/[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 243 by ph10, Thu Sep 13 09:28:14 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    .\" HTML <a name="newlines"></a>
222    .SH NEWLINES
223    .rs
224    .sp
225    PCRE supports five different conventions for indicating line breaks in
226    strings: a single CR (carriage return) character, a single LF (linefeed)
227    character, the two-character sequence CRLF, any of the three preceding, or any
228    Unicode newline sequence. The Unicode newline sequences are the three just
229    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
230    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
231    (paragraph separator, U+2029).
232    .P
233    Each of the first three conventions is used by at least one operating system as
234    its standard newline sequence. When PCRE is built, a default can be specified.
235    The default default is LF, which is the Unix standard. When PCRE is run, the
236    default can be overridden, either when a pattern is compiled, or when it is
237    matched.
238    .P
239    At compile time, the newline convention can be specified by the \fIoptions\fP
240    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
241    start of the pattern itself; this overrides any other settings. See the
242    .\" HREF
243    \fBpcrepattern\fP
244    .\"
245    page for details of the special character sequences.
246    .P
247    In the PCRE documentation the word "newline" is used to mean "the character or
248    pair of characters that indicate a line break". The choice of newline
249    convention affects the handling of the dot, circumflex, and dollar
250    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
251    recognized line ending sequence, the match position advancement for a
252    non-anchored pattern. There is more detail about this in the
253    .\" HTML <a href="#execoptions">
254    .\" </a>
255    section on \fBpcre_exec()\fP options
256    .\"
257    below.
258    .P
259    The choice of newline convention does not affect the interpretation of
260    the \en or \er escape sequences, nor does it affect what \eR matches, which is
261    controlled in a similar way, but by separate options.
262    .
263    .
264  .SH MULTITHREADING  .SH MULTITHREADING
265  .rs  .rs
266  .sp  .sp
# Line 207  which it was compiled. Details are given Line 282  which it was compiled. Details are given
282  .\" HREF  .\" HREF
283  \fBpcreprecompile\fP  \fBpcreprecompile\fP
284  .\"  .\"
285  documentation.  documentation. However, compiling a regular expression with one version of PCRE
286    for use with a different version is not guaranteed to work and may cause
287    crashes.
288  .  .
289  .  .
290  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 238  properties is available; otherwise it is Line 315  properties is available; otherwise it is
315  .sp  .sp
316    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
317  .sp  .sp
318  The output is an integer that is set to the value of the code that is used for  The output is an integer whose value specifies the default character sequence
319  the newline character. It is either linefeed (10) or carriage return (13), and  that is recognized as meaning "newline". The four values that are supported
320  should normally be the standard character for your operating system.  are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF, and -1 for ANY. The
321    default should normally be the standard sequence for your operating system.
322    .sp
323      PCRE_CONFIG_BSR
324    .sp
325    The output is an integer whose value indicates what character sequences the \eR
326    escape sequence matches by default. A value of 0 means that \eR matches any
327    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
328    or CRLF. The default can be overridden when a pattern is compiled or matched.
329  .sp  .sp
330    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
331  .sp  .sp
# Line 266  The output is an integer that gives the Line 351  The output is an integer that gives the
351  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
352  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
353  .sp  .sp
354      PCRE_CONFIG_MATCH_LIMIT_RECURSION
355    .sp
356    The output is an integer that gives the default limit for the depth of
357    recursion when calling the internal matching function in a \fBpcre_exec()\fP
358    execution. Further details are given with \fBpcre_exec()\fP below.
359    .sp
360    PCRE_CONFIG_STACKRECURSE    PCRE_CONFIG_STACKRECURSE
361  .sp  .sp
362  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
363  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
364  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
365  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
366  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
367  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
368    avoiding the use of the stack.
369  .  .
370  .  .
371  .SH "COMPILING A PATTERN"  .SH "COMPILING A PATTERN"
# Line 284  called to manage memory blocks on the he Line 376  called to manage memory blocks on the he
376  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
377  .ti +5n  .ti +5n
378  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
379    .sp
380    .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
381    .ti +5n
382    .B int *\fIerrorcodeptr\fP,
383    .ti +5n
384    .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
385    .ti +5n
386    .B const unsigned char *\fItableptr\fP);
387  .P  .P
388  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
389  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
390  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,
391  that is obtained via \fBpcre_malloc\fP is returned. This contains the compiled  \fIerrorcodeptr\fP, via which a numerical error code can be returned.
392  code and related data. The \fBpcre\fP type is defined for the returned block;  .P
393  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
394  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
395    via \fBpcre_malloc\fP is returned. This contains the compiled code and related
396    data. The \fBpcre\fP type is defined for the returned block; this is a typedef
397    for a structure whose contents are not externally defined. It is up to the
398    caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
399  .P  .P
400  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
401  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
402  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
403  argument, which is an address (see below).  argument, which is an address (see below).
404  .P  .P
405  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
406  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
407  options are described below. Some of them, in particular, those that are  options are described below. Some of them, in particular, those that are
408  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 412  the detailed description in the
412  .\"  .\"
413  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For these options, the contents of the \fIoptions\fP argument
414  specifies their initial settings at the start of compilation and execution. The  specifies their initial settings at the start of compilation and execution. The
415  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
416  time.  matching as well as at compile time.
417  .P  .P
418  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
419  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
420  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
421  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
422  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
423    where the error was discovered is placed in the variable pointed to by
424  \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.
425  .P  .P
426    If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
427    \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
428    returned via this argument in the event of an error. This is in addition to the
429    textual error message. Error codes and messages are listed below.
430    .P
431  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
432  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
433  locale. Otherwise, \fItableptr\fP must be an address that is the result of a  locale. Otherwise, \fItableptr\fP must be an address that is the result of a
# Line 358  facility, see the Line 468  facility, see the
468  .\"  .\"
469  documentation.  documentation.
470  .sp  .sp
471      PCRE_BSR_ANYCRLF
472      PCRE_BSR_UNICODE
473    .sp
474    These options (which are mutually exclusive) control what the \eR escape
475    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
476    match any Unicode newline sequence. The default is specified when PCRE is
477    built. It can be overridden from within the pattern, or by setting an option
478    when a compiled pattern is matched.
479    .sp
480    PCRE_CASELESS    PCRE_CASELESS
481  .sp  .sp
482  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
483  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
484  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
485  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
486  character property support.  matching is always possible. For characters with higher values, the concept of
487    case is supported if PCRE is compiled with Unicode property support, but not
488    otherwise. If you want to use caseless matching for characters 128 and above,
489    you must ensure that PCRE is compiled with Unicode property support as well as
490    with UTF-8 support.
491  .sp  .sp
492    PCRE_DOLLAR_ENDONLY    PCRE_DOLLAR_ENDONLY
493  .sp  .sp
494  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
495  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
496  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
497  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.
498  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
499  a pattern.  pattern.
500  .sp  .sp
501    PCRE_DOTALL    PCRE_DOTALL
502  .sp  .sp
503  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,
504  including newlines. Without it, newlines are excluded. This option is  including those that indicate newline. Without it, a dot does not match when
505  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
506  (?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
507  character, independent of the setting of this option.  negative class such as [^a] always matches newline characters, independent of
508    the setting of this option.
509    .sp
510      PCRE_DUPNAMES
511    .sp
512    If this bit is set, names used to identify capturing subpatterns need not be
513    unique. This can be helpful for certain types of pattern when it is known that
514    only one instance of the named subpattern can ever be matched. There are more
515    details of named subpatterns below; see also the
516    .\" HREF
517    \fBpcrepattern\fP
518    .\"
519    documentation.
520  .sp  .sp
521    PCRE_EXTENDED    PCRE_EXTENDED
522  .sp  .sp
523  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
524  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
525  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
526  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
527  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
528  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
529  .P  .P
530  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
531  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 539  that is incompatible with Perl, but it i
539  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
540  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
541  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
542  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
543  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
544  pattern.  this option. It can also be set by a (?X) option setting within a pattern.
545    .sp
546      PCRE_FIRSTLINE
547    .sp
548    If this option is set, an unanchored pattern is required to match before or at
549    the first newline in the subject string, though the matched text may continue
550    over the newline.
551  .sp  .sp
552    PCRE_MULTILINE    PCRE_MULTILINE
553  .sp  .sp
# Line 418  terminating newline (unless PCRE_DOLLAR_ Line 559  terminating newline (unless PCRE_DOLLAR_
559  Perl.  Perl.
560  .P  .P
561  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
562  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
563  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
564  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
565  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
566  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
567  .sp  .sp
568      PCRE_NEWLINE_CR
569      PCRE_NEWLINE_LF
570      PCRE_NEWLINE_CRLF
571      PCRE_NEWLINE_ANYCRLF
572      PCRE_NEWLINE_ANY
573    .sp
574    These options override the default newline definition that was chosen when PCRE
575    was built. Setting the first or the second specifies that a newline is
576    indicated by a single character (CR or LF, respectively). Setting
577    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
578    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
579    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
580    that any Unicode newline sequence should be recognized. The Unicode newline
581    sequences are the three just mentioned, plus the single characters VT (vertical
582    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
583    separator, U+2028), and PS (paragraph separator, U+2029). The last two are
584    recognized only in UTF-8 mode.
585    .P
586    The newline setting in the options word uses three bits that are treated
587    as a number, giving eight possibilities. Currently only six are used (default
588    plus the five values above). This means that if you set more than one newline
589    option, the combination may or may not be sensible. For example,
590    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
591    other combinations may yield unused numbers and cause an error.
592    .P
593    The only time that a line break is specially recognized when compiling a
594    pattern is if PCRE_EXTENDED is set, and an unescaped # outside a character
595    class is encountered. This indicates a comment that lasts until after the next
596    line break sequence. In other circumstances, line break sequences are treated
597    as literal data, except that in PCRE_EXTENDED mode, both CR and LF are treated
598    as whitespace characters and are therefore ignored.
599    .P
600    The newline option that is set at compile time becomes the default that is used
601    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
602    .sp
603    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
604  .sp  .sp
605  If this option is set, it disables the use of numbered capturing parentheses in  If this option is set, it disables the use of numbered capturing parentheses in
# Line 458  page. Line 634  page.
634    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
635  .sp  .sp
636  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is  When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
637  automatically checked. If an invalid UTF-8 sequence of bytes is found,  automatically checked. There is a discussion about the
638  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcre.html#utf8strings">
639  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
640  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
641  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
642  Note that this option can also be passed to \fBpcre_exec()\fP, to suppress the  in the main
643  UTF-8 validity checking of subject strings.  .\" HREF
644    \fBpcre\fP
645    .\"
646    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_compile()\fP
647    returns an error. If you already know that your pattern is valid, and you want
648    to skip this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK
649    option. When it is set, the effect of passing an invalid UTF-8 string as a
650    pattern is undefined. It may cause your program to crash. Note that this option
651    can also be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress
652    the UTF-8 validity checking of subject strings.
653    .
654    .
655    .SH "COMPILATION ERROR CODES"
656    .rs
657    .sp
658    The following table lists the error codes than may be returned by
659    \fBpcre_compile2()\fP, along with the error messages that may be returned by
660    both compiling functions. As PCRE has developed, some error codes have fallen
661    out of use. To avoid confusion, they have not been re-used.
662    .sp
663       0  no error
664       1  \e at end of pattern
665       2  \ec at end of pattern
666       3  unrecognized character follows \e
667       4  numbers out of order in {} quantifier
668       5  number too big in {} quantifier
669       6  missing terminating ] for character class
670       7  invalid escape sequence in character class
671       8  range out of order in character class
672       9  nothing to repeat
673      10  [this code is not in use]
674      11  internal error: unexpected repeat
675      12  unrecognized character after (?
676      13  POSIX named classes are supported only within a class
677      14  missing )
678      15  reference to non-existent subpattern
679      16  erroffset passed as NULL
680      17  unknown option bit(s) set
681      18  missing ) after comment
682      19  [this code is not in use]
683      20  regular expression too large
684      21  failed to get memory
685      22  unmatched parentheses
686      23  internal error: code overflow
687      24  unrecognized character after (?<
688      25  lookbehind assertion is not fixed length
689      26  malformed number or name after (?(
690      27  conditional group contains more than two branches
691      28  assertion expected after (?(
692      29  (?R or (?[+-]digits must be followed by )
693      30  unknown POSIX class name
694      31  POSIX collating elements are not supported
695      32  this version of PCRE is not compiled with PCRE_UTF8 support
696      33  [this code is not in use]
697      34  character value in \ex{...} sequence is too large
698      35  invalid condition (?(0)
699      36  \eC not allowed in lookbehind assertion
700      37  PCRE does not support \eL, \el, \eN, \eU, or \eu
701      38  number after (?C is > 255
702      39  closing ) for (?C expected
703      40  recursive call could loop indefinitely
704      41  unrecognized character after (?P
705      42  syntax error in subpattern name (missing terminator)
706      43  two named subpatterns have the same name
707      44  invalid UTF-8 string
708      45  support for \eP, \ep, and \eX has not been compiled
709      46  malformed \eP or \ep sequence
710      47  unknown property name after \eP or \ep
711      48  subpattern name is too long (maximum 32 characters)
712      49  too many named subpatterns (maximum 10,000)
713      50  [this code is not in use]
714      51  octal value is greater than \e377 (not in UTF-8 mode)
715      52  internal error: overran compiling workspace
716      53  internal error: previously-checked referenced subpattern not found
717      54  DEFINE group contains more than one branch
718      55  repeating a DEFINE group is not allowed
719      56  inconsistent NEWLINE options
720      57  \eg is not followed by a braced name or an optionally braced
721            non-zero number
722      58  (?+ or (?- or (?(+ or (?(- must be followed by a non-zero number
723  .  .
724  .  .
725  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
726  .rs  .rs
727  .sp  .sp
728  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP
729  .ti +5n  .ti +5n
730  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
731  .PP  .PP
# Line 492  below Line 747  below
747  .\"  .\"
748  in the section on matching a pattern.  in the section on matching a pattern.
749  .P  .P
750  If studying the pattern does not produce any additional information,  If studying the pattern does not produce any additional information
751  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
752  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
753  own \fBpcre_extra\fP block.  own \fBpcre_extra\fP block.
# Line 502  options are defined, and this argument s Line 757  options are defined, and this argument s
757  .P  .P
758  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
759  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
760  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
761  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
762  be sure that it has run successfully.  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
763    sure that it has run successfully.
764  .P  .P
765  This is a typical call to \fBpcre_study\fP():  This is a typical call to \fBpcre_study\fP():
766  .sp  .sp
# Line 525  bytes is created. Line 781  bytes is created.
781  .sp  .sp
782  PCRE handles caseless matching, and determines whether characters are letters,  PCRE handles caseless matching, and determines whether characters are letters,
783  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
784  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
785  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
786  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
787  support.)  support. The use of locales with Unicode is discouraged. If you are handling
788  .P  characters with codes greater than 128, you should either use UTF-8 and
789  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.
790  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  .P
791  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
792  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.
793  default. As more and more applications change to using Unicode, the need for  Normally, the internal tables recognize only ASCII characters. However, when
794  this locale support is expected to die away.  PCRE is built, it is possible to cause the internal tables to be rebuilt in the
795    default "C" locale of the local system, which may cause them to be different.
796    .P
797    The internal tables can always be overridden by tables supplied by the
798    application that calls PCRE. These may be created in a different locale from
799    the default. As more and more applications change to using Unicode, the need
800    for this locale support is expected to die away.
801  .P  .P
802  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
803  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 810  the following code could be used:
810    tables = pcre_maketables();    tables = pcre_maketables();
811    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
812  .sp  .sp
813    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
814    are using Windows, the name for the French locale is "french".
815    .P
816  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
817  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
818  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 859  check against passing an arbitrary memor
859  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:
860  .sp  .sp
861    int rc;    int rc;
862    unsigned long int length;    size_t length;
863    rc = pcre_fullinfo(    rc = pcre_fullinfo(
864      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
865      pe,               /* result of pcre_study(), or NULL */      pe,               /* result of pcre_study(), or NULL */
# Line 615  no back references. Line 880  no back references.
880  Return the number of capturing subpatterns in the pattern. The fourth argument  Return the number of capturing subpatterns in the pattern. The fourth argument
881  should point to an \fBint\fP variable.  should point to an \fBint\fP variable.
882  .sp  .sp
883    PCRE_INFO_DEFAULTTABLES    PCRE_INFO_DEFAULT_TABLES
884  .sp  .sp
885  Return a pointer to the internal default character tables within PCRE. The  Return a pointer to the internal default character tables within PCRE. The
886  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 891  a NULL table pointer.
891    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
892  .sp  .sp
893  Return information about the first byte of any matched string, for a  Return information about the first byte of any matched string, for a
894  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
895  old name is still recognized for backwards compatibility.)  variable. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name is
896    still recognized for backwards compatibility.)
897  .P  .P
898  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
899  (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  
900  .sp  .sp
901  (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
902  starts with "^", or  starts with "^", or
# Line 650  table indicating a fixed set of bytes fo Line 915  table indicating a fixed set of bytes fo
915  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
916  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
917  .sp  .sp
918      PCRE_INFO_HASCRORLF
919    .sp
920    Return 1 if the pattern contains any explicit matches for CR or LF characters,
921    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
922    explicit match is either a literal CR or LF character, or \er or \en.
923    .sp
924      PCRE_INFO_JCHANGED
925    .sp
926    Return 1 if the (?J) option setting is used in the pattern, otherwise 0. The
927    fourth argument should point to an \fBint\fP variable. The (?J) internal option
928    setting changes the local PCRE_DUPNAMES option.
929    .sp
930    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
931  .sp  .sp
932  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 943  is -1.
943  .sp  .sp
944  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
945  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
946  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
947  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
948  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
949  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
950  \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,
951  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
952    values.
953  .P  .P
954  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
955  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 958  length of the longest name. PCRE_INFO_NA
958  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
959  are the number of the capturing parenthesis, most significant byte first. The  are the number of the capturing parenthesis, most significant byte first. The
960  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
961  alphabetical order. For example, consider the following pattern (assume  alphabetical order. When PCRE_DUPNAMES is set, duplicate names are in order of
962    their parentheses numbers. For example, consider the following pattern (assume
963  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  PCRE_EXTENDED is set, so white space - including newlines - is ignored):
964  .sp  .sp
965  .\" JOIN  .\" JOIN
966    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
967    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
968  .sp  .sp
969  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
970  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 976  bytes shows in hexadecimal, and undefine
976    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
977  .sp  .sp
978  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
979  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
980  different for each compiled pattern.  different for each compiled pattern.
981  .sp  .sp
982      PCRE_INFO_OKPARTIAL
983    .sp
984    Return 1 if the pattern can be used for partial matching, otherwise 0. The
985    fourth argument should point to an \fBint\fP variable. The
986    .\" HREF
987    \fBpcrepartial\fP
988    .\"
989    documentation lists the restrictions that apply to patterns when partial
990    matching is used.
991    .sp
992    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
993  .sp  .sp
994  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
995  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
996  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
997  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
998    they are the options that will be in force when matching starts. For example,
999    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1000    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1001  .P  .P
1002  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
1003  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 1052  it is used to pass back information abou
1052  string (see PCRE_INFO_FIRSTBYTE above).  string (see PCRE_INFO_FIRSTBYTE above).
1053  .  .
1054  .  .
1055  .SH "MATCHING A PATTERN"  .SH "REFERENCE COUNTS"
1056    .rs
1057    .sp
1058    .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
1059    .PP
1060    The \fBpcre_refcount()\fP function is used to maintain a reference count in the
1061    data block that contains a compiled pattern. It is provided for the benefit of
1062    applications that operate in an object-oriented manner, where different parts
1063    of the application may be using the same compiled pattern, but you want to free
1064    the block when they are all done.
1065    .P
1066    When a pattern is compiled, the reference count field is initialized to zero.
1067    It is changed only by calling this function, whose action is to add the
1068    \fIadjust\fP value (which may be positive or negative) to it. The yield of the
1069    function is the new value. However, the value of the count is constrained to
1070    lie between 0 and 65535, inclusive. If the new value is outside these limits,
1071    it is forced to the appropriate limit value.
1072    .P
1073    Except when it is zero, the reference count is not correctly preserved if a
1074    pattern is compiled on one host and then transferred to a host whose byte-order
1075    is different. (This seems a highly unlikely scenario.)
1076    .
1077    .
1078    .SH "MATCHING A PATTERN: THE TRADITIONAL FUNCTION"
1079  .rs  .rs
1080  .sp  .sp
1081  .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 1087  string (see PCRE_INFO_FIRSTBYTE above).
1087  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
1088  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1089  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
1090  \fIextra\fP argument.  \fIextra\fP argument. This function is the main matching facility of the
1091    library, and it operates in a Perl-like manner. For specialist use there is
1092    also an alternative matching function, which is described
1093    .\" HTML <a href="#dfamatch">
1094    .\" </a>
1095    below
1096    .\"
1097    in the section about the \fBpcre_dfa_exec()\fP function.
1098  .P  .P
1099  In most applications, the pattern will have been compiled (and optionally  In most applications, the pattern will have been compiled (and optionally
1100  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 1118  Here is an example of a simple call to \
1118      0,              /* start at offset 0 in the subject */      0,              /* start at offset 0 in the subject */
1119      0,              /* default options */      0,              /* default options */
1120      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1121      30);            /* number of elements in the vector (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1122  .  .
1123  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1124  .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 1127  Here is an example of a simple call to \
1127  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
1128  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
1129  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
1130  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
1131  follows:  fields (not necessarily in this order):
1132  .sp  .sp
1133    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1134    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1135    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1136      unsigned long int \fImatch_limit_recursion\fP;
1137    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1138    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1139  .sp  .sp
# Line 819  are set. The flag bits are: Line 1142  are set. The flag bits are:
1142  .sp  .sp
1143    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1144    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1145      PCRE_EXTRA_MATCH_LIMIT_RECURSION
1146    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1147    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1148  .sp  .sp
# Line 833  but which have a very large number of po Line 1157  but which have a very large number of po
1157  classic example is the use of nested unlimited repeats.  classic example is the use of nested unlimited repeats.
1158  .P  .P
1159  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly
1160  (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
1161  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
1162  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
1163  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
1164    string.
1165  .P  .P
1166  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
1167  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
1168  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
1169  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
1170  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
1171  exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_MATCHLIMIT.  PCRE_ERROR_MATCHLIMIT.
1172    .P
1173    The \fImatch_limit_recursion\fP field is similar to \fImatch_limit\fP, but
1174    instead of limiting the total number of times that \fBmatch()\fP is called, it
1175    limits the depth of recursion. The recursion depth is a smaller number than the
1176    total number of calls, because not all calls to \fBmatch()\fP are recursive.
1177    This limit is of use only if it is set smaller than \fImatch_limit\fP.
1178    .P
1179    Limiting the recursion depth limits the amount of stack that can be used, or,
1180    when PCRE has been compiled to use memory on the heap instead of the stack, the
1181    amount of heap memory that can be used.
1182    .P
1183    The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1184    built; the default default is the same value as the default for
1185    \fImatch_limit\fP. You can override the default by suppling \fBpcre_exec()\fP
1186    with a \fBpcre_extra\fP block in which \fImatch_limit_recursion\fP is set, and
1187    PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1188    is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1189  .P  .P
1190  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,
1191  which is described in the  which is described in the
# Line 866  called. See the Line 1208  called. See the
1208  .\"  .\"
1209  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1210  .  .
1211    .\" HTML <a name="execoptions"></a>
1212  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
1213  .rs  .rs
1214  .sp  .sp
1215  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
1216  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,
1217  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.
1218  .sp  .sp
1219    PCRE_ANCHORED    PCRE_ANCHORED
1220  .sp  .sp
# Line 880  matching position. If a pattern was comp Line 1223  matching position. If a pattern was comp
1223  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
1224  matching time.  matching time.
1225  .sp  .sp
1226      PCRE_BSR_ANYCRLF
1227      PCRE_BSR_UNICODE
1228    .sp
1229    These options (which are mutually exclusive) control what the \eR escape
1230    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1231    match any Unicode newline sequence. These options override the choice that was
1232    made or defaulted when the pattern was compiled.
1233    .sp
1234      PCRE_NEWLINE_CR
1235      PCRE_NEWLINE_LF
1236      PCRE_NEWLINE_CRLF
1237      PCRE_NEWLINE_ANYCRLF
1238      PCRE_NEWLINE_ANY
1239    .sp
1240    These options override the newline definition that was chosen or defaulted when
1241    the pattern was compiled. For details, see the description of
1242    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1243    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1244    the way the match position is advanced after a match failure for an unanchored
1245    pattern.
1246    .P
1247    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1248    match attempt for an unanchored pattern fails when the current position is at a
1249    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1250    characters, the match position is advanced by two characters instead of one, in
1251    other words, to after the CRLF.
1252    .P
1253    The above rule is a compromise that makes the most common cases work as
1254    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1255    set), it does not match the string "\er\enA" because, after failing at the
1256    start, it skips both the CR and the LF before retrying. However, the pattern
1257    [\er\en]A does match that string, because it contains an explicit CR or LF
1258    reference, and so advances only by one character after the first failure.
1259    .P
1260    An explicit match for CR of LF is either a literal appearance of one of those
1261    characters, or one of the \er or \en escape sequences. Implicit matches such as
1262    [^X] do not count, nor does \es (which includes CR and LF in the characters
1263    that it matches).
1264    .P
1265    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1266    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1267    .sp
1268    PCRE_NOTBOL    PCRE_NOTBOL
1269  .sp  .sp
1270  This option specifies that first character of the subject string is not the  This option specifies that first character of the subject string is not the
# Line 921  code that demonstrates how to do this in Line 1306  code that demonstrates how to do this in
1306  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8  When PCRE_UTF8 is set at compile time, the validity of the subject as a UTF-8
1307  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1308  The value of \fIstartoffset\fP is also checked to ensure that it points to the  The value of \fIstartoffset\fP is also checked to ensure that it points to the
1309  start of a UTF-8 character. If an invalid UTF-8 sequence of bytes is found,  start of a UTF-8 character. There is a discussion about the validity of UTF-8
1310  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  strings in the
1311  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HTML <a href="pcre.html#utf8strings">
1312    .\" </a>
1313    section on UTF-8 support
1314    .\"
1315    in the main
1316    .\" HREF
1317    \fBpcre\fP
1318    .\"
1319    page. If an invalid UTF-8 sequence of bytes is found, \fBpcre_exec()\fP returns
1320    the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP contains an invalid value,
1321    PCRE_ERROR_BADUTF8_OFFSET is returned.
1322  .P  .P
1323  If you already know that your subject is valid, and you want to skip these  If you already know that your subject is valid, and you want to skip these
1324  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when  checks for performance reasons, you can set the PCRE_NO_UTF8_CHECK option when
# Line 1011  is set to the offset of the first charac Line 1406  is set to the offset of the first charac
1406  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
1407  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
1408  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
1409  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
1410  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
1411  just the first pair of offsets has been set.  capturing subpatterns, the return value from a successful match is 1,
1412  .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.  
1413  .P  .P
1414  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
1415  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 1423  the \fIovector\fP is not big enough to r
1423  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
1424  advisable to supply an \fIovector\fP.  advisable to supply an \fIovector\fP.
1425  .P  .P
1426  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
1427  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1428  \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
1429  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.
1430    .P
1431    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1432    the subject when subpattern \fIn\fP has not been used at all. For example, if
1433    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1434    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1435    happens, both values in the offset pairs corresponding to unused subpatterns
1436    are set to -1.
1437    .P
1438    Offset values that correspond to unused subpatterns at the end of the
1439    expression are also set to -1. For example, if the string "abc" is matched
1440    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1441    return from the function is 2, because the highest used capturing subpattern
1442    number is 1. However, you can refer to the offsets for the second and third
1443    capturing subpatterns if you wish (assuming the vector is large enough, of
1444    course).
1445    .P
1446    Some convenience functions are provided for extracting the captured substrings
1447    as separate strings. These are described below.
1448  .  .
1449  .SS "Return values from \fBpcre_exec()\fP"  .\" HTML <a name="errorlist"></a>
1450    .SS "Error return values from \fBpcre_exec()\fP"
1451  .rs  .rs
1452  .sp  .sp
1453  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 1474  compiled in an environment of one endian
1474  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
1475  not present.  not present.
1476  .sp  .sp
1477    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1478  .sp  .sp
1479  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1480  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 1496  below). It is never returned by \fBpcre_
1496  .sp  .sp
1497    PCRE_ERROR_MATCHLIMIT     (-8)    PCRE_ERROR_MATCHLIMIT     (-8)
1498  .sp  .sp
1499  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
1500  field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1501  description above.  above.
1502  .sp  .sp
1503    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1504  .sp  .sp
# Line 1112  A string that contains an invalid UTF-8 Line 1518  A string that contains an invalid UTF-8
1518  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
1519  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.
1520  .sp  .sp
1521    PCRE_ERROR_PARTIAL (-12)    PCRE_ERROR_PARTIAL        (-12)
1522  .sp  .sp
1523  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
1524  .\" HREF  .\" HREF
# Line 1120  The subject string did not match, but it Line 1526  The subject string did not match, but it
1526  .\"  .\"
1527  documentation for details of partial matching.  documentation for details of partial matching.
1528  .sp  .sp
1529    PCRE_ERROR_BAD_PARTIAL (-13)    PCRE_ERROR_BADPARTIAL     (-13)
1530  .sp  .sp
1531  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
1532  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 1535  are not supported for partial matching.
1535  .\"  .\"
1536  documentation for details of partial matching.  documentation for details of partial matching.
1537  .sp  .sp
1538    PCRE_ERROR_INTERNAL (-14)    PCRE_ERROR_INTERNAL       (-14)
1539  .sp  .sp
1540  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
1541  in PCRE or by overwriting of the compiled pattern.  in PCRE or by overwriting of the compiled pattern.
1542  .sp  .sp
1543    PCRE_ERROR_BADCOUNT (-15)    PCRE_ERROR_BADCOUNT       (-15)
1544  .sp  .sp
1545  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.
1546    .sp
1547      PCRE_ERROR_RECURSIONLIMIT (-21)
1548    .sp
1549    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
1550    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
1551    description above.
1552    .sp
1553      PCRE_ERROR_BADNEWLINE     (-23)
1554    .sp
1555    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
1556    .P
1557    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
1558  .  .
1559  .  .
1560  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1148  This error is given if the value of the Line 1566  This error is given if the value of the
1566  .ti +5n  .ti +5n
1567  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
1568  .PP  .PP
 .br  
1569  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
1570  .ti +5n  .ti +5n
1571  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
1572  .ti +5n  .ti +5n
1573  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
1574  .PP  .PP
 .br  
1575  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
1576  .ti +5n  .ti +5n
1577  .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 1582  Captured substrings can be accessed dire
1582  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
1583  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
1584  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
1585  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
1586  has a further zero added on the end, but the result is not, of course,  .P
1587  a C string.  A substring that contains a binary zero is correctly extracted and has a
1588    further zero added on the end, but the result is not, of course, a C string.
1589    However, you can process such a string by referring to the length that is
1590    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
1591    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
1592    for handling strings containing binary zeros, because the end of the final
1593    string is not independently indicated.
1594  .P  .P
1595  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:
1596  \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 1610  the string is placed in \fIbuffer\fP, wh
1610  \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
1611  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
1612  \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
1613  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
1614  .sp  .sp
1615    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1616  .sp  .sp
# Line 1204  and builds a list of pointers to them. A Line 1626  and builds a list of pointers to them. A
1626  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
1627  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
1628  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
1629  function is zero if all went well, or  function is zero if all went well, or the error code
1630  .sp  .sp
1631    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
1632  .sp  .sp
# Line 1223  a previous call of \fBpcre_get_substring Line 1645  a previous call of \fBpcre_get_substring
1645  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
1646  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
1647  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
1648  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
1649  \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
1650  provided.  provided.
1651  .  .
# Line 1235  provided. Line 1657  provided.
1657  .ti +5n  .ti +5n
1658  .B const char *\fIname\fP);  .B const char *\fIname\fP);
1659  .PP  .PP
 .br  
1660  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
1661  .ti +5n  .ti +5n
1662  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1244  provided. Line 1665  provided.
1665  .ti +5n  .ti +5n
1666  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
1667  .PP  .PP
 .br  
1668  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
1669  .ti +5n  .ti +5n
1670  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1258  For example, for this pattern Line 1678  For example, for this pattern
1678  .sp  .sp
1679    (a+)b(?<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
1680  .sp  .sp
1681  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
1682  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
1683  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
1684    pattern, and the second is the name. The yield of the function is the
1685  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
1686  that name.  that name.
1687  .P  .P
# Line 1268  Given the number, you can extract the su Line 1689  Given the number, you can extract the su
1689  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
1690  two functions that do the whole job.  two functions that do the whole job.
1691  .P  .P
1692  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
1693  \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
1694  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
1695  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
1696  .P  .P
# Line 1279  pattern. This is needed in order to gain Line 1700  pattern. This is needed in order to gain
1700  translation table.  translation table.
1701  .P  .P
1702  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
1703  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
1704  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
1705    the behaviour may not be what you want (see the next section).
1706    .
1707    .
1708    .SH "DUPLICATE SUBPATTERN NAMES"
1709    .rs
1710    .sp
1711    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
1712    .ti +5n
1713    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
1714    .PP
1715    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
1716    are not required to be unique. Normally, patterns with duplicate names are such
1717    that in any one match, only one of the named subpatterns participates. An
1718    example is shown in the
1719    .\" HREF
1720    \fBpcrepattern\fP
1721    .\"
1722    documentation.
1723  .P  .P
1724  .in 0  When duplicates are present, \fBpcre_copy_named_substring()\fP and
1725  Last updated: 09 September 2004  \fBpcre_get_named_substring()\fP return the first substring corresponding to
1726  .br  the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
1727  Copyright (c) 1997-2004 University of Cambridge.  returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
1728    returns one of the numbers that are associated with the name, but it is not
1729    defined which it is.
1730    .P
1731    If you want to get full details of all captured substrings for a given name,
1732    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
1733    argument is the compiled pattern, and the second is the name. The third and
1734    fourth are pointers to variables which are updated by the function. After it
1735    has run, they point to the first and last entries in the name-to-number table
1736    for the given name. The function itself returns the length of each entry, or
1737    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
1738    described above in the section entitled \fIInformation about a pattern\fP.
1739    Given all the relevant entries for the name, you can extract each of their
1740    numbers, and hence the captured data, if any.
1741    .
1742    .
1743    .SH "FINDING ALL POSSIBLE MATCHES"
1744    .rs
1745    .sp
1746    The traditional matching function uses a similar algorithm to Perl, which stops
1747    when it finds the first match, starting at a given point in the subject. If you
1748    want to find all possible matches, or the longest possible match, consider
1749    using the alternative matching function (see below) instead. If you cannot use
1750    the alternative function, but still need to find all possible matches, you
1751    can kludge it up by making use of the callout facility, which is described in
1752    the
1753    .\" HREF
1754    \fBpcrecallout\fP
1755    .\"
1756    documentation.
1757    .P
1758    What you have to do is to insert a callout right at the end of the pattern.
1759    When your callout function is called, extract and save the current matched
1760    substring. Then return 1, which forces \fBpcre_exec()\fP to backtrack and try
1761    other alternatives. Ultimately, when it runs out of matches, \fBpcre_exec()\fP
1762    will yield PCRE_ERROR_NOMATCH.
1763    .
1764    .
1765    .\" HTML <a name="dfamatch"></a>
1766    .SH "MATCHING A PATTERN: THE ALTERNATIVE FUNCTION"
1767    .rs
1768    .sp
1769    .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
1770    .ti +5n
1771    .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
1772    .ti +5n
1773    .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
1774    .ti +5n
1775    .B int *\fIworkspace\fP, int \fIwscount\fP);
1776    .P
1777    The function \fBpcre_dfa_exec()\fP is called to match a subject string against
1778    a compiled pattern, using a matching algorithm that scans the subject string
1779    just once, and does not backtrack. This has different characteristics to the
1780    normal algorithm, and is not compatible with Perl. Some of the features of PCRE
1781    patterns are not supported. Nevertheless, there are times when this kind of
1782    matching can be useful. For a discussion of the two matching algorithms, see
1783    the
1784    .\" HREF
1785    \fBpcrematching\fP
1786    .\"
1787    documentation.
1788    .P
1789    The arguments for the \fBpcre_dfa_exec()\fP function are the same as for
1790    \fBpcre_exec()\fP, plus two extras. The \fIovector\fP argument is used in a
1791    different way, and this is described below. The other common arguments are used
1792    in the same way as for \fBpcre_exec()\fP, so their description is not repeated
1793    here.
1794    .P
1795    The two additional arguments provide workspace for the function. The workspace
1796    vector should contain at least 20 elements. It is used for keeping track of
1797    multiple paths through the pattern tree. More workspace will be needed for
1798    patterns and subjects where there are a lot of potential matches.
1799    .P
1800    Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
1801    .sp
1802      int rc;
1803      int ovector[10];
1804      int wspace[20];
1805      rc = pcre_dfa_exec(
1806        re,             /* result of pcre_compile() */
1807        NULL,           /* we didn't study the pattern */
1808        "some string",  /* the subject string */
1809        11,             /* the length of the subject string */
1810        0,              /* start at offset 0 in the subject */
1811        0,              /* default options */
1812        ovector,        /* vector of integers for substring information */
1813        10,             /* number of elements (NOT size in bytes) */
1814        wspace,         /* working space vector */
1815        20);            /* number of elements (NOT size in bytes) */
1816    .
1817    .SS "Option bits for \fBpcre_dfa_exec()\fP"
1818    .rs
1819    .sp
1820    The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
1821    zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_\fIxxx\fP,
1822    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,
1823    PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are
1824    the same as for \fBpcre_exec()\fP, so their description is not repeated here.
1825    .sp
1826      PCRE_PARTIAL
1827    .sp
1828    This has the same general effect as it does for \fBpcre_exec()\fP, but the
1829    details are slightly different. When PCRE_PARTIAL is set for
1830    \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into
1831    PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no
1832    complete matches, but there is still at least one matching possibility. The
1833    portion of the string that provided the partial match is set as the first
1834    matching string.
1835    .sp
1836      PCRE_DFA_SHORTEST
1837    .sp
1838    Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
1839    soon as it has found one match. Because of the way the alternative algorithm
1840    works, this is necessarily the shortest possible match at the first possible
1841    matching point in the subject string.
1842    .sp
1843      PCRE_DFA_RESTART
1844    .sp
1845    When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns
1846    a partial match, it is possible to call it again, with additional subject
1847    characters, and have it continue with the same match. The PCRE_DFA_RESTART
1848    option requests this action; when it is set, the \fIworkspace\fP and
1849    \fIwscount\fP options must reference the same vector as before because data
1850    about the match so far is left in them after a partial match. There is more
1851    discussion of this facility in the
1852    .\" HREF
1853    \fBpcrepartial\fP
1854    .\"
1855    documentation.
1856    .
1857    .SS "Successful returns from \fBpcre_dfa_exec()\fP"
1858    .rs
1859    .sp
1860    When \fBpcre_dfa_exec()\fP succeeds, it may have matched more than one
1861    substring in the subject. Note, however, that all the matches from one run of
1862    the function start at the same point in the subject. The shorter matches are
1863    all initial substrings of the longer matches. For example, if the pattern
1864    .sp
1865      <.*>
1866    .sp
1867    is matched against the string
1868    .sp
1869      This is <something> <something else> <something further> no more
1870    .sp
1871    the three matched strings are
1872    .sp
1873      <something>
1874      <something> <something else>
1875      <something> <something else> <something further>
1876    .sp
1877    On success, the yield of the function is a number greater than zero, which is
1878    the number of matched substrings. The substrings themselves are returned in
1879    \fIovector\fP. Each string uses two elements; the first is the offset to the
1880    start, and the second is the offset to the end. In fact, all the strings have
1881    the same start offset. (Space could have been saved by giving this only once,
1882    but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
1883    returns data, even though the meaning of the strings is different.)
1884    .P
1885    The strings are returned in reverse order of length; that is, the longest
1886    matching string is given first. If there were too many matches to fit into
1887    \fIovector\fP, the yield of the function is zero, and the vector is filled with
1888    the longest matches.
1889    .
1890    .SS "Error returns from \fBpcre_dfa_exec()\fP"
1891    .rs
1892    .sp
1893    The \fBpcre_dfa_exec()\fP function returns a negative number when it fails.
1894    Many of the errors are the same as for \fBpcre_exec()\fP, and these are
1895    described
1896    .\" HTML <a href="#errorlist">
1897    .\" </a>
1898    above.
1899    .\"
1900    There are in addition the following errors that are specific to
1901    \fBpcre_dfa_exec()\fP:
1902    .sp
1903      PCRE_ERROR_DFA_UITEM      (-16)
1904    .sp
1905    This return is given if \fBpcre_dfa_exec()\fP encounters an item in the pattern
1906    that it does not support, for instance, the use of \eC or a back reference.
1907    .sp
1908      PCRE_ERROR_DFA_UCOND      (-17)
1909    .sp
1910    This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
1911    uses a back reference for the condition, or a test for recursion in a specific
1912    group. These are not supported.
1913    .sp
1914      PCRE_ERROR_DFA_UMLIMIT    (-18)
1915    .sp
1916    This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
1917    block that contains a setting of the \fImatch_limit\fP field. This is not
1918    supported (it is meaningless).
1919    .sp
1920      PCRE_ERROR_DFA_WSSIZE     (-19)
1921    .sp
1922    This return is given if \fBpcre_dfa_exec()\fP runs out of space in the
1923    \fIworkspace\fP vector.
1924    .sp
1925      PCRE_ERROR_DFA_RECURSE    (-20)
1926    .sp
1927    When a recursive subpattern is processed, the matching function calls itself
1928    recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
1929    error is given if the output vector is not large enough. This should be
1930    extremely rare, as a vector of size 1000 is used.
1931    .
1932    .
1933    .SH "SEE ALSO"
1934    .rs
1935    .sp
1936    \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
1937    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
1938    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
1939    .
1940    .
1941    .SH AUTHOR
1942    .rs
1943    .sp
1944    .nf
1945    Philip Hazel
1946    University Computing Service
1947    Cambridge CB2 3QH, England.
1948    .fi
1949    .
1950    .
1951    .SH REVISION
1952    .rs
1953    .sp
1954    .nf
1955    Last updated: 11 September 2007
1956    Copyright (c) 1997-2007 University of Cambridge.
1957    .fi

Legend:
Removed from v.75  
changed lines
  Added in v.243

  ViewVC Help
Powered by ViewVC 1.1.5