
1963    .P
1964    If you want to get full details of all captured substrings for a given name,
1965    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1966    argument is the compiled pattern, and the second is the name. The third and
1967    fourth are pointers to variables which are updated by the function. After it
1968    has run, they point to the first and last entries in the name-to-number table
1969    for the given name. The function itself returns the length of each entry, or
1970    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1971    described above in the section entitled \fIInformation about a pattern\fP.
1972    Given all the relevant entries for the name, you can extract each of their
1973    numbers, and hence the captured data, if any.
1974  .  .
1975  .  .
1976  .SH "FINDING ALL POSSIBLE MATCHES"  .SH "FINDING ALL POSSIBLE MATCHES"
# Line 1478  will yield PCRE_ERROR_NOMATCH. Line 2008  will yield PCRE_ERROR_NOMATCH.
2008  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
2009  .P  .P
2010  The function \fBpcre_dfa_exec()\fP is called to match a subject string against  The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2011  a compiled pattern, using a "DFA" matching algorithm. This has different  a compiled pattern, using a matching algorithm that scans the subject string
2012  characteristics to the normal algorithm, and is not compatible with Perl. Some  just once, and does not backtrack. This has different characteristics to the
2013  of the features of PCRE patterns are not supported. Nevertheless, there are  normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2014  times when this kind of matching can be useful. For a discussion of the two  patterns are not supported. Nevertheless, there are times when this kind of
2015  matching algorithms, see the  matching can be useful. For a discussion of the two matching algorithms, and a
2016    list of features that \fBpcre_dfa_exec()\fP does not support, see the
2017  .\" HREF  .\" HREF
2018  \fBpcrematching\fP  \fBpcrematching\fP
2019  .\"  .\"
# Line 1497  here. Line 2028  here.
2028  The two additional arguments provide workspace for the function. The workspace  The two additional arguments provide workspace for the function. The workspace
2029  vector should contain at least 20 elements. It is used for keeping track of  vector should contain at least 20 elements. It is used for keeping track of
2030  multiple paths through the pattern tree. More workspace will be needed for  multiple paths through the pattern tree. More workspace will be needed for
2031  patterns and subjects where there are a lot of possible matches.  patterns and subjects where there are a lot of potential matches.
2032  .P  .P
2033  Here is an example of a simple call to \fBpcre_exec()\fP:  Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2034  .sp  .sp
2035    int rc;    int rc;
2036    int ovector[10];    int ovector[10];
2037    int wspace[20];    int wspace[20];
2038    rc = pcre_exec(    rc = pcre_dfa_exec(
2039      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
2040      NULL,           /* we didn't study the pattern */      NULL,           /* we didn't study the pattern */
2041      "some string",  /* the subject string */      "some string",  /* the subject string */
# Line 1520  Here is an example of a simple call to \ Line 2051  Here is an example of a simple call to \
2051  .rs  .rs
2052  .sp  .sp
2053  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2054  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,
2055  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2056  PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are  PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2057  the same as for \fBpcre_exec()\fP, so their description is not repeated here.  PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2058  .sp  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2059    PCRE_PARTIAL  so their description is not repeated here.
2060  .sp  .sp
2061  This has the same general effect as it does for \fBpcre_exec()\fP, but the    PCRE_PARTIAL_HARD
2062  details are slightly different. When PCRE_PARTIAL is set for    PCRE_PARTIAL_SOFT
2063  \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into  .sp
2064  PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no  These have the same general effect as they do for \fBpcre_exec()\fP, but the
2065  complete matches, but there is still at least one matching possibility. The  details are slightly different. When PCRE_PARTIAL_HARD is set for
2066  portion of the string that provided the partial match is set as the first  \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2067  matching string.  is reached and there is still at least one matching possibility that requires
2068    additional characters. This happens even if some complete matches have also
2069    been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2070    is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2071    there have been no complete matches, but there is still at least one matching
2072    possibility. The portion of the string that was inspected when the longest
2073    partial match was found is set as the first matching string in both cases.
2074    There is a more detailed discussion of partial and multi-segment matching, with
2075    examples, in the
2076    .\" HREF
2077    \fBpcrepartial\fP
2078    .\"
2079    documentation.
2080  .sp  .sp
2081    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2082  .sp  .sp
2083  Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as  Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2084  soon as it has found one match. Because of the way the DFA algorithm works,  soon as it has found one match. Because of the way the alternative algorithm
2085  this is necessarily the shortest possible match at the first possible matching  works, this is necessarily the shortest possible match at the first possible
2086  point in the subject string.  matching point in the subject string.
2087  .sp  .sp
2088    PCRE_DFA_RESTART    PCRE_DFA_RESTART
2089  .sp  .sp
2090  When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns  When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2091  a partial match, it is possible to call it again, with additional subject  again, with additional subject characters, and have it continue with the same
2092  characters, and have it continue with the same match. The PCRE_DFA_RESTART  match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2093  option requests this action; when it is set, the \fIworkspace\fP and  \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2094  \fIwscount\fP options must reference the same vector as before because data  before because data about the match so far is left in them after a partial
2095  about the match so far is left in them after a partial match. There is more  match. There is more discussion of this facility in the
 discussion of this facility in the  
2096  .\" HREF  .\" HREF
2097  \fBpcrepartial\fP  \fBpcrepartial\fP
2098  .\"  .\"
# Line 1579  the three matched strings are Line 2121  the three matched strings are
2121  On success, the yield of the function is a number greater than zero, which is  On success, the yield of the function is a number greater than zero, which is
2122  the number of matched substrings. The substrings themselves are returned in  the number of matched substrings. The substrings themselves are returned in
2123  \fIovector\fP. Each string uses two elements; the first is the offset to the  \fIovector\fP. Each string uses two elements; the first is the offset to the
2124  start, and the second is the offset to the end. All the strings have the same  start, and the second is the offset to the end. In fact, all the strings have
2125  start offset. (Space could have been saved by giving this only once, but it was  the same start offset. (Space could have been saved by giving this only once,
2126  decided to retain some compatibility with the way \fBpcre_exec()\fP returns  but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2127  data, even though the meaning of the strings is different.)  returns data, even though the meaning of the strings is different.)
2128  .P  .P
2129  The strings are returned in reverse order of length; that is, the longest  The strings are returned in reverse order of length; that is, the longest
2130  matching string is given first. If there were too many matches to fit into  matching string is given first. If there were too many matches to fit into
# Line 1609  that it does not support, for instance, Line 2151  that it does not support, for instance,
2151  .sp  .sp
2152    PCRE_ERROR_DFA_UCOND      (-17)    PCRE_ERROR_DFA_UCOND      (-17)
2153  .sp  .sp
2154  This return is given if \fBpcre_dfa_exec()\fP encounters a condition item in a  This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2155  pattern that uses a back reference for the condition. This is not supported.  uses a back reference for the condition, or a test for recursion in a specific
2156    group. These are not supported.
2157  .sp  .sp
2158    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2159  .sp  .sp
# Line 1629  When a recursive subpattern is processed Line 2172  When a recursive subpattern is processed
2172  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2173  error is given if the output vector is not large enough. This should be  error is given if the output vector is not large enough. This should be
2174  extremely rare, as a vector of size 1000 is used.  extremely rare, as a vector of size 1000 is used.
2175  .P  .
2176  .in 0  .
2177  Last updated: 16 May 2005  .SH "SEE ALSO"
2178  .br  .rs
2179  Copyright (c) 1997-2005 University of Cambridge.  .sp
2180    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2181    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2182    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2183    .
2184    .
2185    .SH AUTHOR
2186    .rs
2187    .sp
2188    .nf
2189    Philip Hazel
2190    University Computing Service
2191    Cambridge CB2 3QH, England.
2192    .fi
2193    .
2194    .
2195    .SH REVISION
2196    .rs
2197    .sp
2198    .nf
2199    Last updated: 01 November 2010
2200    Copyright (c) 1997-2010 University of Cambridge.
2201    .fi

Legend:
Removed from v.77  
changed lines
  Added in v.566

  ViewVC Help
Powered by ViewVC 1.1.5