/[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 196 by ph10, Mon Jul 30 13:44:50 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 463  automatically checked. If an invalid UTF Line 606  automatically checked. If an invalid UTF
606  valid, and you want to skip this check for performance reasons, you can set the  valid, and you want to skip this check for performance reasons, you can set the
607  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid
608  UTF-8 string as a pattern is undefined. It may cause your program to crash.  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  Note that this option can also be passed to \fBpcre_exec()\fP and
610  UTF-8 validity checking of subject strings.  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject
611    strings.
612    .
613    .
614    .SH "COMPILATION ERROR CODES"
615    .rs
616    .sp
617    The following table lists the error codes than may be returned by
618    \fBpcre_compile2()\fP, along with the error messages that may be returned by
619    both compiling functions. As PCRE has developed, some error codes have fallen
620    out of use. To avoid confusion, they have not been re-used.
621    .sp
622       0  no error
623       1  \e at end of pattern
624       2  \ec at end of pattern
625       3  unrecognized character follows \e
626       4  numbers out of order in {} quantifier
627       5  number too big in {} quantifier
628       6  missing terminating ] for character class
629       7  invalid escape sequence in character class
630       8  range out of order in character class
631       9  nothing to repeat
632      10  [this code is not in use]
633      11  internal error: unexpected repeat
634      12  unrecognized character after (?
635      13  POSIX named classes are supported only within a class
636      14  missing )
637      15  reference to non-existent subpattern
638      16  erroffset passed as NULL
639      17  unknown option bit(s) set
640      18  missing ) after comment
641      19  [this code is not in use]
642      20  regular expression too large
643      21  failed to get memory
644      22  unmatched parentheses
645      23  internal error: code overflow
646      24  unrecognized character after (?<
647      25  lookbehind assertion is not fixed length
648      26  malformed number or name after (?(
649      27  conditional group contains more than two branches
650      28  assertion expected after (?(
651      29  (?R or (?[+-]digits must be followed by )
652      30  unknown POSIX class name
653      31  POSIX collating elements are not supported
654      32  this version of PCRE is not compiled with PCRE_UTF8 support
655      33  [this code is not in use]
656      34  character value in \ex{...} sequence is too large
657      35  invalid condition (?(0)
658      36  \eC not allowed in lookbehind assertion
659      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
660      38  number after (?C is > 255
661      39  closing ) for (?C expected
662      40  recursive call could loop indefinitely
663      41  unrecognized character after (?P
664      42  syntax error in subpattern name (missing terminator)
665      43  two named subpatterns have the same name
666      44  invalid UTF-8 string
667      45  support for \eP, \ep, and \eX has not been compiled
668      46  malformed \eP or \ep sequence
669      47  unknown property name after \eP or \ep
670      48  subpattern name is too long (maximum 32 characters)
671      49  too many named subpatterns (maximum 10,000)
672      50  repeated subpattern is too long
673      51  octal value is greater than \e377 (not in UTF-8 mode)
674      52  internal error: overran compiling workspace
675      53  internal error: previously-checked referenced subpattern not found
676      54  DEFINE group contains more than one branch
677      55  repeating a DEFINE group is not allowed
678      56  inconsistent NEWLINE options"
679      57  \eg is not followed by a braced name or an optionally braced
680            non-zero number
681      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
682  .  .
683  .  .
684  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
685  .rs  .rs
686  .sp  .sp
687  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
688  .ti +5n  .ti +5n
689  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
690  .PP  .PP
# Line 492  below Line 706  below
706  .\"  .\"
707  in the section on matching a pattern.  in the section on matching a pattern.
708  .P  .P
709  If studying the pattern does not produce any additional information,  If studying the pattern does not produce any additional information
710  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
711  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
712  own \fBpcre_extra\fP block.  own \fBpcre_extra\fP block.
# Line 502  options are defined, and this argument s Line 716  options are defined, and this argument s
716  .P  .P
717  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
718  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
719  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
720  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
721  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
722    sure that it has run successfully.
723  .P  .P
724  This is a typical call to \fBpcre_study\fP():  This is a typical call to \fBpcre_study\fP():
725  .sp  .sp
# Line 525  bytes is created. Line 740  bytes is created.
740  .sp  .sp
741  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
742  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
743  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
744  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
745  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
746  support.)  support. The use of locales with Unicode is discouraged. If you are handling
747  .P  characters with codes greater than 128, you should either use UTF-8 and
748  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.
749  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  .P
750  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
751  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.
752  default. As more and more applications change to using Unicode, the need for  Normally, the internal tables recognize only ASCII characters. However, when
753  this locale support is expected to die away.  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
754    default "C" locale of the local system, which may cause them to be different.
755    .P
756    The internal tables can always be overridden by tables supplied by the
757    application that calls PCRE. These may be created in a different locale from
758    the default. As more and more applications change to using Unicode, the need
759    for this locale support is expected to die away.
760  .P  .P
761  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
762  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 769  the following code could be used:
769    tables = pcre_maketables();    tables = pcre_maketables();
770    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
771  .sp  .sp
772    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
773    are using Windows, the name for the French locale is "french".
774    .P
775  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
776  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
777  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 818  check against passing an arbitrary memor
818  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
819  .sp  .sp
820    int rc;    int rc;
821    unsigned long int length;    size_t length;
822    rc = pcre_fullinfo(    rc = pcre_fullinfo(
823      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
824      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
# Line 615  no back references. Line 839  no back references.
839  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
840  should point to an \fBint\fP variable.  should point to an \fBint\fP variable.
841  .sp  .sp
842    PCRE_INFO_DEFAULTTABLES    PCRE_INFO_DEFAULT_TABLES
843  .sp  .sp
844  Return a pointer to the internal default character tables within PCRE. The  Return a pointer to the internal default character tables within PCRE. The
845  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 850  a NULL table pointer.
850    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
851  .sp  .sp
852  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
853  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
854  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
855    still recognized for backwards compatibility.)
856  .P  .P
857  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
858  (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  
859  .sp  .sp
860  (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
861  starts with "^", or  starts with "^", or
# Line 650  table indicating a fixed set of bytes fo Line 874  table indicating a fixed set of bytes fo
874  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
875  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
876  .sp  .sp
877      PCRE_INFO_JCHANGED
878    .sp
879    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
880    fourth argument should point to an \fBint\fP variable. The (?J) internal option
881    setting changes the local PCRE_DUPNAMES option.
882    .sp
883    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
884  .sp  .sp
885  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 896  is -1.
896  .sp  .sp
897  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
898  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
899  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
900  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
901  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
902  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
903  \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,
904  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
905    values.
906  .P  .P
907  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
908  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 911  length of the longest name. PCRE_INFO_NA
911  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
912  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
913  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
914  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
915    their parentheses numbers. For example, consider the following pattern (assume
916  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
917  .sp  .sp
918  .\" JOIN  .\" JOIN
919    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
920    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
921  .sp  .sp
922  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
923  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 929  bytes shows in hexadecimal, and undefine
929    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
930  .sp  .sp
931  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
932  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
933  different for each compiled pattern.  different for each compiled pattern.
934  .sp  .sp
935      PCRE_INFO_OKPARTIAL
936    .sp
937    Return 1 if the pattern can be used for partial matching, otherwise 0. The
938    fourth argument should point to an \fBint\fP variable. The
939    .\" HREF
940    \fBpcrepartial\fP
941    .\"
942    documentation lists the restrictions that apply to patterns when partial
943    matching is used.
944    .sp
945    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
946  .sp  .sp
947  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
948  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
949  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
950  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
951    they are the options that will be in force when matching starts. For example,
952    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
953    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
954  .P  .P
955  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
956  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 1005  it is used to pass back information abou
1005  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1006  .  .
1007  .  .
1008  .SH "MATCHING A PATTERN"  .SH "REFERENCE COUNTS"
1009    .rs
1010    .sp
1011    .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1012    .PP
1013    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1014    data block that contains a compiled pattern. It is provided for the benefit of
1015    applications that operate in an object-oriented manner, where different parts
1016    of the application may be using the same compiled pattern, but you want to free
1017    the block when they are all done.
1018    .P
1019    When a pattern is compiled, the reference count field is initialized to zero.
1020    It is changed only by calling this function, whose action is to add the
1021    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1022    function is the new value. However, the value of the count is constrained to
1023    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1024    it is forced to the appropriate limit value.
1025    .P
1026    Except when it is zero, the reference count is not correctly preserved if a
1027    pattern is compiled on one host and then transferred to a host whose byte-order
1028    is different. (This seems a highly unlikely scenario.)
1029    .
1030    .
1031    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1032  .rs  .rs
1033  .sp  .sp
1034  .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 1040  string (see PCRE_INFO_FIRSTBYTE above).
1040  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
1041  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1042  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
1043  \fIextra\fP argument.  \fIextra\fP argument. This function is the main matching facility of the
1044    library, and it operates in a Perl-like manner. For specialist use there is
1045    also an alternative matching function, which is described
1046    .\" HTML <a href="#dfamatch">
1047    .\" </a>
1048    below
1049    .\"
1050    in the section about the \fBpcre_dfa_exec()\fP function.
1051  .P  .P
1052  In most applications, the pattern will have been compiled (and optionally  In most applications, the pattern will have been compiled (and optionally
1053  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 1071  Here is an example of a simple call to \
1071      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1072      0,              /* default options */      0,              /* default options */
1073      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1074      30);            /* number of elements in the vector (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1075  .  .
1076  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1077  .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 1080  Here is an example of a simple call to \
1080  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
1081  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
1082  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
1083  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
1084  follows:  fields (not necessarily in this order):
1085  .sp  .sp
1086    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1087    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1088    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1089      unsigned long int \fImatch_limit_recursion\fP;
1090    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1091    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1092  .sp  .sp
# Line 819  are set. The flag bits are: Line 1095  are set. The flag bits are:
1095  .sp  .sp
1096    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1097    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1098      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1099    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1100    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1101  .sp  .sp
# Line 833  but which have a very large number of po Line 1110  but which have a very large number of po
1110  classic example is the use of nested unlimited repeats.  classic example is the use of nested unlimited repeats.
1111  .P  .P
1112  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1113  (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
1114  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
1115  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
1116  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
1117    string.
1118  .P  .P
1119  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
1120  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
1121  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
1122  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
1123  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
1124  exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1125    .P
1126    The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1127    instead of limiting the total number of times that \fBmatch()\fP is called, it
1128    limits the depth of recursion. The recursion depth is a smaller number than the
1129    total number of calls, because not all calls to \fBmatch()\fP are recursive.
1130    This limit is of use only if it is set smaller than \fImatch_limit\fP.
1131    .P
1132    Limiting the recursion depth limits the amount of stack that can be used, or,
1133    when PCRE has been compiled to use memory on the heap instead of the stack, the
1134    amount of heap memory that can be used.
1135    .P
1136    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1137    built; the default default is the same value as the default for
1138    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1139    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1140    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1141    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1142  .P  .P
1143  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,
1144  which is described in the  which is described in the
# Line 870  documentation for a discussion of saving Line 1165  documentation for a discussion of saving
1165  .rs  .rs
1166  .sp  .sp
1167  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
1168  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,
1169  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1170  .sp  .sp
1171    PCRE_ANCHORED    PCRE_ANCHORED
1172  .sp  .sp
# Line 880  matching position. If a pattern was comp Line 1175  matching position. If a pattern was comp
1175  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
1176  matching time.  matching time.
1177  .sp  .sp
1178      PCRE_NEWLINE_CR
1179      PCRE_NEWLINE_LF
1180      PCRE_NEWLINE_CRLF
1181      PCRE_NEWLINE_ANYCRLF
1182      PCRE_NEWLINE_ANY
1183    .sp
1184    These options override the newline definition that was chosen or defaulted when
1185    the pattern was compiled. For details, see the description of
1186    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1187    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1188    the way the match position is advanced after a match failure for an unanchored
1189    pattern. When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
1190    set, and a match attempt fails when the current position is at a CRLF sequence,
1191    the match position is advanced by two characters instead of one, in other
1192    words, to after the CRLF.
1193    .sp
1194    PCRE_NOTBOL    PCRE_NOTBOL
1195  .sp  .sp
1196  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 1011  is set to the offset of the first charac Line 1322  is set to the offset of the first charac
1322  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
1323  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
1324  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
1325  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
1326  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
1327  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1328  .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.  
1329  .P  .P
1330  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
1331  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 1339  the \fIovector\fP is not big enough to r
1339  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
1340  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1341  .P  .P
1342  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
1343  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1344  \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
1345  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.
1346    .P
1347    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1348    the subject when subpattern \fIn\fP has not been used at all. For example, if
1349    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1350    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1351    happens, both values in the offset pairs corresponding to unused subpatterns
1352    are set to -1.
1353    .P
1354    Offset values that correspond to unused subpatterns at the end of the
1355    expression are also set to -1. For example, if the string "abc" is matched
1356    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1357    return from the function is 2, because the highest used capturing subpattern
1358    number is 1. However, you can refer to the offsets for the second and third
1359    capturing subpatterns if you wish (assuming the vector is large enough, of
1360    course).
1361    .P
1362    Some convenience functions are provided for extracting the captured substrings
1363    as separate strings. These are described below.
1364  .  .
1365  .SS "Return values from \fBpcre_exec()\fP"  .\" HTML <a name="errorlist"></a>
1366    .SS "Error return values from \fBpcre_exec()\fP"
1367  .rs  .rs
1368  .sp  .sp
1369  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 1390  compiled in an environment of one endian
1390  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
1391  not present.  not present.
1392  .sp  .sp
1393    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1394  .sp  .sp
1395  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1396  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 1412  below). It is never returned by \fBpcre_
1412  .sp  .sp
1413    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1414  .sp  .sp
1415  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
1416  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1417  description above.  above.
1418  .sp  .sp
1419    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1420  .sp  .sp
# Line 1112  A string that contains an invalid UTF-8 Line 1434  A string that contains an invalid UTF-8
1434  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
1435  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.
1436  .sp  .sp
1437    PCRE_ERROR_PARTIAL (-12)    PCRE_ERROR_PARTIAL        (-12)
1438  .sp  .sp
1439  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
1440  .\" HREF  .\" HREF
# Line 1120  The subject string did not match, but it Line 1442  The subject string did not match, but it
1442  .\"  .\"
1443  documentation for details of partial matching.  documentation for details of partial matching.
1444  .sp  .sp
1445    PCRE_ERROR_BAD_PARTIAL (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1446  .sp  .sp
1447  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
1448  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 1451  are not supported for partial matching.
1451  .\"  .\"
1452  documentation for details of partial matching.  documentation for details of partial matching.
1453  .sp  .sp
1454    PCRE_ERROR_INTERNAL (-14)    PCRE_ERROR_INTERNAL       (-14)
1455  .sp  .sp
1456  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
1457  in PCRE or by overwriting of the compiled pattern.  in PCRE or by overwriting of the compiled pattern.
1458  .sp  .sp
1459    PCRE_ERROR_BADCOUNT (-15)    PCRE_ERROR_BADCOUNT       (-15)
1460  .sp  .sp
1461  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.
1462    .sp
1463      PCRE_ERROR_RECURSIONLIMIT (-21)
1464    .sp
1465    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1466    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1467    description above.
1468    .sp
1469      PCRE_ERROR_NULLWSLIMIT    (-22)
1470    .sp
1471    When a parenthesized subpattern that can match an empty substring is repeated
1472    with an unbounded upper limit, the subject position at the start of the group
1473    must be remembered, so that a test for an empty string can be made when the end
1474    of the group is reached. The repetition loop is automatically terminated when
1475    an empty string is matched.
1476    .P
1477    Some workspace is required for remembering the starting position. For most
1478    capturing subpatterns, an internal recursive call of the match() function is
1479    used, and in this case, the workspace is taken from the runtime stack. However,
1480    for non-capturing subpatterns, this is not available. Instead, some fixed
1481    workspace (taken from the top-level stack) is used. If it runs out, the
1482    PCRE_ERROR_NULLWSLIMIT error is given. The size of the fixed workspace allows
1483    for up to 1000 repetitions of a non-capturing subpattern that might match an
1484    empty string. If this limit is a problem for you, you can sometimes get round
1485    it by changing a non-capturing subpattern into a capturing one, but that will
1486    increase the amount of runtime stack that is used.
1487    .sp
1488      PCRE_ERROR_BADNEWLINE     (-23)
1489    .sp
1490    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1491    .P
1492    Error numbers -16 to -20 are not used by \fBpcre_exec()\fP.
1493  .  .
1494  .  .
1495  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1148  This error is given if the value of the Line 1501  This error is given if the value of the
1501  .ti +5n  .ti +5n
1502  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
1503  .PP  .PP
 .br  
1504  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1505  .ti +5n  .ti +5n
1506  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1507  .ti +5n  .ti +5n
1508  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
1509  .PP  .PP
 .br  
1510  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
1511  .ti +5n  .ti +5n
1512  .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 1517  Captured substrings can be accessed dire
1517  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1518  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1519  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1520  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1521  has a further zero added on the end, but the result is not, of course,  .P
1522  a C string.  A substring that contains a binary zero is correctly extracted and has a
1523    further zero added on the end, but the result is not, of course, a C string.
1524    However, you can process such a string by referring to the length that is
1525    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1526    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1527    for handling strings containing binary zeros, because the end of the final
1528    string is not independently indicated.
1529  .P  .P
1530  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:
1531  \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 1545  the string is placed in \fIbuffer\fP, wh
1545  \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
1546  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1547  \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
1548  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1549  .sp  .sp
1550    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1551  .sp  .sp
# Line 1204  and builds a list of pointers to them. A Line 1561  and builds a list of pointers to them. A
1561  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
1562  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
1563  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
1564  function is zero if all went well, or  function is zero if all went well, or the error code
1565  .sp  .sp
1566    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1567  .sp  .sp
# Line 1223  a previous call of \fBpcre_get_substring Line 1580  a previous call of \fBpcre_get_substring
1580  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1581  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
1582  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
1583  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1584  \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
1585  provided.  provided.
1586  .  .
# Line 1235  provided. Line 1592  provided.
1592  .ti +5n  .ti +5n
1593  .B const char *\fIname\fP);  .B const char *\fIname\fP);
1594  .PP  .PP
 .br  
1595  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1596  .ti +5n  .ti +5n
1597  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1244  provided. Line 1600  provided.
1600  .ti +5n  .ti +5n
1601  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1602  .PP  .PP
 .br  
1603  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_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 1258  For example, for this pattern Line 1613  For example, for this pattern
1613  .sp  .sp
1614    (a+)b(?<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
1615  .sp  .sp
1616  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
1617  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
1618  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
1619    pattern, and the second is the name. The yield of the function is the
1620  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
1621  that name.  that name.
1622  .P  .P
# Line 1268  Given the number, you can extract the su Line 1624  Given the number, you can extract the su
1624  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
1625  two functions that do the whole job.  two functions that do the whole job.
1626  .P  .P
1627  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
1628  \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
1629  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
1630  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
1631  .P  .P
# Line 1279  pattern. This is needed in order to gain Line 1635  pattern. This is needed in order to gain
1635  translation table.  translation table.
1636  .P  .P
1637  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1638  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1639  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1640    the behaviour may not be what you want (see the next section).
1641    .
1642    .
1643    .SH "DUPLICATE SUBPATTERN NAMES"
1644    .rs
1645    .sp
1646    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1647    .ti +5n
1648    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1649    .PP
1650    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1651    are not required to be unique. Normally, patterns with duplicate names are such
1652    that in any one match, only one of the named subpatterns participates. An
1653    example is shown in the
1654    .\" HREF
1655    \fBpcrepattern\fP
1656    .\"
1657    documentation. When duplicates are present, \fBpcre_copy_named_substring()\fP
1658    and \fBpcre_get_named_substring()\fP return the first substring corresponding
1659    to the given name that is set. If none are set, an empty string is returned.
1660    The \fBpcre_get_stringnumber()\fP function returns one of the numbers that are
1661    associated with the name, but it is not defined which it is.
1662    .sp
1663    If you want to get full details of all captured substrings for a given name,
1664    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1665    argument is the compiled pattern, and the second is the name. The third and
1666    fourth are pointers to variables which are updated by the function. After it
1667    has run, they point to the first and last entries in the name-to-number table
1668    for the given name. The function itself returns the length of each entry, or
1669    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1670    described above in the section entitled \fIInformation about a pattern\fP.
1671    Given all the relevant entries for the name, you can extract each of their
1672    numbers, and hence the captured data, if any.
1673    .
1674    .
1675    .SH "FINDING ALL POSSIBLE MATCHES"
1676    .rs
1677    .sp
1678    The traditional matching function uses a similar algorithm to Perl, which stops
1679    when it finds the first match, starting at a given point in the subject. If you
1680    want to find all possible matches, or the longest possible match, consider
1681    using the alternative matching function (see below) instead. If you cannot use
1682    the alternative function, but still need to find all possible matches, you
1683    can kludge it up by making use of the callout facility, which is described in
1684    the
1685    .\" HREF
1686    \fBpcrecallout\fP
1687    .\"
1688    documentation.
1689    .P
1690    What you have to do is to insert a callout right at the end of the pattern.
1691    When your callout function is called, extract and save the current matched
1692    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1693    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1694    will yield PCRE_ERROR_NOMATCH.
1695    .
1696    .
1697    .\" HTML <a name="dfamatch"></a>
1698    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1699    .rs
1700    .sp
1701    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1702    .ti +5n
1703    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1704    .ti +5n
1705    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1706    .ti +5n
1707    .B int *\fIworkspace\fP, int \fIwscount\fP);
1708    .P
1709    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1710    a compiled pattern, using a matching algorithm that scans the subject string
1711    just once, and does not backtrack. This has different characteristics to the
1712    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1713    patterns are not supported. Nevertheless, there are times when this kind of
1714    matching can be useful. For a discussion of the two matching algorithms, see
1715    the
1716    .\" HREF
1717    \fBpcrematching\fP
1718    .\"
1719    documentation.
1720    .P
1721    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1722    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1723    different way, and this is described below. The other common arguments are used
1724    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1725    here.
1726    .P
1727    The two additional arguments provide workspace for the function. The workspace
1728    vector should contain at least 20 elements. It is used for keeping track of
1729    multiple paths through the pattern tree. More workspace will be needed for
1730    patterns and subjects where there are a lot of potential matches.
1731  .P  .P
1732  .in 0  Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1733  Last updated: 09 September 2004  .sp
1734  .br    int rc;
1735  Copyright (c) 1997-2004 University of Cambridge.    int ovector[10];
1736      int wspace[20];
1737      rc = pcre_dfa_exec(
1738        re,             /* result of pcre_compile() */
1739        NULL,           /* we didn't study the pattern */
1740        "some string",  /* the subject string */
1741        11,             /* the length of the subject string */
1742        0,              /* start at offset 0 in the subject */
1743        0,              /* default options */
1744        ovector,        /* vector of integers for substring information */
1745        10,             /* number of elements (NOT size in bytes) */
1746        wspace,         /* working space vector */
1747        20);            /* number of elements (NOT size in bytes) */
1748    .
1749    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1750    .rs
1751    .sp
1752    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1753    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1754    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1755    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1756    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1757    .sp
1758      PCRE_PARTIAL
1759    .sp
1760    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1761    details are slightly different. When PCRE_PARTIAL is set for
1762    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1763    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1764    complete matches, but there is still at least one matching possibility. The
1765    portion of the string that provided the partial match is set as the first
1766    matching string.
1767    .sp
1768      PCRE_DFA_SHORTEST
1769    .sp
1770    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1771    soon as it has found one match. Because of the way the alternative algorithm
1772    works, this is necessarily the shortest possible match at the first possible
1773    matching point in the subject string.
1774    .sp
1775      PCRE_DFA_RESTART
1776    .sp
1777    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1778    a partial match, it is possible to call it again, with additional subject
1779    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1780    option requests this action; when it is set, the \fIworkspace\fP and
1781    \fIwscount\fP options must reference the same vector as before because data
1782    about the match so far is left in them after a partial match. There is more
1783    discussion of this facility in the
1784    .\" HREF
1785    \fBpcrepartial\fP
1786    .\"
1787    documentation.
1788    .
1789    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1790    .rs
1791    .sp
1792    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1793    substring in the subject. Note, however, that all the matches from one run of
1794    the function start at the same point in the subject. The shorter matches are
1795    all initial substrings of the longer matches. For example, if the pattern
1796    .sp
1797      <.*>
1798    .sp
1799    is matched against the string
1800    .sp
1801      This is <something> <something else> <something further> no more
1802    .sp
1803    the three matched strings are
1804    .sp
1805      <something>
1806      <something> <something else>
1807      <something> <something else> <something further>
1808    .sp
1809    On success, the yield of the function is a number greater than zero, which is
1810    the number of matched substrings. The substrings themselves are returned in
1811    \fIovector\fP. Each string uses two elements; the first is the offset to the
1812    start, and the second is the offset to the end. In fact, all the strings have
1813    the same start offset. (Space could have been saved by giving this only once,
1814    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1815    returns data, even though the meaning of the strings is different.)
1816    .P
1817    The strings are returned in reverse order of length; that is, the longest
1818    matching string is given first. If there were too many matches to fit into
1819    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1820    the longest matches.
1821    .
1822    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1823    .rs
1824    .sp
1825    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1826    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1827    described
1828    .\" HTML <a href="#errorlist">
1829    .\" </a>
1830    above.
1831    .\"
1832    There are in addition the following errors that are specific to
1833    \fBpcre_dfa_exec()\fP:
1834    .sp
1835      PCRE_ERROR_DFA_UITEM      (-16)
1836    .sp
1837    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1838    that it does not support, for instance, the use of \eC or a back reference.
1839    .sp
1840      PCRE_ERROR_DFA_UCOND      (-17)
1841    .sp
1842    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1843    uses a back reference for the condition, or a test for recursion in a specific
1844    group. These are not supported.
1845    .sp
1846      PCRE_ERROR_DFA_UMLIMIT    (-18)
1847    .sp
1848    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1849    block that contains a setting of the \fImatch_limit\fP field. This is not
1850    supported (it is meaningless).
1851    .sp
1852      PCRE_ERROR_DFA_WSSIZE     (-19)
1853    .sp
1854    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1855    \fIworkspace\fP vector.
1856    .sp
1857      PCRE_ERROR_DFA_RECURSE    (-20)
1858    .sp
1859    When a recursive subpattern is processed, the matching function calls itself
1860    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1861    error is given if the output vector is not large enough. This should be
1862    extremely rare, as a vector of size 1000 is used.
1863    .
1864    .
1865    .SH "SEE ALSO"
1866    .rs
1867    .sp
1868    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1869    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1870    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1871    .
1872    .
1873    .SH AUTHOR
1874    .rs
1875    .sp
1876    .nf
1877    Philip Hazel
1878    University Computing Service
1879    Cambridge CB2 3QH, England.
1880    .fi
1881    .
1882    .
1883    .SH REVISION
1884    .rs
1885    .sp
1886    .nf
1887    Last updated: 30 July 2007
1888    Copyright (c) 1997-2007 University of Cambridge.
1889    .fi

Legend:
Removed from v.75  
changed lines
  Added in v.196

  ViewVC Help
Powered by ViewVC 1.1.5