/[pcre]/code/trunk/doc/pcreapi.3
ViewVC logotype

Diff of /code/trunk/doc/pcreapi.3

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 75 by nigel, Sat Feb 24 21:40:37 2007 UTC revision 225 by ph10, Mon Aug 20 14:38:34 2007 UTC
# Line 1  Line 1 
1  .TH PCRE 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH "PCRE NATIVE API"  .SH "PCRE NATIVE API"
# Line 7  PCRE - Perl-compatible regular expressio Line 7  PCRE - Perl-compatible regular expressio
7  .B #include <pcre.h>  .B #include <pcre.h>
8  .PP  .PP
9  .SM  .SM
 .br  
10  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
11  .ti +5n  .ti +5n
12  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
13  .ti +5n  .ti +5n
14  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
15  .PP  .PP
16  .br  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
17    .ti +5n
18    .B int *\fIerrorcodeptr\fP,
19    .ti +5n
20    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
21    .ti +5n
22    .B const unsigned char *\fItableptr\fP);
23    .PP
24  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
25  .ti +5n  .ti +5n
26  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
27  .PP  .PP
 .br  
28  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
29  .ti +5n  .ti +5n
30  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
31  .ti +5n  .ti +5n
32  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
33  .PP  .PP
34  .br  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
35    .ti +5n
36    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
37    .ti +5n
38    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
39    .ti +5n
40    .B int *\fIworkspace\fP, int \fIwscount\fP);
41    .PP
42  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
43  .ti +5n  .ti +5n
44  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 35  PCRE - Perl-compatible regular expressio Line 47  PCRE - Perl-compatible regular expressio
47  .ti +5n  .ti +5n
48  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
49  .PP  .PP
 .br  
50  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
51  .ti +5n  .ti +5n
52  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
53  .ti +5n  .ti +5n
54  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
55  .PP  .PP
 .br  
56  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
57  .ti +5n  .ti +5n
58  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 51  PCRE - Perl-compatible regular expressio Line 61  PCRE - Perl-compatible regular expressio
61  .ti +5n  .ti +5n
62  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
63  .PP  .PP
 .br  
64  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
65  .ti +5n  .ti +5n
66  .B const char *\fIname\fP);  .B const char *\fIname\fP);
67  .PP  .PP
68  .br  .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
69    .ti +5n
70    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
71    .PP
72  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
73  .ti +5n  .ti +5n
74  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
75  .ti +5n  .ti +5n
76  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
77  .PP  .PP
 .br  
78  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
79  .ti +5n  .ti +5n
80  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
81  .PP  .PP
 .br  
82  .B void pcre_free_substring(const char *\fIstringptr\fP);  .B void pcre_free_substring(const char *\fIstringptr\fP);
83  .PP  .PP
 .br  
84  .B void pcre_free_substring_list(const char **\fIstringptr\fP);  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
85  .PP  .PP
 .br  
86  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
87  .PP  .PP
 .br  
88  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_fullinfo(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
89  .ti +5n  .ti +5n
90  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
91  .PP  .PP
 .br  
92  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int
93  .B *\fIfirstcharptr\fP);  .B *\fIfirstcharptr\fP);
94  .PP  .PP
95  .br  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
96    .PP
97  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
98  .PP  .PP
 .br  
99  .B char *pcre_version(void);  .B char *pcre_version(void);
100  .PP  .PP
 .br  
101  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
102  .PP  .PP
 .br  
103  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
104  .PP  .PP
 .br  
105  .B void *(*pcre_stack_malloc)(size_t);  .B void *(*pcre_stack_malloc)(size_t);
106  .PP  .PP
 .br  
107  .B void (*pcre_stack_free)(void *);  .B void (*pcre_stack_free)(void *);
108  .PP  .PP
 .br  
109  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
110  .  .
111  .  .
112  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
113  .rs  .rs
114  .sp  .sp
115  PCRE has its own native API, which is described in this document. There is also  PCRE has its own native API, which is described in this document. There are
116  a set of wrapper functions that correspond to the POSIX regular expression API.  also some wrapper functions that correspond to the POSIX regular expression
117  These are described in the  API. These are described in the
118  .\" HREF  .\" HREF
119  \fBpcreposix\fP  \fBpcreposix\fP
120  .\"  .\"
121  documentation.  documentation. Both of these APIs define a set of C function calls. A C++
122    wrapper is distributed with PCRE. It is documented in the
123    .\" HREF
124    \fBpcrecpp\fP
125    .\"
126    page.
127  .P  .P
128  The native API function prototypes are defined in the header file \fBpcre.h\fP,  The native API C function prototypes are defined in the header file
129  and on Unix systems the library itself is called \fBlibpcre\fP. It can  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.
130  normally be accessed by adding \fB-lpcre\fP to the command for linking an  It can normally be accessed by adding \fB-lpcre\fP to the command for linking
131  application that uses PCRE. The header file defines the macros PCRE_MAJOR and  an application that uses PCRE. The header file defines the macros PCRE_MAJOR
132  PCRE_MINOR to contain the major and minor release numbers for the library.  and PCRE_MINOR to contain the major and minor release numbers for the library.
133  Applications can use these to include support for different releases of PCRE.  Applications can use these to include support for different releases of PCRE.
134  .P  .P
135  The functions \fBpcre_compile()\fP, \fBpcre_study()\fP, and \fBpcre_exec()\fP  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
136  are used for compiling and matching regular expressions. A sample program that  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
137  demonstrates the simplest way of using them is provided in the file called  in a Perl-compatible manner. A sample program that demonstrates the simplest
138  \fIpcredemo.c\fP in the source distribution. The  way of using them is provided in the file called \fIpcredemo.c\fP in the source
139    distribution. The
140  .\" HREF  .\" HREF
141  \fBpcresample\fP  \fBpcresample\fP
142  .\"  .\"
143  documentation describes how to run it.  documentation describes how to run it.
144  .P  .P
145    A second matching function, \fBpcre_dfa_exec()\fP, which is not
146    Perl-compatible, is also provided. This uses a different algorithm for the
147    matching. The alternative algorithm finds all possible matches (at a given
148    point in the subject), and scans the subject just once. However, this algorithm
149    does not return captured substrings. A description of the two matching
150    algorithms and their advantages and disadvantages is given in the
151    .\" HREF
152    \fBpcrematching\fP
153    .\"
154    documentation.
155    .P
156  In addition to the main compiling and matching functions, there are convenience  In addition to the main compiling and matching functions, there are convenience
157  functions for extracting captured substrings from a matched subject string.  functions for extracting captured substrings from a subject string that is
158  They are:  matched by \fBpcre_exec()\fP. They are:
159  .sp  .sp
160    \fBpcre_copy_substring()\fP    \fBpcre_copy_substring()\fP
161    \fBpcre_copy_named_substring()\fP    \fBpcre_copy_named_substring()\fP
# Line 145  They are: Line 163  They are:
163    \fBpcre_get_named_substring()\fP    \fBpcre_get_named_substring()\fP
164    \fBpcre_get_substring_list()\fP    \fBpcre_get_substring_list()\fP
165    \fBpcre_get_stringnumber()\fP    \fBpcre_get_stringnumber()\fP
166      \fBpcre_get_stringtable_entries()\fP
167  .sp  .sp
168  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
169  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
170  .P  .P
171  The function \fBpcre_maketables()\fP is used to build a set of character tables  The function \fBpcre_maketables()\fP is used to build a set of character tables
172  in the current locale for passing to \fBpcre_compile()\fP or \fBpcre_exec()\fP.  in the current locale for passing to \fBpcre_compile()\fP, \fBpcre_exec()\fP,
173  This is an optional facility that is provided for specialist use. Most  or \fBpcre_dfa_exec()\fP. This is an optional facility that is provided for
174  commonly, no special tables are passed, in which case internal tables that are  specialist use. Most commonly, no special tables are passed, in which case
175  generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
176  .P  .P
177  The function \fBpcre_fullinfo()\fP is used to find out information about a  The function \fBpcre_fullinfo()\fP is used to find out information about a
178  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only
# Line 161  some of the available information, but i Line 180  some of the available information, but i
180  The function \fBpcre_version()\fP returns a pointer to a string containing the  The function \fBpcre_version()\fP returns a pointer to a string containing the
181  version of PCRE and its date of release.  version of PCRE and its date of release.
182  .P  .P
183    The function \fBpcre_refcount()\fP maintains a reference count in a data block
184    containing a compiled pattern. This is provided for the benefit of
185    object-oriented applications.
186    .P
187  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain  The global variables \fBpcre_malloc\fP and \fBpcre_free\fP initially contain
188  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,  the entry points of the standard \fBmalloc()\fP and \fBfree()\fP functions,
189  respectively. PCRE calls the memory management functions via these variables,  respectively. PCRE calls the memory management functions via these variables,
# Line 170  should be done before calling any PCRE f Line 193  should be done before calling any PCRE f
193  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also  The global variables \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are also
194  indirections to memory management functions. These special functions are used  indirections to memory management functions. These special functions are used
195  only when PCRE is compiled to use the heap for remembering data, instead of  only when PCRE is compiled to use the heap for remembering data, instead of
196  recursive function calls. This is a non-standard way of building PCRE, for use  recursive function calls, when running the \fBpcre_exec()\fP function. See the
197  in environments that have limited stacks. Because of the greater use of memory  .\" HREF
198  management, it runs more slowly. Separate functions are provided so that  \fBpcrebuild\fP
199  special-purpose external code can be used for this case. When used, these  .\"
200  functions are always called in a stack-like manner (last obtained, first  documentation for details of how to do this. It is a non-standard way of
201  freed), and always for memory blocks of the same size.  building PCRE, for use in environments that have limited stacks. Because of the
202    greater use of memory management, it runs more slowly. Separate functions are
203    provided so that special-purpose external code can be used for this case. When
204    used, these functions are always called in a stack-like manner (last obtained,
205    first freed), and always for memory blocks of the same size. There is a
206    discussion about PCRE's stack usage in the
207    .\" HREF
208    \fBpcrestack\fP
209    .\"
210    documentation.
211  .P  .P
212  The global variable \fBpcre_callout\fP initially contains NULL. It can be set  The global variable \fBpcre_callout\fP initially contains NULL. It can be set
213  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
# Line 186  points during a matching operation. Deta Line 218  points during a matching operation. Deta
218  documentation.  documentation.
219  .  .
220  .  .
221    .SH NEWLINES
222    .rs
223    .sp
224    PCRE supports five different conventions for indicating line breaks in
225    strings: a single CR (carriage return) character, a single LF (linefeed)
226    character, the two-character sequence CRLF, any of the three preceding, or any
227    Unicode newline sequence. The Unicode newline sequences are the three just
228    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
229    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
230    (paragraph separator, U+2029).
231    .P
232    Each of the first three conventions is used by at least one operating system as
233    its standard newline sequence. When PCRE is built, a default can be specified.
234    The default default is LF, which is the Unix standard. When PCRE is run, the
235    default can be overridden, either when a pattern is compiled, or when it is
236    matched.
237    .P
238    In the PCRE documentation the word "newline" is used to mean "the character or
239    pair of characters that indicate a line break". The choice of newline
240    convention affects the handling of the dot, circumflex, and dollar
241    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
242    recognized line ending sequence, the match position advancement for a
243    non-anchored pattern. The choice of newline convention does not affect the
244    interpretation of the \en or \er escape sequences.
245    .
246    .
247  .SH MULTITHREADING  .SH MULTITHREADING
248  .rs  .rs
249  .sp  .sp
# Line 207  which it was compiled. Details are given Line 265  which it was compiled. Details are given
265  .\" HREF  .\" HREF
266  \fBpcreprecompile\fP  \fBpcreprecompile\fP
267  .\"  .\"
268  documentation.  documentation. However, compiling a regular expression with one version of PCRE
269    for use with a different version is not guaranteed to work and may cause
270    crashes.
271  .  .
272  .  .
273  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 238  properties is available; otherwise it is Line 298  properties is available; otherwise it is
298  .sp  .sp
299    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
300  .sp  .sp
301  The output is an integer that is set to the value of the code that is used for  The output is an integer whose value specifies the default character sequence
302  the newline character. It is either linefeed (10) or carriage return (13), and  that is recognized as meaning "newline". The four values that are supported
303  should normally be the standard character for your operating system.  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
304    default should normally be the standard sequence for your operating system.
305  .sp  .sp
306    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
307  .sp  .sp
# Line 266  The output is an integer that gives the Line 327  The output is an integer that gives the
327  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
328  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
329  .sp  .sp
330      PCRE_CONFIG_MATCH_LIMIT_RECURSION
331    .sp
332    The output is an integer that gives the default limit for the depth of
333    recursion when calling the internal matching function in a \fBpcre_exec()\fP
334    execution. Further details are given with \fBpcre_exec()\fP below.
335    .sp
336    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
337  .sp  .sp
338  The output is an integer that is set to one if internal recursion is  The output is an integer that is set to one if internal recursion when running
339  implemented by recursive function calls that use the stack to remember their  \fBpcre_exec()\fP is implemented by recursive function calls that use the stack
340  state. This is the usual way that PCRE is compiled. The output is zero if PCRE  to remember their state. This is the usual way that PCRE is compiled. The
341  was compiled to use blocks of data on the heap instead of recursive function  output is zero if PCRE was compiled to use blocks of data on the heap instead
342  calls. In this case, \fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP are  of recursive function calls. In this case, \fBpcre_stack_malloc\fP and
343  called to manage memory blocks on the heap, thus avoiding the use of the stack.  \fBpcre_stack_free\fP are called to manage memory blocks on the heap, thus
344    avoiding the use of the stack.
345  .  .
346  .  .
347  .SH "COMPILING A PATTERN"  .SH "COMPILING A PATTERN"
# Line 284  called to manage memory blocks on the he Line 352  called to manage memory blocks on the he
352  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
353  .ti +5n  .ti +5n
354  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
355    .sp
356    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
357    .ti +5n
358    .B int *\fIerrorcodeptr\fP,
359    .ti +5n
360    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
361    .ti +5n
362    .B const unsigned char *\fItableptr\fP);
363  .P  .P
364  The function \fBpcre_compile()\fP is called to compile a pattern into an  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
365  internal form. The pattern is a C string terminated by a binary zero, and  called to compile a pattern into an internal form. The only difference between
366  is passed in the \fIpattern\fP argument. A pointer to a single block of memory  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
367  that is obtained via \fBpcre_malloc\fP is returned. This contains the compiled  \fIerrorcodeptr\fP, via which a numerical error code can be returned.
368  code and related data. The \fBpcre\fP type is defined for the returned block;  .P
369  this is a typedef for a structure whose contents are not externally defined. It  The pattern is a C string terminated by a binary zero, and is passed in the
370  is up to the caller to free the memory when it is no longer required.  \fIpattern\fP argument. A pointer to a single block of memory that is obtained
371    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
372    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
373    for a structure whose contents are not externally defined. It is up to the
374    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
375  .P  .P
376  Although the compiled code of a PCRE regex is relocatable, that is, it does not  Although the compiled code of a PCRE regex is relocatable, that is, it does not
377  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
378  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
379  argument, which is an address (see below).  argument, which is an address (see below).
380  .P  .P
381  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
382  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
383  options are described below. Some of them, in particular, those that are  options are described below. Some of them, in particular, those that are
384  compatible with Perl, can also be set and unset from within the pattern (see  compatible with Perl, can also be set and unset from within the pattern (see
# Line 308  the detailed description in the Line 388  the detailed description in the
388  .\"  .\"
389  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For these options, the contents of the \fIoptions\fP argument
390  specifies their initial settings at the start of compilation and execution. The  specifies their initial settings at the start of compilation and execution. The
391  PCRE_ANCHORED option can be set at the time of matching as well as at compile  PCRE_ANCHORED and PCRE_NEWLINE_\fIxxx\fP options can be set at the time of
392  time.  matching as well as at compile time.
393  .P  .P
394  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
395  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
396  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
397  error message. The offset from the start of the pattern to the character where  error message. This is a static string that is part of the library. You must
398  the error was discovered is placed in the variable pointed to by  not try to free it. The offset from the start of the pattern to the character
399    where the error was discovered is placed in the variable pointed to by
400  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.
401  .P  .P
402    If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
403    \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
404    returned via this argument in the event of an error. This is in addition to the
405    textual error message. Error codes and messages are listed below.
406    .P
407  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of  If the final argument, \fItableptr\fP, is NULL, PCRE uses a default set of
408  character tables that are built when PCRE is compiled, using the default C  character tables that are built when PCRE is compiled, using the default C
409  locale. Otherwise, \fItableptr\fP must be an address that is the result of a  locale. Otherwise, \fItableptr\fP must be an address that is the result of a
# Line 362  documentation. Line 448  documentation.
448  .sp  .sp
449  If this bit is set, letters in the pattern match both upper and lower case  If this bit is set, letters in the pattern match both upper and lower case
450  letters. It is equivalent to Perl's /i option, and it can be changed within a  letters. It is equivalent to Perl's /i option, and it can be changed within a
451  pattern by a (?i) option setting. When running in UTF-8 mode, case support for  pattern by a (?i) option setting. In UTF-8 mode, PCRE always understands the
452  high-valued characters is available only when PCRE is built with Unicode  concept of case for characters whose values are less than 128, so caseless
453  character property support.  matching is always possible. For characters with higher values, the concept of
454    case is supported if PCRE is compiled with Unicode property support, but not
455    otherwise. If you want to use caseless matching for characters 128 and above,
456    you must ensure that PCRE is compiled with Unicode property support as well as
457    with UTF-8 support.
458  .sp  .sp
459    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
460  .sp  .sp
461  If this bit is set, a dollar metacharacter in the pattern matches only at the  If this bit is set, a dollar metacharacter in the pattern matches only at the
462  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
463  immediately before the final character if it is a newline (but not before any  immediately before a newline at the end of the string (but not before any other
464  other newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is  newlines). The PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
465  set. There is no equivalent to this option in Perl, and no way to set it within  There is no equivalent to this option in Perl, and no way to set it within a
466  a pattern.  pattern.
467  .sp  .sp
468    PCRE_DOTALL    PCRE_DOTALL
469  .sp  .sp
470  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharater in the pattern matches all characters,
471  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
472  equivalent to Perl's /s option, and it can be changed within a pattern by a  the current position is at a newline. This option is equivalent to Perl's /s
473  (?s) option setting. A negative class such as [^a] always matches a newline  option, and it can be changed within a pattern by a (?s) option setting. A
474  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
475    the setting of this option.
476    .sp
477      PCRE_DUPNAMES
478    .sp
479    If this bit is set, names used to identify capturing subpatterns need not be
480    unique. This can be helpful for certain types of pattern when it is known that
481    only one instance of the named subpattern can ever be matched. There are more
482    details of named subpatterns below; see also the
483    .\" HREF
484    \fBpcrepattern\fP
485    .\"
486    documentation.
487  .sp  .sp
488    PCRE_EXTENDED    PCRE_EXTENDED
489  .sp  .sp
490  If this bit is set, whitespace data characters in the pattern are totally  If this bit is set, whitespace data characters in the pattern are totally
491  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
492  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
493  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
494  inclusive, are also ignored. This is equivalent to Perl's /x option, and it can  ignored. This is equivalent to Perl's /x option, and it can be changed within a
495  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
496  .P  .P
497  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
498  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
# Line 404  that is incompatible with Perl, but it i Line 506  that is incompatible with Perl, but it i
506  set, any backslash in a pattern that is followed by a letter that has no  set, any backslash in a pattern that is followed by a letter that has no
507  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
508  expansion. By default, as in Perl, a backslash followed by a letter with no  expansion. By default, as in Perl, a backslash followed by a letter with no
509  special meaning is treated as a literal. There are at present no other features  special meaning is treated as a literal. (Perl can, however, be persuaded to
510  controlled by this option. It can also be set by a (?X) option setting within a  give a warning for this.) There are at present no other features controlled by
511  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
512    .sp
513      PCRE_FIRSTLINE
514    .sp
515    If this option is set, an unanchored pattern is required to match before or at
516    the first newline in the subject string, though the matched text may continue
517    over the newline.
518  .sp  .sp
519    PCRE_MULTILINE    PCRE_MULTILINE
520  .sp  .sp
# Line 418  terminating newline (unless PCRE_DOLLAR_ Line 526  terminating newline (unless PCRE_DOLLAR_
526  Perl.  Perl.
527  .P  .P
528  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs  When PCRE_MULTILINE it is set, the "start of line" and "end of line" constructs
529  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
530  string, respectively, as well as at the very start and end. This is equivalent  subject string, respectively, as well as at the very start and end. This is
531  to Perl's /m option, and it can be changed within a pattern by a (?m) option  equivalent to Perl's /m option, and it can be changed within a pattern by a
532  setting. If there are no "\en" characters in a subject string, or no  (?m) option setting. If there are no newlines in a subject string, or no
533  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
534  .sp  .sp
535      PCRE_NEWLINE_CR
536      PCRE_NEWLINE_LF
537      PCRE_NEWLINE_CRLF
538      PCRE_NEWLINE_ANYCRLF
539      PCRE_NEWLINE_ANY
540    .sp
541    These options override the default newline definition that was chosen when PCRE
542    was built. Setting the first or the second specifies that a newline is
543    indicated by a single character (CR or LF, respectively). Setting
544    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
545    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
546    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
547    that any Unicode newline sequence should be recognized. The Unicode newline
548    sequences are the three just mentioned, plus the single characters VT (vertical
549    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
550    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
551    recognized only in UTF-8 mode.
552    .P
553    The newline setting in the options word uses three bits that are treated
554    as a number, giving eight possibilities. Currently only six are used (default
555    plus the five values above). This means that if you set more than one newline
556    option, the combination may or may not be sensible. For example,
557    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
558    other combinations may yield unused numbers and cause an error.
559    .P
560    The only time that a line break is specially recognized when compiling a
561    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
562    class is encountered. This indicates a comment that lasts until after the next
563    line break sequence. In other circumstances, line break sequences are treated
564    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
565    as whitespace characters and are therefore ignored.
566    .P
567    The newline option that is set at compile time becomes the default that is used
568    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
569    .sp
570    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
571  .sp  .sp
572  If this option is set, it disables the use of numbered capturing parentheses in  If this option is set, it disables the use of numbered capturing parentheses in
# Line 458  page. Line 601  page.
601    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
602  .sp  .sp
603  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
604  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
605  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcre.html#utf8strings">
606  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
607  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
608  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
609  Note that this option can also be passed to \fBpcre_exec()\fP, to suppress the  in the main
610  UTF-8 validity checking of subject strings.  .\" HREF
611    \fBpcre\fP
612    .\"
613    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
614    returns an error. If you already know that your pattern is valid, and you want
615    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
616    option. When it is set, the effect of passing an invalid UTF-8 string as a
617    pattern is undefined. It may cause your program to crash. Note that this option
618    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
619    the UTF-8 validity checking of subject strings.
620    .
621    .
622    .SH "COMPILATION ERROR CODES"
623    .rs
624    .sp
625    The following table lists the error codes than may be returned by
626    \fBpcre_compile2()\fP, along with the error messages that may be returned by
627    both compiling functions. As PCRE has developed, some error codes have fallen
628    out of use. To avoid confusion, they have not been re-used.
629    .sp
630       0  no error
631       1  \e at end of pattern
632       2  \ec at end of pattern
633       3  unrecognized character follows \e
634       4  numbers out of order in {} quantifier
635       5  number too big in {} quantifier
636       6  missing terminating ] for character class
637       7  invalid escape sequence in character class
638       8  range out of order in character class
639       9  nothing to repeat
640      10  [this code is not in use]
641      11  internal error: unexpected repeat
642      12  unrecognized character after (?
643      13  POSIX named classes are supported only within a class
644      14  missing )
645      15  reference to non-existent subpattern
646      16  erroffset passed as NULL
647      17  unknown option bit(s) set
648      18  missing ) after comment
649      19  [this code is not in use]
650      20  regular expression too large
651      21  failed to get memory
652      22  unmatched parentheses
653      23  internal error: code overflow
654      24  unrecognized character after (?<
655      25  lookbehind assertion is not fixed length
656      26  malformed number or name after (?(
657      27  conditional group contains more than two branches
658      28  assertion expected after (?(
659      29  (?R or (?[+-]digits must be followed by )
660      30  unknown POSIX class name
661      31  POSIX collating elements are not supported
662      32  this version of PCRE is not compiled with PCRE_UTF8 support
663      33  [this code is not in use]
664      34  character value in \ex{...} sequence is too large
665      35  invalid condition (?(0)
666      36  \eC not allowed in lookbehind assertion
667      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
668      38  number after (?C is > 255
669      39  closing ) for (?C expected
670      40  recursive call could loop indefinitely
671      41  unrecognized character after (?P
672      42  syntax error in subpattern name (missing terminator)
673      43  two named subpatterns have the same name
674      44  invalid UTF-8 string
675      45  support for \eP, \ep, and \eX has not been compiled
676      46  malformed \eP or \ep sequence
677      47  unknown property name after \eP or \ep
678      48  subpattern name is too long (maximum 32 characters)
679      49  too many named subpatterns (maximum 10,000)
680      50  [this code is not in use]
681      51  octal value is greater than \e377 (not in UTF-8 mode)
682      52  internal error: overran compiling workspace
683      53  internal error: previously-checked referenced subpattern not found
684      54  DEFINE group contains more than one branch
685      55  repeating a DEFINE group is not allowed
686      56  inconsistent NEWLINE options"
687      57  \eg is not followed by a braced name or an optionally braced
688            non-zero number
689      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
690  .  .
691  .  .
692  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
693  .rs  .rs
694  .sp  .sp
695  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
696  .ti +5n  .ti +5n
697  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
698  .PP  .PP
# Line 492  below Line 714  below
714  .\"  .\"
715  in the section on matching a pattern.  in the section on matching a pattern.
716  .P  .P
717  If studying the pattern does not produce any additional information,  If studying the pattern does not produce any additional information
718  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
719  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its
720  own \fBpcre_extra\fP block.  own \fBpcre_extra\fP block.
# Line 502  options are defined, and this argument s Line 724  options are defined, and this argument s
724  .P  .P
725  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
726  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
727  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
728  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
729  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
730    sure that it has run successfully.
731  .P  .P
732  This is a typical call to \fBpcre_study\fP():  This is a typical call to \fBpcre_study\fP():
733  .sp  .sp
# Line 525  bytes is created. Line 748  bytes is created.
748  .sp  .sp
749  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
750  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
751  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
752  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but
753  can be tested with \ep if PCRE is built with Unicode character property  can be tested with \ep if PCRE is built with Unicode character property
754  support.)  support. The use of locales with Unicode is discouraged. If you are handling
755  .P  characters with codes greater than 128, you should either use UTF-8 and
756  An internal set of tables is created in the default C locale when PCRE is  Unicode, or use locales, but not try to mix the two.
757  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  .P
758  and is sufficient for many applications. An alternative set of tables can,  PCRE contains an internal set of tables that are used when the final argument
759  however, be supplied. These may be created in a different locale from the  of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
760  default. As more and more applications change to using Unicode, the need for  Normally, the internal tables recognize only ASCII characters. However, when
761  this locale support is expected to die away.  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
762    default "C" locale of the local system, which may cause them to be different.
763    .P
764    The internal tables can always be overridden by tables supplied by the
765    application that calls PCRE. These may be created in a different locale from
766    the default. As more and more applications change to using Unicode, the need
767    for this locale support is expected to die away.
768  .P  .P
769  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
770  which has no arguments, in the relevant locale. The result can then be passed  which has no arguments, in the relevant locale. The result can then be passed
# Line 548  the following code could be used: Line 777  the following code could be used:
777    tables = pcre_maketables();    tables = pcre_maketables();
778    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
779  .sp  .sp
780    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
781    are using Windows, the name for the French locale is "french".
782    .P
783  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
784  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
785  that the memory containing the tables remains available for as long as it is  that the memory containing the tables remains available for as long as it is
# Line 594  check against passing an arbitrary memor Line 826  check against passing an arbitrary memor
826  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
827  .sp  .sp
828    int rc;    int rc;
829    unsigned long int length;    size_t length;
830    rc = pcre_fullinfo(    rc = pcre_fullinfo(
831      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
832      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
# Line 615  no back references. Line 847  no back references.
847  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
848  should point to an \fBint\fP variable.  should point to an \fBint\fP variable.
849  .sp  .sp
850    PCRE_INFO_DEFAULTTABLES    PCRE_INFO_DEFAULT_TABLES
851  .sp  .sp
852  Return a pointer to the internal default character tables within PCRE. The  Return a pointer to the internal default character tables within PCRE. The
853  fourth argument should point to an \fBunsigned char *\fP variable. This  fourth argument should point to an \fBunsigned char *\fP variable. This
# Line 626  a NULL table pointer. Line 858  a NULL table pointer.
858    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
859  .sp  .sp
860  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
861  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
862  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
863    still recognized for backwards compatibility.)
864  .P  .P
865  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
866  (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  
867  .sp  .sp
868  (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
869  starts with "^", or  starts with "^", or
# Line 650  table indicating a fixed set of bytes fo Line 882  table indicating a fixed set of bytes fo
882  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
883  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
884  .sp  .sp
885      PCRE_INFO_JCHANGED
886    .sp
887    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
888    fourth argument should point to an \fBint\fP variable. The (?J) internal option
889    setting changes the local PCRE_DUPNAMES option.
890    .sp
891    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
892  .sp  .sp
893  Return the value of the rightmost literal byte that must exist in any matched  Return the value of the rightmost literal byte that must exist in any matched
# Line 666  is -1. Line 904  is -1.
904  .sp  .sp
905  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
906  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
907  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
908  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
909  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
910  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
911  \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,
912  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
913    values.
914  .P  .P
915  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
916  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each  the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size of each
# Line 680  length of the longest name. PCRE_INFO_NA Line 919  length of the longest name. PCRE_INFO_NA
919  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
920  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
921  rest of the entry is the corresponding name, zero terminated. The names are in  rest of the entry is the corresponding name, zero terminated. The names are in
922  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
923    their parentheses numbers. For example, consider the following pattern (assume
924  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
925  .sp  .sp
926  .\" JOIN  .\" JOIN
927    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
928    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
929  .sp  .sp
930  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
931  in the table is eight bytes long. The table is as follows, with non-printing  in the table is eight bytes long. The table is as follows, with non-printing
# Line 697  bytes shows in hexadecimal, and undefine Line 937  bytes shows in hexadecimal, and undefine
937    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
938  .sp  .sp
939  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
940  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
941  different for each compiled pattern.  different for each compiled pattern.
942  .sp  .sp
943      PCRE_INFO_OKPARTIAL
944    .sp
945    Return 1 if the pattern can be used for partial matching, otherwise 0. The
946    fourth argument should point to an \fBint\fP variable. The
947    .\" HREF
948    \fBpcrepartial\fP
949    .\"
950    documentation lists the restrictions that apply to patterns when partial
951    matching is used.
952    .sp
953    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
954  .sp  .sp
955  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
956  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
957  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
958  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
959    they are the options that will be in force when matching starts. For example,
960    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
961    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
962  .P  .P
963  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
964  alternatives begin with one of the following:  alternatives begin with one of the following:
# Line 760  it is used to pass back information abou Line 1013  it is used to pass back information abou
1013  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1014  .  .
1015  .  .
1016  .SH "MATCHING A PATTERN"  .SH "REFERENCE COUNTS"
1017    .rs
1018    .sp
1019    .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1020    .PP
1021    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1022    data block that contains a compiled pattern. It is provided for the benefit of
1023    applications that operate in an object-oriented manner, where different parts
1024    of the application may be using the same compiled pattern, but you want to free
1025    the block when they are all done.
1026    .P
1027    When a pattern is compiled, the reference count field is initialized to zero.
1028    It is changed only by calling this function, whose action is to add the
1029    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1030    function is the new value. However, the value of the count is constrained to
1031    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1032    it is forced to the appropriate limit value.
1033    .P
1034    Except when it is zero, the reference count is not correctly preserved if a
1035    pattern is compiled on one host and then transferred to a host whose byte-order
1036    is different. (This seems a highly unlikely scenario.)
1037    .
1038    .
1039    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1040  .rs  .rs
1041  .sp  .sp
1042  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
# Line 772  string (see PCRE_INFO_FIRSTBYTE above). Line 1048  string (see PCRE_INFO_FIRSTBYTE above).
1048  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
1049  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1050  pattern has been studied, the result of the study should be passed in the  pattern has been studied, the result of the study should be passed in the
1051  \fIextra\fP argument.  \fIextra\fP argument. This function is the main matching facility of the
1052    library, and it operates in a Perl-like manner. For specialist use there is
1053    also an alternative matching function, which is described
1054    .\" HTML <a href="#dfamatch">
1055    .\" </a>
1056    below
1057    .\"
1058    in the section about the \fBpcre_dfa_exec()\fP function.
1059  .P  .P
1060  In most applications, the pattern will have been compiled (and optionally  In most applications, the pattern will have been compiled (and optionally
1061  studied) in the same process that calls \fBpcre_exec()\fP. However, it is  studied) in the same process that calls \fBpcre_exec()\fP. However, it is
# Line 796  Here is an example of a simple call to \ Line 1079  Here is an example of a simple call to \
1079      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1080      0,              /* default options */      0,              /* default options */
1081      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1082      30);            /* number of elements in the vector (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1083  .  .
1084  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1085  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
# Line 805  Here is an example of a simple call to \ Line 1088  Here is an example of a simple call to \
1088  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
1089  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
1090  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
1091  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
1092  follows:  fields (not necessarily in this order):
1093  .sp  .sp
1094    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1095    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1096    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1097      unsigned long int \fImatch_limit_recursion\fP;
1098    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1099    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1100  .sp  .sp
# Line 819  are set. The flag bits are: Line 1103  are set. The flag bits are:
1103  .sp  .sp
1104    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1105    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1106      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1107    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1108    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1109  .sp  .sp
# Line 833  but which have a very large number of po Line 1118  but which have a very large number of po
1118  classic example is the use of nested unlimited repeats.  classic example is the use of nested unlimited repeats.
1119  .P  .P
1120  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1121  (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
1122  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
1123  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
1124  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
1125    string.
1126  .P  .P
1127  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
1128  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
1129  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
1130  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
1131  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
1132  exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1133    .P
1134    The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1135    instead of limiting the total number of times that \fBmatch()\fP is called, it
1136    limits the depth of recursion. The recursion depth is a smaller number than the
1137    total number of calls, because not all calls to \fBmatch()\fP are recursive.
1138    This limit is of use only if it is set smaller than \fImatch_limit\fP.
1139    .P
1140    Limiting the recursion depth limits the amount of stack that can be used, or,
1141    when PCRE has been compiled to use memory on the heap instead of the stack, the
1142    amount of heap memory that can be used.
1143    .P
1144    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1145    built; the default default is the same value as the default for
1146    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1147    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1148    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1149    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1150  .P  .P
1151  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,
1152  which is described in the  which is described in the
# Line 870  documentation for a discussion of saving Line 1173  documentation for a discussion of saving
1173  .rs  .rs
1174  .sp  .sp
1175  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
1176  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,
1177  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1178  .sp  .sp
1179    PCRE_ANCHORED    PCRE_ANCHORED
1180  .sp  .sp
# Line 880  matching position. If a pattern was comp Line 1183  matching position. If a pattern was comp
1183  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
1184  matching time.  matching time.
1185  .sp  .sp
1186      PCRE_NEWLINE_CR
1187      PCRE_NEWLINE_LF
1188      PCRE_NEWLINE_CRLF
1189      PCRE_NEWLINE_ANYCRLF
1190      PCRE_NEWLINE_ANY
1191    .sp
1192    These options override the newline definition that was chosen or defaulted when
1193    the pattern was compiled. For details, see the description of
1194    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1195    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1196    the way the match position is advanced after a match failure for an unanchored
1197    pattern. When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
1198    set, and a match attempt fails when the current position is at a CRLF sequence,
1199    the match position is advanced by two characters instead of one, in other
1200    words, to after the CRLF.
1201    .P
1202    Anomalous effects can occur when CRLF is a valid newline sequence and explicit
1203    \er or \en escapes appear in the pattern. For example, the string "\er\enA"
1204    matches the unanchored pattern \enA but not [X\en]A. This happens because, in
1205    the first case, PCRE knows that the match must start with \en, and so it skips
1206    there before trying to match. In the second case, it has no knowledge about the
1207    starting character, so it starts matching at the beginning of the string, and
1208    on failing, skips over the CRLF as described above. However, if the pattern is
1209    studied, the match succeeds, because then PCRE once again knows where to start.
1210    .sp
1211    PCRE_NOTBOL    PCRE_NOTBOL
1212  .sp  .sp
1213  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 921  code that demonstrates how to do this in Line 1249  code that demonstrates how to do this in
1249  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
1250  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1251  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
1252  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
1253  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  strings in the
1254  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HTML <a href="pcre.html#utf8strings">
1255    .\" </a>
1256    section on UTF-8 support
1257    .\"
1258    in the main
1259    .\" HREF
1260    \fBpcre\fP
1261    .\"
1262    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1263    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1264    PCRE_ERROR_BADUTF8_OFFSET is returned.
1265  .P  .P
1266  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
1267  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 1011  is set to the offset of the first charac Line 1349  is set to the offset of the first charac
1349  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the
1350  subject string matched by the entire pattern. The next pair is used for the  subject string matched by the entire pattern. The next pair is used for the
1351  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP
1352  is the number of pairs that have been set. If there are no capturing  is one more than the highest numbered pair that has been set. For example, if
1353  subpatterns, the return value from a successful match is 1, indicating that  two substrings have been captured, the returned value is 3. If there are no
1354  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1355  .P  indicating that just the first pair of offsets has been set.
 Some convenience functions are provided for extracting the captured substrings  
 as separate strings. These are described in the following section.  
 .P  
 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.  
1356  .P  .P
1357  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
1358  string that it matched that is returned.  string that it matched that is returned.
# Line 1036  the \fIovector\fP is not big enough to r Line 1366  the \fIovector\fP is not big enough to r
1366  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
1367  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1368  .P  .P
1369  Note that \fBpcre_info()\fP can be used to find out how many capturing  The \fBpcre_info()\fP function can be used to find out how many capturing
1370  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1371  \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
1372  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.
1373    .P
1374    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1375    the subject when subpattern \fIn\fP has not been used at all. For example, if
1376    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1377    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1378    happens, both values in the offset pairs corresponding to unused subpatterns
1379    are set to -1.
1380    .P
1381    Offset values that correspond to unused subpatterns at the end of the
1382    expression are also set to -1. For example, if the string "abc" is matched
1383    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1384    return from the function is 2, because the highest used capturing subpattern
1385    number is 1. However, you can refer to the offsets for the second and third
1386    capturing subpatterns if you wish (assuming the vector is large enough, of
1387    course).
1388    .P
1389    Some convenience functions are provided for extracting the captured substrings
1390    as separate strings. These are described below.
1391  .  .
1392  .SS "Return values from \fBpcre_exec()\fP"  .\" HTML <a name="errorlist"></a>
1393    .SS "Error return values from \fBpcre_exec()\fP"
1394  .rs  .rs
1395  .sp  .sp
1396  If \fBpcre_exec()\fP fails, it returns a negative number. The following are  If \fBpcre_exec()\fP fails, it returns a negative number. The following are
# Line 1068  compiled in an environment of one endian Line 1417  compiled in an environment of one endian
1417  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
1418  not present.  not present.
1419  .sp  .sp
1420    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1421  .sp  .sp
1422  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1423  compiled pattern. This error could be caused by a bug in PCRE or by overwriting  compiled pattern. This error could be caused by a bug in PCRE or by overwriting
# Line 1090  below). It is never returned by \fBpcre_ Line 1439  below). It is never returned by \fBpcre_
1439  .sp  .sp
1440    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1441  .sp  .sp
1442  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
1443  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1444  description above.  above.
1445  .sp  .sp
1446    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1447  .sp  .sp
# Line 1112  A string that contains an invalid UTF-8 Line 1461  A string that contains an invalid UTF-8
1461  The UTF-8 byte sequence that was passed as a subject was valid, but the value  The UTF-8 byte sequence that was passed as a subject was valid, but the value
1462  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.
1463  .sp  .sp
1464    PCRE_ERROR_PARTIAL (-12)    PCRE_ERROR_PARTIAL        (-12)
1465  .sp  .sp
1466  The subject string did not match, but it did match partially. See the  The subject string did not match, but it did match partially. See the
1467  .\" HREF  .\" HREF
# Line 1120  The subject string did not match, but it Line 1469  The subject string did not match, but it
1469  .\"  .\"
1470  documentation for details of partial matching.  documentation for details of partial matching.
1471  .sp  .sp
1472    PCRE_ERROR_BAD_PARTIAL (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1473  .sp  .sp
1474  The PCRE_PARTIAL option was used with a compiled pattern containing items that  The PCRE_PARTIAL option was used with a compiled pattern containing items that
1475  are not supported for partial matching. See the  are not supported for partial matching. See the
# Line 1129  are not supported for partial matching. Line 1478  are not supported for partial matching.
1478  .\"  .\"
1479  documentation for details of partial matching.  documentation for details of partial matching.
1480  .sp  .sp
1481    PCRE_ERROR_INTERNAL (-14)    PCRE_ERROR_INTERNAL       (-14)
1482  .sp  .sp
1483  An unexpected internal error has occurred. This error could be caused by a bug  An unexpected internal error has occurred. This error could be caused by a bug
1484  in PCRE or by overwriting of the compiled pattern.  in PCRE or by overwriting of the compiled pattern.
1485  .sp  .sp
1486    PCRE_ERROR_BADCOUNT (-15)    PCRE_ERROR_BADCOUNT       (-15)
1487  .sp  .sp
1488  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.
1489    .sp
1490      PCRE_ERROR_RECURSIONLIMIT (-21)
1491    .sp
1492    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1493    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1494    description above.
1495    .sp
1496      PCRE_ERROR_BADNEWLINE     (-23)
1497    .sp
1498    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1499    .P
1500    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1501  .  .
1502  .  .
1503  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1148  This error is given if the value of the Line 1509  This error is given if the value of the
1509  .ti +5n  .ti +5n
1510  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
1511  .PP  .PP
 .br  
1512  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1513  .ti +5n  .ti +5n
1514  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1515  .ti +5n  .ti +5n
1516  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
1517  .PP  .PP
 .br  
1518  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
1519  .ti +5n  .ti +5n
1520  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
# Line 1166  Captured substrings can be accessed dire Line 1525  Captured substrings can be accessed dire
1525  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1526  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1527  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1528  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1529  has a further zero added on the end, but the result is not, of course,  .P
1530  a C string.  A substring that contains a binary zero is correctly extracted and has a
1531    further zero added on the end, but the result is not, of course, a C string.
1532    However, you can process such a string by referring to the length that is
1533    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1534    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1535    for handling strings containing binary zeros, because the end of the final
1536    string is not independently indicated.
1537  .P  .P
1538  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:
1539  \fIsubject\fP is the subject string that has just been successfully matched,  \fIsubject\fP is the subject string that has just been successfully matched,
# Line 1188  the string is placed in \fIbuffer\fP, wh Line 1553  the string is placed in \fIbuffer\fP, wh
1553  \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
1554  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1555  \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
1556  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1557  .sp  .sp
1558    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1559  .sp  .sp
# Line 1204  and builds a list of pointers to them. A Line 1569  and builds a list of pointers to them. A
1569  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
1570  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
1571  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
1572  function is zero if all went well, or  function is zero if all went well, or the error code
1573  .sp  .sp
1574    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1575  .sp  .sp
# Line 1223  a previous call of \fBpcre_get_substring Line 1588  a previous call of \fBpcre_get_substring
1588  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1589  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
1590  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
1591  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1592  \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
1593  provided.  provided.
1594  .  .
# Line 1235  provided. Line 1600  provided.
1600  .ti +5n  .ti +5n
1601  .B const char *\fIname\fP);  .B const char *\fIname\fP);
1602  .PP  .PP
 .br  
1603  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1604  .ti +5n  .ti +5n
1605  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1244  provided. Line 1608  provided.
1608  .ti +5n  .ti +5n
1609  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1610  .PP  .PP
 .br  
1611  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1612  .ti +5n  .ti +5n
1613  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1258  For example, for this pattern Line 1621  For example, for this pattern
1621  .sp  .sp
1622    (a+)b(?<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
1623  .sp  .sp
1624  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
1625  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
1626  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
1627    pattern, and the second is the name. The yield of the function is the
1628  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
1629  that name.  that name.
1630  .P  .P
# Line 1268  Given the number, you can extract the su Line 1632  Given the number, you can extract the su
1632  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
1633  two functions that do the whole job.  two functions that do the whole job.
1634  .P  .P
1635  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
1636  \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
1637  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
1638  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
1639  .P  .P
# Line 1279  pattern. This is needed in order to gain Line 1643  pattern. This is needed in order to gain
1643  translation table.  translation table.
1644  .P  .P
1645  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1646  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1647  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1648    the behaviour may not be what you want (see the next section).
1649    .
1650    .
1651    .SH "DUPLICATE SUBPATTERN NAMES"
1652    .rs
1653    .sp
1654    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1655    .ti +5n
1656    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1657    .PP
1658    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1659    are not required to be unique. Normally, patterns with duplicate names are such
1660    that in any one match, only one of the named subpatterns participates. An
1661    example is shown in the
1662    .\" HREF
1663    \fBpcrepattern\fP
1664    .\"
1665    documentation.
1666    .P
1667    When duplicates are present, \fBpcre_copy_named_substring()\fP and
1668    \fBpcre_get_named_substring()\fP return the first substring corresponding to
1669    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1670    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1671    returns one of the numbers that are associated with the name, but it is not
1672    defined which it is.
1673    .P
1674    If you want to get full details of all captured substrings for a given name,
1675    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1676    argument is the compiled pattern, and the second is the name. The third and
1677    fourth are pointers to variables which are updated by the function. After it
1678    has run, they point to the first and last entries in the name-to-number table
1679    for the given name. The function itself returns the length of each entry, or
1680    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1681    described above in the section entitled \fIInformation about a pattern\fP.
1682    Given all the relevant entries for the name, you can extract each of their
1683    numbers, and hence the captured data, if any.
1684    .
1685    .
1686    .SH "FINDING ALL POSSIBLE MATCHES"
1687    .rs
1688    .sp
1689    The traditional matching function uses a similar algorithm to Perl, which stops
1690    when it finds the first match, starting at a given point in the subject. If you
1691    want to find all possible matches, or the longest possible match, consider
1692    using the alternative matching function (see below) instead. If you cannot use
1693    the alternative function, but still need to find all possible matches, you
1694    can kludge it up by making use of the callout facility, which is described in
1695    the
1696    .\" HREF
1697    \fBpcrecallout\fP
1698    .\"
1699    documentation.
1700    .P
1701    What you have to do is to insert a callout right at the end of the pattern.
1702    When your callout function is called, extract and save the current matched
1703    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1704    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1705    will yield PCRE_ERROR_NOMATCH.
1706    .
1707    .
1708    .\" HTML <a name="dfamatch"></a>
1709    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1710    .rs
1711    .sp
1712    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1713    .ti +5n
1714    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1715    .ti +5n
1716    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1717    .ti +5n
1718    .B int *\fIworkspace\fP, int \fIwscount\fP);
1719    .P
1720    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1721    a compiled pattern, using a matching algorithm that scans the subject string
1722    just once, and does not backtrack. This has different characteristics to the
1723    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1724    patterns are not supported. Nevertheless, there are times when this kind of
1725    matching can be useful. For a discussion of the two matching algorithms, see
1726    the
1727    .\" HREF
1728    \fBpcrematching\fP
1729    .\"
1730    documentation.
1731    .P
1732    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1733    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1734    different way, and this is described below. The other common arguments are used
1735    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1736    here.
1737    .P
1738    The two additional arguments provide workspace for the function. The workspace
1739    vector should contain at least 20 elements. It is used for keeping track of
1740    multiple paths through the pattern tree. More workspace will be needed for
1741    patterns and subjects where there are a lot of potential matches.
1742    .P
1743    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1744    .sp
1745      int rc;
1746      int ovector[10];
1747      int wspace[20];
1748      rc = pcre_dfa_exec(
1749        re,             /* result of pcre_compile() */
1750        NULL,           /* we didn't study the pattern */
1751        "some string",  /* the subject string */
1752        11,             /* the length of the subject string */
1753        0,              /* start at offset 0 in the subject */
1754        0,              /* default options */
1755        ovector,        /* vector of integers for substring information */
1756        10,             /* number of elements (NOT size in bytes) */
1757        wspace,         /* working space vector */
1758        20);            /* number of elements (NOT size in bytes) */
1759    .
1760    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1761    .rs
1762    .sp
1763    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1764    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1765    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1766    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1767    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1768    .sp
1769      PCRE_PARTIAL
1770    .sp
1771    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1772    details are slightly different. When PCRE_PARTIAL is set for
1773    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1774    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1775    complete matches, but there is still at least one matching possibility. The
1776    portion of the string that provided the partial match is set as the first
1777    matching string.
1778    .sp
1779      PCRE_DFA_SHORTEST
1780    .sp
1781    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1782    soon as it has found one match. Because of the way the alternative algorithm
1783    works, this is necessarily the shortest possible match at the first possible
1784    matching point in the subject string.
1785    .sp
1786      PCRE_DFA_RESTART
1787    .sp
1788    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1789    a partial match, it is possible to call it again, with additional subject
1790    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1791    option requests this action; when it is set, the \fIworkspace\fP and
1792    \fIwscount\fP options must reference the same vector as before because data
1793    about the match so far is left in them after a partial match. There is more
1794    discussion of this facility in the
1795    .\" HREF
1796    \fBpcrepartial\fP
1797    .\"
1798    documentation.
1799    .
1800    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1801    .rs
1802    .sp
1803    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1804    substring in the subject. Note, however, that all the matches from one run of
1805    the function start at the same point in the subject. The shorter matches are
1806    all initial substrings of the longer matches. For example, if the pattern
1807    .sp
1808      <.*>
1809    .sp
1810    is matched against the string
1811    .sp
1812      This is <something> <something else> <something further> no more
1813    .sp
1814    the three matched strings are
1815    .sp
1816      <something>
1817      <something> <something else>
1818      <something> <something else> <something further>
1819    .sp
1820    On success, the yield of the function is a number greater than zero, which is
1821    the number of matched substrings. The substrings themselves are returned in
1822    \fIovector\fP. Each string uses two elements; the first is the offset to the
1823    start, and the second is the offset to the end. In fact, all the strings have
1824    the same start offset. (Space could have been saved by giving this only once,
1825    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1826    returns data, even though the meaning of the strings is different.)
1827  .P  .P
1828  .in 0  The strings are returned in reverse order of length; that is, the longest
1829  Last updated: 09 September 2004  matching string is given first. If there were too many matches to fit into
1830  .br  \fIovector\fP, the yield of the function is zero, and the vector is filled with
1831  Copyright (c) 1997-2004 University of Cambridge.  the longest matches.
1832    .
1833    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1834    .rs
1835    .sp
1836    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1837    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1838    described
1839    .\" HTML <a href="#errorlist">
1840    .\" </a>
1841    above.
1842    .\"
1843    There are in addition the following errors that are specific to
1844    \fBpcre_dfa_exec()\fP:
1845    .sp
1846      PCRE_ERROR_DFA_UITEM      (-16)
1847    .sp
1848    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1849    that it does not support, for instance, the use of \eC or a back reference.
1850    .sp
1851      PCRE_ERROR_DFA_UCOND      (-17)
1852    .sp
1853    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1854    uses a back reference for the condition, or a test for recursion in a specific
1855    group. These are not supported.
1856    .sp
1857      PCRE_ERROR_DFA_UMLIMIT    (-18)
1858    .sp
1859    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1860    block that contains a setting of the \fImatch_limit\fP field. This is not
1861    supported (it is meaningless).
1862    .sp
1863      PCRE_ERROR_DFA_WSSIZE     (-19)
1864    .sp
1865    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1866    \fIworkspace\fP vector.
1867    .sp
1868      PCRE_ERROR_DFA_RECURSE    (-20)
1869    .sp
1870    When a recursive subpattern is processed, the matching function calls itself
1871    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1872    error is given if the output vector is not large enough. This should be
1873    extremely rare, as a vector of size 1000 is used.
1874    .
1875    .
1876    .SH "SEE ALSO"
1877    .rs
1878    .sp
1879    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1880    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1881    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1882    .
1883    .
1884    .SH AUTHOR
1885    .rs
1886    .sp
1887    .nf
1888    Philip Hazel
1889    University Computing Service
1890    Cambridge CB2 3QH, England.
1891    .fi
1892    .
1893    .
1894    .SH REVISION
1895    .rs
1896    .sp
1897    .nf
1898    Last updated: 20 August 2007
1899    Copyright (c) 1997-2007 University of Cambridge.
1900    .fi

Legend:
Removed from v.75  
changed lines
  Added in v.225

  ViewVC Help
Powered by ViewVC 1.1.5