/[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 87 by nigel, Sat Feb 24 21:41:21 2007 UTC revision 887 by ph10, Tue Jan 17 14:32:32 2012 UTC
# Line 1  Line 1 
1  .TH PCREAPI 3  .TH PCREAPI 3
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
 .SH "PCRE NATIVE API"  
 .rs  
4  .sp  .sp
5  .B #include <pcre.h>  .B #include <pcre.h>
6  .PP  .
7    .
8    .SH "PCRE NATIVE API BASIC FUNCTIONS"
9    .rs
10    .sp
11  .SM  .SM
 .br  
12  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile(const char *\fIpattern\fP, int \fIoptions\fP,
13  .ti +5n  .ti +5n
14  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,  .B const char **\fIerrptr\fP, int *\fIerroffset\fP,
15  .ti +5n  .ti +5n
16  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
17  .PP  .PP
 .br  
18  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,  .B pcre *pcre_compile2(const char *\fIpattern\fP, int \fIoptions\fP,
19  .ti +5n  .ti +5n
20  .B int *\fIerrorcodeptr\fP,  .B int *\fIerrorcodeptr\fP,
# Line 23  PCRE - Perl-compatible regular expressio Line 23  PCRE - Perl-compatible regular expressio
23  .ti +5n  .ti +5n
24  .B const unsigned char *\fItableptr\fP);  .B const unsigned char *\fItableptr\fP);
25  .PP  .PP
 .br  
26  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,  .B pcre_extra *pcre_study(const pcre *\fIcode\fP, int \fIoptions\fP,
27  .ti +5n  .ti +5n
28  .B const char **\fIerrptr\fP);  .B const char **\fIerrptr\fP);
29  .PP  .PP
30  .br  .B void pcre_free_study(pcre_extra *\fIextra\fP);
31    .PP
32  .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,"
33  .ti +5n  .ti +5n
34  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
35  .ti +5n  .ti +5n
36  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP);
37  .PP  .PP
 .br  
38  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"  .B int pcre_dfa_exec(const pcre *\fIcode\fP, "const pcre_extra *\fIextra\fP,"
39  .ti +5n  .ti +5n
40  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,  .B "const char *\fIsubject\fP," int \fIlength\fP, int \fIstartoffset\fP,
# Line 43  PCRE - Perl-compatible regular expressio Line 42  PCRE - Perl-compatible regular expressio
42  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,  .B int \fIoptions\fP, int *\fIovector\fP, int \fIovecsize\fP,
43  .ti +5n  .ti +5n
44  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
45  .PP  .
46  .br  .
47    .SH "PCRE NATIVE API STRING EXTRACTION FUNCTIONS"
48    .rs
49    .sp
50  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
51  .ti +5n  .ti +5n
52  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 53  PCRE - Perl-compatible regular expressio Line 55  PCRE - Perl-compatible regular expressio
55  .ti +5n  .ti +5n
56  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
57  .PP  .PP
 .br  
58  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_copy_substring(const char *\fIsubject\fP, int *\fIovector\fP,
59  .ti +5n  .ti +5n
60  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP, char *\fIbuffer\fP,
61  .ti +5n  .ti +5n
62  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
63  .PP  .PP
 .br  
64  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
65  .ti +5n  .ti +5n
66  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 69  PCRE - Perl-compatible regular expressio Line 69  PCRE - Perl-compatible regular expressio
69  .ti +5n  .ti +5n
70  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
71  .PP  .PP
 .br  
72  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,  .B int pcre_get_stringnumber(const pcre *\fIcode\fP,
73  .ti +5n  .ti +5n
74  .B const char *\fIname\fP);  .B const char *\fIname\fP);
75  .PP  .PP
76  .br  .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
77    .ti +5n
78    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
79    .PP
80  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
81  .ti +5n  .ti +5n
82  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
83  .ti +5n  .ti +5n
84  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
85  .PP  .PP
 .br  
86  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
87  .ti +5n  .ti +5n
88  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
89  .PP  .PP
 .br  
90  .B void pcre_free_substring(const char *\fIstringptr\fP);  .B void pcre_free_substring(const char *\fIstringptr\fP);
91  .PP  .PP
 .br  
92  .B void pcre_free_substring_list(const char **\fIstringptr\fP);  .B void pcre_free_substring_list(const char **\fIstringptr\fP);
93    .
94    .
95    .SH "PCRE NATIVE API AUXILIARY FUNCTIONS"
96    .rs
97    .sp
98    .B pcre_jit_stack *pcre_jit_stack_alloc(int \fIstartsize\fP, int \fImaxsize\fP);
99    .PP
100    .B void pcre_jit_stack_free(pcre_jit_stack *\fIstack\fP);
101    .PP
102    .B void pcre_assign_jit_stack(pcre_extra *\fIextra\fP,
103    .ti +5n
104    .B pcre_jit_callback \fIcallback\fP, void *\fIdata\fP);
105  .PP  .PP
 .br  
106  .B const unsigned char *pcre_maketables(void);  .B const unsigned char *pcre_maketables(void);
107  .PP  .PP
 .br  
108  .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,"
109  .ti +5n  .ti +5n
110  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
111  .PP  .PP
 .br  
 .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  
 .B *\fIfirstcharptr\fP);  
 .PP  
 .br  
112  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);  .B int pcre_refcount(pcre *\fIcode\fP, int \fIadjust\fP);
113  .PP  .PP
 .br  
114  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);  .B int pcre_config(int \fIwhat\fP, void *\fIwhere\fP);
115  .PP  .PP
116  .br  .B const char *pcre_version(void);
 .B char *pcre_version(void);  
117  .PP  .PP
118  .br  .B int pcre_pattern_to_host_byte_order(pcre *\fIcode\fP,
119    .ti +5n
120    .B pcre_extra *\fIextra\fP, const unsigned char *\fItables\fP);
121    .
122    .
123    .SH "PCRE NATIVE API INDIRECTED FUNCTIONS"
124    .rs
125    .sp
126  .B void *(*pcre_malloc)(size_t);  .B void *(*pcre_malloc)(size_t);
127  .PP  .PP
 .br  
128  .B void (*pcre_free)(void *);  .B void (*pcre_free)(void *);
129  .PP  .PP
 .br  
130  .B void *(*pcre_stack_malloc)(size_t);  .B void *(*pcre_stack_malloc)(size_t);
131  .PP  .PP
 .br  
132  .B void (*pcre_stack_free)(void *);  .B void (*pcre_stack_free)(void *);
133  .PP  .PP
 .br  
134  .B int (*pcre_callout)(pcre_callout_block *);  .B int (*pcre_callout)(pcre_callout_block *);
135  .  .
136  .  .
137    .SH "PCRE 8-BIT AND 16-BIT LIBRARIES"
138    .rs
139    .sp
140    From release 8.30, PCRE can be compiled as a library for handling 16-bit
141    character strings as well as, or instead of, the original library that handles
142    8-bit character strings. To avoid too much complication, this document
143    describes the 8-bit versions of the functions, with only occasional references
144    to the 16-bit library.
145    .P
146    The 16-bit functions operate in the same way as their 8-bit counterparts; they
147    just use different data types for their arguments and results, and their names
148    start with \fBpcre16_\fP instead of \fBpcre_\fP. For every option that has UTF8
149    in its name (for example, PCRE_UTF8), there is a corresponding 16-bit name with
150    UTF8 replaced by UTF16. This facility is in fact just cosmetic; the 16-bit
151    option names define the same bit values.
152    .P
153    References to bytes and UTF-8 in this document should be read as references to
154    16-bit data quantities and UTF-16 when using the 16-bit library, unless
155    specified otherwise. More details of the specific differences for the 16-bit
156    library are given in the
157    .\" HREF
158    \fBpcre16\fP
159    .\"
160    page.
161    .
162    .
163  .SH "PCRE API OVERVIEW"  .SH "PCRE API OVERVIEW"
164  .rs  .rs
165  .sp  .sp
166  PCRE has its own native API, which is described in this document. There is  PCRE has its own native API, which is described in this document. There are
167  also a set of wrapper functions that correspond to the POSIX regular expression  also some wrapper functions (for the 8-bit library only) that correspond to the
168  API. These are described in the  POSIX regular expression API, but they do not give access to all the
169    functionality. They are described in the
170  .\" HREF  .\" HREF
171  \fBpcreposix\fP  \fBpcreposix\fP
172  .\"  .\"
173  documentation. Both of these APIs define a set of C function calls. A C++  documentation. Both of these APIs define a set of C function calls. A C++
174  wrapper is distributed with PCRE. It is documented in the  wrapper (again for the 8-bit library only) is also distributed with PCRE. It is
175    documented in the
176  .\" HREF  .\" HREF
177  \fBpcrecpp\fP  \fBpcrecpp\fP
178  .\"  .\"
179  page.  page.
180  .P  .P
181  The native API C function prototypes are defined in the header file  The native API C function prototypes are defined in the header file
182  \fBpcre.h\fP, and on Unix systems the library itself is called \fBlibpcre\fP.  \fBpcre.h\fP, and on Unix-like systems the (8-bit) library itself is called
183  It can normally be accessed by adding \fB-lpcre\fP to the command for linking  \fBlibpcre\fP. It can normally be accessed by adding \fB-lpcre\fP to the
184  an application that uses PCRE. The header file defines the macros PCRE_MAJOR  command for linking an application that uses PCRE. The header file defines the
185  and PCRE_MINOR to contain the major and minor release numbers for the library.  macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release numbers
186  Applications can use these to include support for different releases of PCRE.  for the library. Applications can use these to include support for different
187    releases of PCRE.
188    .P
189    In a Windows environment, if you want to statically link an application program
190    against a non-dll \fBpcre.a\fP file, you must define PCRE_STATIC before
191    including \fBpcre.h\fP or \fBpcrecpp.h\fP, because otherwise the
192    \fBpcre_malloc()\fP and \fBpcre_free()\fP exported functions will be declared
193    \fB__declspec(dllimport)\fP, with unwanted results.
194  .P  .P
195  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,  The functions \fBpcre_compile()\fP, \fBpcre_compile2()\fP, \fBpcre_study()\fP,
196  and \fBpcre_exec()\fP are used for compiling and matching regular expressions  and \fBpcre_exec()\fP are used for compiling and matching regular expressions
197  in a Perl-compatible manner. A sample program that demonstrates the simplest  in a Perl-compatible manner. A sample program that demonstrates the simplest
198  way of using them is provided in the file called \fIpcredemo.c\fP in the source  way of using them is provided in the file called \fIpcredemo.c\fP in the PCRE
199  distribution. The  source distribution. A listing of this program is given in the
200    .\" HREF
201    \fBpcredemo\fP
202    .\"
203    documentation, and the
204  .\" HREF  .\" HREF
205  \fBpcresample\fP  \fBpcresample\fP
206  .\"  .\"
207  documentation describes how to run it.  documentation describes how to compile and run it.
208    .P
209    Just-in-time compiler support is an optional feature of PCRE that can be built
210    in appropriate hardware environments. It greatly speeds up the matching
211    performance of many patterns. Simple programs can easily request that it be
212    used if available, by setting an option that is ignored when it is not
213    relevant. More complicated programs might need to make use of the functions
214    \fBpcre_jit_stack_alloc()\fP, \fBpcre_jit_stack_free()\fP, and
215    \fBpcre_assign_jit_stack()\fP in order to control the JIT code's memory usage.
216    These functions are discussed in the
217    .\" HREF
218    \fBpcrejit\fP
219    .\"
220    documentation.
221  .P  .P
222  A second matching function, \fBpcre_dfa_exec()\fP, which is not  A second matching function, \fBpcre_dfa_exec()\fP, which is not
223  Perl-compatible, is also provided. This uses a different algorithm for the  Perl-compatible, is also provided. This uses a different algorithm for the
224  matching. This allows it to find all possible matches (at a given point in the  matching. The alternative algorithm finds all possible matches (at a given
225  subject), not just one. However, this algorithm does not return captured  point in the subject), and scans the subject just once (unless there are
226    lookbehind assertions). However, this algorithm does not return captured
227  substrings. A description of the two matching algorithms and their advantages  substrings. A description of the two matching algorithms and their advantages
228  and disadvantages is given in the  and disadvantages is given in the
229  .\" HREF  .\" HREF
# Line 183  matched by \fBpcre_exec()\fP. They are: Line 241  matched by \fBpcre_exec()\fP. They are:
241    \fBpcre_get_named_substring()\fP    \fBpcre_get_named_substring()\fP
242    \fBpcre_get_substring_list()\fP    \fBpcre_get_substring_list()\fP
243    \fBpcre_get_stringnumber()\fP    \fBpcre_get_stringnumber()\fP
244      \fBpcre_get_stringtable_entries()\fP
245  .sp  .sp
246  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also  \fBpcre_free_substring()\fP and \fBpcre_free_substring_list()\fP are also
247  provided, to free the memory used for extracted strings.  provided, to free the memory used for extracted strings.
# Line 194  specialist use. Most commonly, no specia Line 253  specialist use. Most commonly, no specia
253  internal tables that are generated when PCRE is built are used.  internal tables that are generated when PCRE is built are used.
254  .P  .P
255  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
256  compiled pattern; \fBpcre_info()\fP is an obsolete version that returns only  compiled pattern. The function \fBpcre_version()\fP returns a pointer to a
257  some of the available information, but is retained for backwards compatibility.  string containing the version of PCRE and its date of release.
 The function \fBpcre_version()\fP returns a pointer to a string containing the  
 version of PCRE and its date of release.  
258  .P  .P
259  The function \fBpcre_refcount()\fP maintains a reference count in a data block  The function \fBpcre_refcount()\fP maintains a reference count in a data block
260  containing a compiled pattern. This is provided for the benefit of  containing a compiled pattern. This is provided for the benefit of
# Line 212  should be done before calling any PCRE f Line 269  should be done before calling any PCRE f
269  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
270  indirections to memory management functions. These special functions are used  indirections to memory management functions. These special functions are used
271  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
272  recursive function calls, when running the \fBpcre_exec()\fP function. This is  recursive function calls, when running the \fBpcre_exec()\fP function. See the
273  a non-standard way of building PCRE, for use in environments that have limited  .\" HREF
274  stacks. Because of the greater use of memory management, it runs more slowly.  \fBpcrebuild\fP
275  Separate functions are provided so that special-purpose external code can be  .\"
276  used for this case. When used, these functions are always called in a  documentation for details of how to do this. It is a non-standard way of
277  stack-like manner (last obtained, first freed), and always for memory blocks of  building PCRE, for use in environments that have limited stacks. Because of the
278  the same size.  greater use of memory management, it runs more slowly. Separate functions are
279    provided so that special-purpose external code can be used for this case. When
280    used, these functions are always called in a stack-like manner (last obtained,
281    first freed), and always for memory blocks of the same size. There is a
282    discussion about PCRE's stack usage in the
283    .\" HREF
284    \fBpcrestack\fP
285    .\"
286    documentation.
287  .P  .P
288  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
289  by the caller to a "callout" function, which PCRE will then call at specified  by the caller to a "callout" function, which PCRE will then call at specified
# Line 229  points during a matching operation. Deta Line 294  points during a matching operation. Deta
294  documentation.  documentation.
295  .  .
296  .  .
297    .\" HTML <a name="newlines"></a>
298    .SH NEWLINES
299    .rs
300    .sp
301    PCRE supports five different conventions for indicating line breaks in
302    strings: a single CR (carriage return) character, a single LF (linefeed)
303    character, the two-character sequence CRLF, any of the three preceding, or any
304    Unicode newline sequence. The Unicode newline sequences are the three just
305    mentioned, plus the single characters VT (vertical tab, U+000B), FF (formfeed,
306    U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
307    (paragraph separator, U+2029).
308    .P
309    Each of the first three conventions is used by at least one operating system as
310    its standard newline sequence. When PCRE is built, a default can be specified.
311    The default default is LF, which is the Unix standard. When PCRE is run, the
312    default can be overridden, either when a pattern is compiled, or when it is
313    matched.
314    .P
315    At compile time, the newline convention can be specified by the \fIoptions\fP
316    argument of \fBpcre_compile()\fP, or it can be specified by special text at the
317    start of the pattern itself; this overrides any other settings. See the
318    .\" HREF
319    \fBpcrepattern\fP
320    .\"
321    page for details of the special character sequences.
322    .P
323    In the PCRE documentation the word "newline" is used to mean "the character or
324    pair of characters that indicate a line break". The choice of newline
325    convention affects the handling of the dot, circumflex, and dollar
326    metacharacters, the handling of #-comments in /x mode, and, when CRLF is a
327    recognized line ending sequence, the match position advancement for a
328    non-anchored pattern. There is more detail about this in the
329    .\" HTML <a href="#execoptions">
330    .\" </a>
331    section on \fBpcre_exec()\fP options
332    .\"
333    below.
334    .P
335    The choice of newline convention does not affect the interpretation of
336    the \en or \er escape sequences, nor does it affect what \eR matches, which is
337    controlled in a similar way, but by separate options.
338    .
339    .
340  .SH MULTITHREADING  .SH MULTITHREADING
341  .rs  .rs
342  .sp  .sp
# Line 239  callout function pointed to by \fBpcre_c Line 347  callout function pointed to by \fBpcre_c
347  .P  .P
348  The compiled form of a regular expression is not altered during matching, so  The compiled form of a regular expression is not altered during matching, so
349  the same compiled pattern can safely be used by several threads at once.  the same compiled pattern can safely be used by several threads at once.
350    .P
351    If the just-in-time optimization feature is being used, it needs separate
352    memory stack areas for each thread. See the
353    .\" HREF
354    \fBpcrejit\fP
355    .\"
356    documentation for more details.
357  .  .
358  .  .
359  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"  .SH "SAVING PRECOMPILED PATTERNS FOR LATER USE"
# Line 250  which it was compiled. Details are given Line 365  which it was compiled. Details are given
365  .\" HREF  .\" HREF
366  \fBpcreprecompile\fP  \fBpcreprecompile\fP
367  .\"  .\"
368  documentation.  documentation, which includes a description of the
369    \fBpcre_pattern_to_host_byte_order()\fP function. However, compiling a regular
370    expression with one version of PCRE for use with a different version is not
371    guaranteed to work and may cause crashes.
372  .  .
373  .  .
374  .SH "CHECKING BUILD-TIME OPTIONS"  .SH "CHECKING BUILD-TIME OPTIONS"
# Line 267  documentation has more details about the Line 385  documentation has more details about the
385  .P  .P
386  The first argument for \fBpcre_config()\fP is an integer, specifying which  The first argument for \fBpcre_config()\fP is an integer, specifying which
387  information is required; the second argument is a pointer to a variable into  information is required; the second argument is a pointer to a variable into
388  which the information is placed. The following information is available:  which the information is placed. The returned value is zero on success, or the
389    negative error code PCRE_ERROR_BADOPTION if the value in the first argument is
390    not recognized. The following information is available:
391  .sp  .sp
392    PCRE_CONFIG_UTF8    PCRE_CONFIG_UTF8
393  .sp  .sp
394  The output is an integer that is set to one if UTF-8 support is available;  The output is an integer that is set to one if UTF-8 support is available;
395  otherwise it is set to zero.  otherwise it is set to zero. If this option is given to the 16-bit version of
396    this function, \fBpcre16_config()\fP, the result is PCRE_ERROR_BADOPTION.
397    .sp
398      PCRE_CONFIG_UTF16
399    .sp
400    The output is an integer that is set to one if UTF-16 support is available;
401    otherwise it is set to zero. This value should normally be given to the 16-bit
402    version of this function, \fBpcre16_config()\fP. If it is given to the 8-bit
403    version of this function, the result is PCRE_ERROR_BADOPTION.
404  .sp  .sp
405    PCRE_CONFIG_UNICODE_PROPERTIES    PCRE_CONFIG_UNICODE_PROPERTIES
406  .sp  .sp
407  The output is an integer that is set to one if support for Unicode character  The output is an integer that is set to one if support for Unicode character
408  properties is available; otherwise it is set to zero.  properties is available; otherwise it is set to zero.
409  .sp  .sp
410      PCRE_CONFIG_JIT
411    .sp
412    The output is an integer that is set to one if support for just-in-time
413    compiling is available; otherwise it is set to zero.
414    .sp
415      PCRE_CONFIG_JITTARGET
416    .sp
417    The output is a pointer to a zero-terminated "const char *" string. If JIT
418    support is available, the string contains the name of the architecture for
419    which the JIT compiler is configured, for example "x86 32bit (little endian +
420    unaligned)". If JIT support is not available, the result is NULL.
421    .sp
422    PCRE_CONFIG_NEWLINE    PCRE_CONFIG_NEWLINE
423  .sp  .sp
424  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
425  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
426  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.
427    Though they are derived from ASCII, the same values are returned in EBCDIC
428    environments. The default should normally correspond to the standard sequence
429    for your operating system.
430    .sp
431      PCRE_CONFIG_BSR
432    .sp
433    The output is an integer whose value indicates what character sequences the \eR
434    escape sequence matches by default. A value of 0 means that \eR matches any
435    Unicode line ending sequence; a value of 1 means that \eR matches only CR, LF,
436    or CRLF. The default can be overridden when a pattern is compiled or matched.
437  .sp  .sp
438    PCRE_CONFIG_LINK_SIZE    PCRE_CONFIG_LINK_SIZE
439  .sp  .sp
440  The output is an integer that contains the number of bytes used for internal  The output is an integer that contains the number of bytes used for internal
441  linkage in compiled regular expressions. The value is 2, 3, or 4. Larger values  linkage in compiled regular expressions. For the 8-bit library, the value can
442  allow larger regular expressions to be compiled, at the expense of slower  be 2, 3, or 4. For the 16-bit library, the value is either 2 or 4 and is still
443  matching. The default value of 2 is sufficient for all but the most massive  a number of bytes. The default value of 2 is sufficient for all but the most
444  patterns, since it allows the compiled pattern to be up to 64K in size.  massive patterns, since it allows the compiled pattern to be up to 64K in size.
445    Larger values allow larger regular expressions to be compiled, at the expense
446    of slower matching.
447  .sp  .sp
448    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD    PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
449  .sp  .sp
# Line 305  documentation. Line 457  documentation.
457  .sp  .sp
458    PCRE_CONFIG_MATCH_LIMIT    PCRE_CONFIG_MATCH_LIMIT
459  .sp  .sp
460  The output is an integer that gives the default limit for the number of  The output is a long integer that gives the default limit for the number of
461  internal matching function calls in a \fBpcre_exec()\fP execution. Further  internal matching function calls in a \fBpcre_exec()\fP execution. Further
462  details are given with \fBpcre_exec()\fP below.  details are given with \fBpcre_exec()\fP below.
463  .sp  .sp
464    PCRE_CONFIG_MATCH_LIMIT_RECURSION    PCRE_CONFIG_MATCH_LIMIT_RECURSION
465  .sp  .sp
466  The output is an integer that gives the default limit for the depth of  The output is a long integer that gives the default limit for the depth of
467  recursion when calling the internal matching function in a \fBpcre_exec()\fP  recursion when calling the internal matching function in a \fBpcre_exec()\fP
468  execution. Further details are given with \fBpcre_exec()\fP below.  execution. Further details are given with \fBpcre_exec()\fP below.
469  .sp  .sp
# Line 346  avoiding the use of the stack. Line 498  avoiding the use of the stack.
498  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be  Either of the functions \fBpcre_compile()\fP or \fBpcre_compile2()\fP can be
499  called to compile a pattern into an internal form. The only difference between  called to compile a pattern into an internal form. The only difference between
500  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,  the two interfaces is that \fBpcre_compile2()\fP has an additional argument,
501  \fIerrorcodeptr\fP, via which a numerical error code can be returned.  \fIerrorcodeptr\fP, via which a numerical error code can be returned. To avoid
502    too much repetition, we refer just to \fBpcre_compile()\fP below, but the
503    information applies equally to \fBpcre_compile2()\fP.
504  .P  .P
505  The pattern is a C string terminated by a binary zero, and is passed in the  The pattern is a C string terminated by a binary zero, and is passed in the
506  \fIpattern\fP argument. A pointer to a single block of memory that is obtained  \fIpattern\fP argument. A pointer to a single block of memory that is obtained
507  via \fBpcre_malloc\fP is returned. This contains the compiled code and related  via \fBpcre_malloc\fP is returned. This contains the compiled code and related
508  data. The \fBpcre\fP type is defined for the returned block; this is a typedef  data. The \fBpcre\fP type is defined for the returned block; this is a typedef
509  for a structure whose contents are not externally defined. It is up to the  for a structure whose contents are not externally defined. It is up to the
510  caller to free the memory when it is no longer required.  caller to free the memory (via \fBpcre_free\fP) when it is no longer required.
511  .P  .P
512  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
513  depend on memory location, the complete \fBpcre\fP data block is not  depend on memory location, the complete \fBpcre\fP data block is not
514  fully relocatable, because it may contain a copy of the \fItableptr\fP  fully relocatable, because it may contain a copy of the \fItableptr\fP
515  argument, which is an address (see below).  argument, which is an address (see below).
516  .P  .P
517  The \fIoptions\fP argument contains independent bits that affect the  The \fIoptions\fP argument contains various bit settings that affect the
518  compilation. It should be zero if no options are required. The available  compilation. It should be zero if no options are required. The available
519  options are described below. Some of them, in particular, those that are  options are described below. Some of them (in particular, those that are
520  compatible with Perl, can also be set and unset from within the pattern (see  compatible with Perl, but some others as well) can also be set and unset from
521  the detailed description in the  within the pattern (see the detailed description in the
522  .\" HREF  .\" HREF
523  \fBpcrepattern\fP  \fBpcrepattern\fP
524  .\"  .\"
525  documentation). For these options, the contents of the \fIoptions\fP argument  documentation). For those options that can be different in different parts of
526  specifies their initial settings at the start of compilation and execution. The  the pattern, the contents of the \fIoptions\fP argument specifies their
527  PCRE_ANCHORED option can be set at the time of matching as well as at compile  settings at the start of compilation and execution. The PCRE_ANCHORED,
528  time.  PCRE_BSR_\fIxxx\fP, PCRE_NEWLINE_\fIxxx\fP, PCRE_NO_UTF8_CHECK, and
529    PCRE_NO_START_OPT options can be set at the time of matching as well as at
530    compile time.
531  .P  .P
532  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.  If \fIerrptr\fP is NULL, \fBpcre_compile()\fP returns NULL immediately.
533  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns  Otherwise, if compilation of a pattern fails, \fBpcre_compile()\fP returns
534  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
535  error message. This is a static string that is part of the library. You must  error message. This is a static string that is part of the library. You must
536  not try to free it. The offset from the start of the pattern to the character  not try to free it. Normally, the offset from the start of the pattern to the
537  where the error was discovered is placed in the variable pointed to by  byte that was being processed when the error was discovered is placed in the
538  \fIerroffset\fP, which must not be NULL. If it is, an immediate error is given.  variable pointed to by \fIerroffset\fP, which must not be NULL (if it is, an
539    immediate error is given). However, for an invalid UTF-8 string, the offset is
540    that of the first byte of the failing character.
541    .P
542    Some errors are not detected until the whole pattern has been scanned; in these
543    cases, the offset passed back is the length of the pattern. Note that the
544    offset is in bytes, not characters, even in UTF-8 mode. It may sometimes point
545    into the middle of a UTF-8 character.
546  .P  .P
547  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the  If \fBpcre_compile2()\fP is used instead of \fBpcre_compile()\fP, and the
548  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is  \fIerrorcodeptr\fP argument is not NULL, a non-zero error code number is
# Line 426  facility, see the Line 589  facility, see the
589  .\"  .\"
590  documentation.  documentation.
591  .sp  .sp
592      PCRE_BSR_ANYCRLF
593      PCRE_BSR_UNICODE
594    .sp
595    These options (which are mutually exclusive) control what the \eR escape
596    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
597    match any Unicode newline sequence. The default is specified when PCRE is
598    built. It can be overridden from within the pattern, or by setting an option
599    when a compiled pattern is matched.
600    .sp
601    PCRE_CASELESS    PCRE_CASELESS
602  .sp  .sp
603  If this bit is set, letters in the pattern match both upper and lower case  If this bit is set, letters in the pattern match both upper and lower case
# Line 442  with UTF-8 support. Line 614  with UTF-8 support.
614  .sp  .sp
615  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
616  end of the subject string. Without this option, a dollar also matches  end of the subject string. Without this option, a dollar also matches
617  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
618  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.
619  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
620  a pattern.  pattern.
621  .sp  .sp
622    PCRE_DOTALL    PCRE_DOTALL
623  .sp  .sp
624  If this bit is set, a dot metacharater in the pattern matches all characters,  If this bit is set, a dot metacharacter in the pattern matches a character of
625  including newlines. Without it, newlines are excluded. This option is  any value, including one that indicates a newline. However, it only ever
626    matches one character, even if newlines are coded as CRLF. Without this option,
627    a dot does not match when the current position is at a newline. This option is
628  equivalent to Perl's /s option, and it can be changed within a pattern by a  equivalent to Perl's /s option, and it can be changed within a pattern by a
629  (?s) option setting. A negative class such as [^a] always matches a newline  (?s) option setting. A negative class such as [^a] always matches newline
630  character, independent of the setting of this option.  characters, independent of the setting of this option.
631    .sp
632      PCRE_DUPNAMES
633    .sp
634    If this bit is set, names used to identify capturing subpatterns need not be
635    unique. This can be helpful for certain types of pattern when it is known that
636    only one instance of the named subpattern can ever be matched. There are more
637    details of named subpatterns below; see also the
638    .\" HREF
639    \fBpcrepattern\fP
640    .\"
641    documentation.
642  .sp  .sp
643    PCRE_EXTENDED    PCRE_EXTENDED
644  .sp  .sp
645  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
646  ignored except when escaped or inside a character class. Whitespace does not  ignored except when escaped or inside a character class. Whitespace does not
647  include the VT character (code 11). In addition, characters between an  include the VT character (code 11). In addition, characters between an
648  unescaped # outside a character class and the next newline character,  unescaped # outside a character class and the next newline, inclusive, are also
649  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
650  be changed within a pattern by a (?x) option setting.  pattern by a (?x) option setting.
651    .P
652    Which characters are interpreted as newlines is controlled by the options
653    passed to \fBpcre_compile()\fP or by a special sequence at the start of the
654    pattern, as described in the section entitled
655    .\" HTML <a href="pcrepattern.html#newlines">
656    .\" </a>
657    "Newline conventions"
658    .\"
659    in the \fBpcrepattern\fP documentation. Note that the end of this type of
660    comment is a literal newline sequence in the pattern; escape sequences that
661    happen to represent a newline do not count.
662  .P  .P
663  This option makes it possible to include comments inside complicated patterns.  This option makes it possible to include comments inside complicated patterns.
664  Note, however, that this applies only to data characters. Whitespace characters  Note, however, that this applies only to data characters. Whitespace characters
665  may never appear within special character sequences in a pattern, for example  may never appear within special character sequences in a pattern, for example
666  within the sequence (?( which introduces a conditional subpattern.  within the sequence (?( that introduces a conditional subpattern.
667  .sp  .sp
668    PCRE_EXTRA    PCRE_EXTRA
669  .sp  .sp
# Line 476  that is incompatible with Perl, but it i Line 672  that is incompatible with Perl, but it i
672  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
673  special meaning causes an error, thus reserving these combinations for future  special meaning causes an error, thus reserving these combinations for future
674  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
675  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
676  controlled by this option. It can also be set by a (?X) option setting within a  give an error for this, by running it with the -w option.) There are at present
677  pattern.  no other features controlled by this option. It can also be set by a (?X)
678    option setting within a pattern.
679  .sp  .sp
680    PCRE_FIRSTLINE    PCRE_FIRSTLINE
681  .sp  .sp
682  If this option is set, an unanchored pattern is required to match before or at  If this option is set, an unanchored pattern is required to match before or at
683  the first newline character in the subject string, though the matched text may  the first newline in the subject string, though the matched text may continue
684  continue over the newline.  over the newline.
685    .sp
686      PCRE_JAVASCRIPT_COMPAT
687    .sp
688    If this option is set, PCRE's behaviour is changed in some ways so that it is
689    compatible with JavaScript rather than Perl. The changes are as follows:
690    .P
691    (1) A lone closing square bracket in a pattern causes a compile-time error,
692    because this is illegal in JavaScript (by default it is treated as a data
693    character). Thus, the pattern AB]CD becomes illegal when this option is set.
694    .P
695    (2) At run time, a back reference to an unset subpattern group matches an empty
696    string (by default this causes the current matching alternative to fail). A
697    pattern such as (\e1)(a) succeeds when this option is set (assuming it can find
698    an "a" in the subject), whereas it fails by default, for Perl compatibility.
699    .P
700    (3) \eU matches an upper case "U" character; by default \eU causes a compile
701    time error (Perl uses \eU to upper case subsequent characters).
702    .P
703    (4) \eu matches a lower case "u" character unless it is followed by four
704    hexadecimal digits, in which case the hexadecimal number defines the code point
705    to match. By default, \eu causes a compile time error (Perl uses it to upper
706    case the following character).
707    .P
708    (5) \ex matches a lower case "x" character unless it is followed by two
709    hexadecimal digits, in which case the hexadecimal number defines the code point
710    to match. By default, as in Perl, a hexadecimal number is always expected after
711    \ex, but it may have zero, one, or two digits (so, for example, \exz matches a
712    binary zero character followed by z).
713  .sp  .sp
714    PCRE_MULTILINE    PCRE_MULTILINE
715  .sp  .sp
# Line 496  terminating newline (unless PCRE_DOLLAR_ Line 721  terminating newline (unless PCRE_DOLLAR_
721  Perl.  Perl.
722  .P  .P
723  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
724  match immediately following or immediately before any newline in the subject  match immediately following or immediately before internal newlines in the
725  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
726  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
727  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
728  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.  occurrences of ^ or $ in a pattern, setting PCRE_MULTILINE has no effect.
729  .sp  .sp
730      PCRE_NEWLINE_CR
731      PCRE_NEWLINE_LF
732      PCRE_NEWLINE_CRLF
733      PCRE_NEWLINE_ANYCRLF
734      PCRE_NEWLINE_ANY
735    .sp
736    These options override the default newline definition that was chosen when PCRE
737    was built. Setting the first or the second specifies that a newline is
738    indicated by a single character (CR or LF, respectively). Setting
739    PCRE_NEWLINE_CRLF specifies that a newline is indicated by the two-character
740    CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies that any of the three
741    preceding sequences should be recognized. Setting PCRE_NEWLINE_ANY specifies
742    that any Unicode newline sequence should be recognized. The Unicode newline
743    sequences are the three just mentioned, plus the single characters VT (vertical
744    tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS (line
745    separator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
746    library, the last two are recognized only in UTF-8 mode.
747    .P
748    The newline setting in the options word uses three bits that are treated
749    as a number, giving eight possibilities. Currently only six are used (default
750    plus the five values above). This means that if you set more than one newline
751    option, the combination may or may not be sensible. For example,
752    PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to PCRE_NEWLINE_CRLF, but
753    other combinations may yield unused numbers and cause an error.
754    .P
755    The only time that a line break in a pattern is specially recognized when
756    compiling is when PCRE_EXTENDED is set. CR and LF are whitespace characters,
757    and so are ignored in this mode. Also, an unescaped # outside a character class
758    indicates a comment that lasts until after the next line break sequence. In
759    other circumstances, line break sequences in patterns are treated as literal
760    data.
761    .P
762    The newline option that is set at compile time becomes the default that is used
763    for \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, but it can be overridden.
764    .sp
765    PCRE_NO_AUTO_CAPTURE    PCRE_NO_AUTO_CAPTURE
766  .sp  .sp
767  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 510  were followed by ?: but named parenthese Line 770  were followed by ?: but named parenthese
770  they acquire numbers in the usual way). There is no equivalent of this option  they acquire numbers in the usual way). There is no equivalent of this option
771  in Perl.  in Perl.
772  .sp  .sp
773      NO_START_OPTIMIZE
774    .sp
775    This is an option that acts at matching time; that is, it is really an option
776    for \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. If it is set at compile time,
777    it is remembered with the compiled pattern and assumed at matching time. For
778    details see the discussion of PCRE_NO_START_OPTIMIZE
779    .\" HTML <a href="#execoptions">
780    .\" </a>
781    below.
782    .\"
783    .sp
784      PCRE_UCP
785    .sp
786    This option changes the way PCRE processes \eB, \eb, \eD, \ed, \eS, \es, \eW,
787    \ew, and some of the POSIX character classes. By default, only ASCII characters
788    are recognized, but if PCRE_UCP is set, Unicode properties are used instead to
789    classify characters. More details are given in the section on
790    .\" HTML <a href="pcre.html#genericchartypes">
791    .\" </a>
792    generic character types
793    .\"
794    in the
795    .\" HREF
796    \fBpcrepattern\fP
797    .\"
798    page. If you set PCRE_UCP, matching one of the items it affects takes much
799    longer. The option is available only if PCRE has been compiled with Unicode
800    property support.
801    .sp
802    PCRE_UNGREEDY    PCRE_UNGREEDY
803  .sp  .sp
804  This option inverts the "greediness" of the quantifiers so that they are not  This option inverts the "greediness" of the quantifiers so that they are not
# Line 519  with Perl. It can also be set by a (?U) Line 808  with Perl. It can also be set by a (?U)
808    PCRE_UTF8    PCRE_UTF8
809  .sp  .sp
810  This option causes PCRE to regard both the pattern and the subject as strings  This option causes PCRE to regard both the pattern and the subject as strings
811  of UTF-8 characters instead of single-byte character strings. However, it is  of UTF-8 characters instead of single-byte strings. However, it is available
812  available only when PCRE is built to include UTF-8 support. If not, the use  only when PCRE is built to include UTF support. If not, the use of this option
813  of this option provokes an error. Details of how this option changes the  provokes an error. Details of how this option changes the behaviour of PCRE are
814  behaviour of PCRE are given in the  given in the
 .\" HTML <a href="pcre.html#utf8support">  
 .\" </a>  
 section on UTF-8 support  
 .\"  
 in the main  
815  .\" HREF  .\" HREF
816  \fBpcre\fP  \fBpcreunicode\fP
817  .\"  .\"
818  page.  page.
819  .sp  .sp
820    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
821  .sp  .sp
822  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
823  automatically checked. If an invalid UTF-8 sequence of bytes is found,  string is automatically checked. There is a discussion about the
824  \fBpcre_compile()\fP returns an error. If you already know that your pattern is  .\" HTML <a href="pcreunicode.html#utf8strings">
825  valid, and you want to skip this check for performance reasons, you can set the  .\" </a>
826  PCRE_NO_UTF8_CHECK option. When it is set, the effect of passing an invalid  validity of UTF-8 strings
827  UTF-8 string as a pattern is undefined. It may cause your program to crash.  .\"
828  Note that this option can also be passed to \fBpcre_exec()\fP and  in the
829  \fBpcre_dfa_exec()\fP, to suppress the UTF-8 validity checking of subject  .\" HREF
830  strings.  \fBpcreunicode\fP
831    .\"
832    page. If an invalid UTF-8 sequence is found, \fBpcre_compile()\fP returns an
833    error. If you already know that your pattern is valid, and you want to skip
834    this check for performance reasons, you can set the PCRE_NO_UTF8_CHECK option.
835    When it is set, the effect of passing an invalid UTF-8 string as a pattern is
836    undefined. It may cause your program to crash. Note that this option can also
837    be passed to \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP, to suppress the
838    validity checking of subject strings.
839  .  .
840  .  .
841  .SH "COMPILATION ERROR CODES"  .SH "COMPILATION ERROR CODES"
# Line 551  strings. Line 843  strings.
843  .sp  .sp
844  The following table lists the error codes than may be returned by  The following table lists the error codes than may be returned by
845  \fBpcre_compile2()\fP, along with the error messages that may be returned by  \fBpcre_compile2()\fP, along with the error messages that may be returned by
846  both compiling functions.  both compiling functions. Note that error messages are always 8-bit ASCII
847    strings, even in 16-bit mode. As PCRE has developed, some error codes have
848    fallen out of use. To avoid confusion, they have not been re-used.
849  .sp  .sp
850     0  no error     0  no error
851     1  \e at end of pattern     1  \e at end of pattern
# Line 563  both compiling functions. Line 857  both compiling functions.
857     7  invalid escape sequence in character class     7  invalid escape sequence in character class
858     8  range out of order in character class     8  range out of order in character class
859     9  nothing to repeat     9  nothing to repeat
860    10  operand of unlimited repeat could match the empty string    10  [this code is not in use]
861    11  internal error: unexpected repeat    11  internal error: unexpected repeat
862    12  unrecognized character after (?    12  unrecognized character after (? or (?-
863    13  POSIX named classes are supported only within a class    13  POSIX named classes are supported only within a class
864    14  missing )    14  missing )
865    15  reference to non-existent subpattern    15  reference to non-existent subpattern
866    16  erroffset passed as NULL    16  erroffset passed as NULL
867    17  unknown option bit(s) set    17  unknown option bit(s) set
868    18  missing ) after comment    18  missing ) after comment
869    19  parentheses nested too deeply    19  [this code is not in use]
870    20  regular expression too large    20  regular expression is too large
871    21  failed to get memory    21  failed to get memory
872    22  unmatched parentheses    22  unmatched parentheses
873    23  internal error: code overflow    23  internal error: code overflow
874    24  unrecognized character after (?<    24  unrecognized character after (?<
875    25  lookbehind assertion is not fixed length    25  lookbehind assertion is not fixed length
876    26  malformed number after (?(    26  malformed number or name after (?(
877    27  conditional group contains more than two branches    27  conditional group contains more than two branches
878    28  assertion expected after (?(    28  assertion expected after (?(
879    29  (?R or (?digits must be followed by )    29  (?R or (?[+-]digits must be followed by )
880    30  unknown POSIX class name    30  unknown POSIX class name
881    31  POSIX collating elements are not supported    31  POSIX collating elements are not supported
882    32  this version of PCRE is not compiled with PCRE_UTF8 support    32  this version of PCRE is compiled without UTF support
883    33  spare error    33  [this code is not in use]
884    34  character value in \ex{...} sequence is too large    34  character value in \ex{...} sequence is too large
885    35  invalid condition (?(0)    35  invalid condition (?(0)
886    36  \eC not allowed in lookbehind assertion    36  \eC not allowed in lookbehind assertion
887    37  PCRE does not support \eL, \el, \eN, \eU, or \eu    37  PCRE does not support \eL, \el, \eN{name}, \eU, or \eu
888    38  number after (?C is > 255    38  number after (?C is > 255
889    39  closing ) for (?C expected    39  closing ) for (?C expected
890    40  recursive call could loop indefinitely    40  recursive call could loop indefinitely
891    41  unrecognized character after (?P    41  unrecognized character after (?P
892    42  syntax error after (?P    42  syntax error in subpattern name (missing terminator)
893    43  two named groups have the same name    43  two named subpatterns have the same name
894    44  invalid UTF-8 string    44  invalid UTF-8 string (specifically UTF-8)
895    45  support for \eP, \ep, and \eX has not been compiled    45  support for \eP, \ep, and \eX has not been compiled
896    46  malformed \eP or \ep sequence    46  malformed \eP or \ep sequence
897    47  unknown property name after \eP or \ep    47  unknown property name after \eP or \ep
898      48  subpattern name is too long (maximum 32 characters)
899      49  too many named subpatterns (maximum 10000)
900      50  [this code is not in use]
901      51  octal value is greater than \e377 in 8-bit non-UTF-8 mode
902      52  internal error: overran compiling workspace
903      53  internal error: previously-checked referenced subpattern
904            not found
905      54  DEFINE group contains more than one branch
906      55  repeating a DEFINE group is not allowed
907      56  inconsistent NEWLINE options
908      57  \eg is not followed by a braced, angle-bracketed, or quoted
909            name/number or by a plain number
910      58  a numbered reference must not be zero
911      59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
912      60  (*VERB) not recognized
913      61  number is too big
914      62  subpattern name expected
915      63  digit expected after (?+
916      64  ] is an invalid data character in JavaScript compatibility mode
917      65  different names for subpatterns of the same number are
918            not allowed
919      66  (*MARK) must have an argument
920      67  this version of PCRE is not compiled with Unicode property
921            support
922      68  \ec must be followed by an ASCII character
923      69  \ek is not followed by a braced, angle-bracketed, or quoted name
924      70  internal error: unknown opcode in find_fixedlength()
925      71  \eN is not supported in a class
926      72  too many forward references
927      73  disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
928      74  invalid UTF-16 string (specifically UTF-16)
929    .sp
930    The numbers 32 and 10000 in errors 48 and 49 are defaults; different values may
931    be used if the limits were changed when PCRE was built.
932  .  .
933  .  .
934    .\" HTML <a name="studyingapattern"></a>
935  .SH "STUDYING A PATTERN"  .SH "STUDYING A PATTERN"
936  .rs  .rs
937  .sp  .sp
# Line 619  help speed up matching, \fBpcre_study()\ Line 948  help speed up matching, \fBpcre_study()\
948  results of the study.  results of the study.
949  .P  .P
950  The returned value from \fBpcre_study()\fP can be passed directly to  The returned value from \fBpcre_study()\fP can be passed directly to
951  \fBpcre_exec()\fP. However, a \fBpcre_extra\fP block also contains other  \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. However, a \fBpcre_extra\fP block
952  fields that can be set by the caller before the block is passed; these are  also contains other fields that can be set by the caller before the block is
953  described  passed; these are described
954  .\" HTML <a href="#extradata">  .\" HTML <a href="#extradata">
955  .\" </a>  .\" </a>
956  below  below
957  .\"  .\"
958  in the section on matching a pattern.  in the section on matching a pattern.
959  .P  .P
960  If studying the pattern does not produce any additional information  If studying the pattern does not produce any useful information,
961  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program  \fBpcre_study()\fP returns NULL. In that circumstance, if the calling program
962  wants to pass any of the other fields to \fBpcre_exec()\fP, it must set up its  wants to pass any of the other fields to \fBpcre_exec()\fP or
963  own \fBpcre_extra\fP block.  \fBpcre_dfa_exec()\fP, it must set up its own \fBpcre_extra\fP block.
964  .P  .P
965  The second argument of \fBpcre_study()\fP contains option bits. At present, no  The second argument of \fBpcre_study()\fP contains option bits. There is only
966  options are defined, and this argument should always be zero.  one option: PCRE_STUDY_JIT_COMPILE. If this is set, and the just-in-time
967    compiler is available, the pattern is further compiled into machine code that
968    executes much faster than the \fBpcre_exec()\fP matching function. If
969    the just-in-time compiler is not available, this option is ignored. All other
970    bits in the \fIoptions\fP argument must be zero.
971    .P
972    JIT compilation is a heavyweight optimization. It can take some time for
973    patterns to be analyzed, and for one-off matches and simple patterns the
974    benefit of faster execution might be offset by a much slower study time.
975    Not all patterns can be optimized by the JIT compiler. For those that cannot be
976    handled, matching automatically falls back to the \fBpcre_exec()\fP
977    interpreter. For more details, see the
978    .\" HREF
979    \fBpcrejit\fP
980    .\"
981    documentation.
982  .P  .P
983  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
984  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
# Line 643  static string that is part of the librar Line 987  static string that is part of the librar
987  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be  should test the error pointer for NULL after calling \fBpcre_study()\fP, to be
988  sure that it has run successfully.  sure that it has run successfully.
989  .P  .P
990  This is a typical call to \fBpcre_study\fP():  When you are finished with a pattern, you can free the memory used for the
991    study data by calling \fBpcre_free_study()\fP. This function was added to the
992    API for release 8.20. For earlier versions, the memory could be freed with
993    \fBpcre_free()\fP, just like the pattern itself. This will still work in cases
994    where PCRE_STUDY_JIT_COMPILE is not used, but it is advisable to change to the
995    new function when convenient.
996    .P
997    This is a typical way in which \fBpcre_study\fP() is used (except that in a
998    real application there should be tests for errors):
999  .sp  .sp
1000    pcre_extra *pe;    int rc;
1001    pe = pcre_study(    pcre *re;
1002      pcre_extra *sd;
1003      re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
1004      sd = pcre_study(
1005      re,             /* result of pcre_compile() */      re,             /* result of pcre_compile() */
1006      0,              /* no options exist */      0,              /* no options */
1007      &error);        /* set to NULL or points to a message */      &error);        /* set to NULL or points to a message */
1008  .sp    rc = pcre_exec(   /* see below for details of pcre_exec() options */
1009  At present, studying a pattern is useful only for non-anchored patterns that do      re, sd, "subject", 7, 0, 0, ovector, 30);
1010  not have a single fixed starting character. A bitmap of possible starting    ...
1011  bytes is created.    pcre_free_study(sd);
1012      pcre_free(re);
1013    .sp
1014    Studying a pattern does two things: first, a lower bound for the length of
1015    subject string that is needed to match the pattern is computed. This does not
1016    mean that there are any strings of that length that match, but it does
1017    guarantee that no shorter strings match. The value is used by
1018    \fBpcre_exec()\fP and \fBpcre_dfa_exec()\fP to avoid wasting time by trying to
1019    match strings that are shorter than the lower bound. You can find out the value
1020    in a calling program via the \fBpcre_fullinfo()\fP function.
1021    .P
1022    Studying a pattern is also useful for non-anchored patterns that do not have a
1023    single fixed starting character. A bitmap of possible starting bytes is
1024    created. This speeds up finding a position in the subject at which to start
1025    matching. (In 16-bit mode, the bitmap is used for 16-bit values less than 256.)
1026    .P
1027    These two optimizations apply to both \fBpcre_exec()\fP and
1028    \fBpcre_dfa_exec()\fP. However, they are not used by \fBpcre_exec()\fP if
1029    \fBpcre_study()\fP is called with the PCRE_STUDY_JIT_COMPILE option, and
1030    just-in-time compiling is successful. The optimizations can be disabled by
1031    setting the PCRE_NO_START_OPTIMIZE option when calling \fBpcre_exec()\fP or
1032    \fBpcre_dfa_exec()\fP. You might want to do this if your pattern contains
1033    callouts or (*MARK) (which cannot be handled by the JIT compiler), and you want
1034    to make use of these facilities in cases where matching fails. See the
1035    discussion of PCRE_NO_START_OPTIMIZE
1036    .\" HTML <a href="#execoptions">
1037    .\" </a>
1038    below.
1039    .\"
1040  .  .
1041  .  .
1042  .\" HTML <a name="localesupport"></a>  .\" HTML <a name="localesupport"></a>
1043  .SH "LOCALE SUPPORT"  .SH "LOCALE SUPPORT"
1044  .rs  .rs
1045  .sp  .sp
1046  PCRE handles caseless matching, and determines whether characters are letters  PCRE handles caseless matching, and determines whether characters are letters,
1047  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
1048  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
1049  less than 128. Higher-valued codes never match escapes such as \ew or \ed, but  with codes less than 128. By default, higher-valued codes never match escapes
1050  can be tested with \ep if PCRE is built with Unicode character property  such as \ew or \ed, but they can be tested with \ep if PCRE is built with
1051  support. The use of locales with Unicode is discouraged.  Unicode character property support. Alternatively, the PCRE_UCP option can be
1052  .P  set at compile time; this causes \ew and friends to use Unicode property
1053  An internal set of tables is created in the default C locale when PCRE is  support instead of built-in tables. The use of locales with Unicode is
1054  built. This is used when the final argument of \fBpcre_compile()\fP is NULL,  discouraged. If you are handling characters with codes greater than 128, you
1055  and is sufficient for many applications. An alternative set of tables can,  should either use UTF-8 and Unicode, or use locales, but not try to mix the
1056  however, be supplied. These may be created in a different locale from the  two.
1057  default. As more and more applications change to using Unicode, the need for  .P
1058  this locale support is expected to die away.  PCRE contains an internal set of tables that are used when the final argument
1059    of \fBpcre_compile()\fP is NULL. These are sufficient for many applications.
1060    Normally, the internal tables recognize only ASCII characters. However, when
1061    PCRE is built, it is possible to cause the internal tables to be rebuilt in the
1062    default "C" locale of the local system, which may cause them to be different.
1063    .P
1064    The internal tables can always be overridden by tables supplied by the
1065    application that calls PCRE. These may be created in a different locale from
1066    the default. As more and more applications change to using Unicode, the need
1067    for this locale support is expected to die away.
1068  .P  .P
1069  External tables are built by calling the \fBpcre_maketables()\fP function,  External tables are built by calling the \fBpcre_maketables()\fP function,
1070  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 685  the following code could be used: Line 1077  the following code could be used:
1077    tables = pcre_maketables();    tables = pcre_maketables();
1078    re = pcre_compile(..., tables);    re = pcre_compile(..., tables);
1079  .sp  .sp
1080    The locale name "fr_FR" is used on Linux and other Unix-like systems; if you
1081    are using Windows, the name for the French locale is "french".
1082    .P
1083  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
1084  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
1085  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 703  one in which it was compiled. Passing ta Line 1098  one in which it was compiled. Passing ta
1098  below in the section on matching a pattern.  below in the section on matching a pattern.
1099  .  .
1100  .  .
1101    .\" HTML <a name="infoaboutpattern"></a>
1102  .SH "INFORMATION ABOUT A PATTERN"  .SH "INFORMATION ABOUT A PATTERN"
1103  .rs  .rs
1104  .sp  .sp
# Line 711  below in the section on matching a patte Line 1107  below in the section on matching a patte
1107  .B int \fIwhat\fP, void *\fIwhere\fP);  .B int \fIwhat\fP, void *\fIwhere\fP);
1108  .PP  .PP
1109  The \fBpcre_fullinfo()\fP function returns information about a compiled  The \fBpcre_fullinfo()\fP function returns information about a compiled
1110  pattern. It replaces the obsolete \fBpcre_info()\fP function, which is  pattern. It replaces the \fBpcre_info()\fP function, which was removed from the
1111  nevertheless retained for backwards compability (and is documented below).  library at version 8.30, after more than 10 years of obsolescence.
1112  .P  .P
1113  The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled  The first argument for \fBpcre_fullinfo()\fP is a pointer to the compiled
1114  pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if  pattern. The second argument is the result of \fBpcre_study()\fP, or NULL if
# Line 721  information is required, and the fourth Line 1117  information is required, and the fourth
1117  to receive the data. The yield of the function is zero for success, or one of  to receive the data. The yield of the function is zero for success, or one of
1118  the following negative numbers:  the following negative numbers:
1119  .sp  .sp
1120    PCRE_ERROR_NULL       the argument \fIcode\fP was NULL    PCRE_ERROR_NULL           the argument \fIcode\fP was NULL
1121                          the argument \fIwhere\fP was NULL                              the argument \fIwhere\fP was NULL
1122    PCRE_ERROR_BADMAGIC   the "magic number" was not found    PCRE_ERROR_BADMAGIC       the "magic number" was not found
1123    PCRE_ERROR_BADOPTION  the value of \fIwhat\fP was invalid    PCRE_ERROR_BADENDIANNESS  the pattern was compiled with different
1124                                endianness
1125      PCRE_ERROR_BADOPTION      the value of \fIwhat\fP was invalid
1126  .sp  .sp
1127  The "magic number" is placed at the start of each compiled pattern as an simple  The "magic number" is placed at the start of each compiled pattern as an simple
1128  check against passing an arbitrary memory pointer. Here is a typical call of  check against passing an arbitrary memory pointer. The endianness error can
1129  \fBpcre_fullinfo()\fP, to obtain the length of the compiled pattern:  occur if a compiled pattern is saved and reloaded on a different host. Here is
1130    a typical call of \fBpcre_fullinfo()\fP, to obtain the length of the compiled
1131    pattern:
1132  .sp  .sp
1133    int rc;    int rc;
1134    unsigned long int length;    size_t length;
1135    rc = pcre_fullinfo(    rc = pcre_fullinfo(
1136      re,               /* result of pcre_compile() */      re,               /* result of pcre_compile() */
1137      pe,               /* result of pcre_study(), or NULL */      sd,               /* result of pcre_study(), or NULL */
1138      PCRE_INFO_SIZE,   /* what is required */      PCRE_INFO_SIZE,   /* what is required */
1139      &length);         /* where to put the data */      &length);         /* where to put the data */
1140  .sp  .sp
# Line 762  a NULL table pointer. Line 1162  a NULL table pointer.
1162  .sp  .sp
1163    PCRE_INFO_FIRSTBYTE    PCRE_INFO_FIRSTBYTE
1164  .sp  .sp
1165  Return information about the first byte of any matched string, for a  Return information about the first data unit of any matched string, for a
1166  non-anchored pattern. (This option used to be called PCRE_INFO_FIRSTCHAR; the  non-anchored pattern. (The name of this option refers to the 8-bit library,
1167  old name is still recognized for backwards compatibility.)  where data units are bytes.) The fourth argument should point to an \fBint\fP
1168  .P  variable.
1169  If there is a fixed first byte, for example, from a pattern such as  .P
1170  (cat|cow|coyote), it is returned in the integer pointed to by \fIwhere\fP.  If there is a fixed first value, for example, the letter "c" from a pattern
1171  Otherwise, if either  such as (cat|cow|coyote), its value is returned. In the 8-bit library, the
1172    value is always less than 256; in the 16-bit library the value can be up to
1173    0xffff.
1174    .P
1175    If there is no fixed first value, and if either
1176  .sp  .sp
1177  (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
1178  starts with "^", or  starts with "^", or
# Line 783  returned. For anchored patterns, -2 is r Line 1187  returned. For anchored patterns, -2 is r
1187    PCRE_INFO_FIRSTTABLE    PCRE_INFO_FIRSTTABLE
1188  .sp  .sp
1189  If the pattern was studied, and this resulted in the construction of a 256-bit  If the pattern was studied, and this resulted in the construction of a 256-bit
1190  table indicating a fixed set of bytes for the first byte in any matching  table indicating a fixed set of values for the first data unit in any matching
1191  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
1192  fourth argument should point to an \fBunsigned char *\fP variable.  fourth argument should point to an \fBunsigned char *\fP variable.
1193  .sp  .sp
1194      PCRE_INFO_HASCRORLF
1195    .sp
1196    Return 1 if the pattern contains any explicit matches for CR or LF characters,
1197    otherwise 0. The fourth argument should point to an \fBint\fP variable. An
1198    explicit match is either a literal CR or LF character, or \er or \en.
1199    .sp
1200      PCRE_INFO_JCHANGED
1201    .sp
1202    Return 1 if the (?J) or (?-J) option setting is used in the pattern, otherwise
1203    0. The fourth argument should point to an \fBint\fP variable. (?J) and
1204    (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1205    .sp
1206      PCRE_INFO_JIT
1207    .sp
1208    Return 1 if the pattern was studied with the PCRE_STUDY_JIT_COMPILE option, and
1209    just-in-time compiling was successful. The fourth argument should point to an
1210    \fBint\fP variable. A return value of 0 means that JIT support is not available
1211    in this version of PCRE, or that the pattern was not studied with the
1212    PCRE_STUDY_JIT_COMPILE option, or that the JIT compiler could not handle this
1213    particular pattern. See the
1214    .\" HREF
1215    \fBpcrejit\fP
1216    .\"
1217    documentation for details of what can and cannot be handled.
1218    .sp
1219      PCRE_INFO_JITSIZE
1220    .sp
1221    If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1222    return the size of the JIT compiled code, otherwise return zero. The fourth
1223    argument should point to a \fBsize_t\fP variable.
1224    .sp
1225    PCRE_INFO_LASTLITERAL    PCRE_INFO_LASTLITERAL
1226  .sp  .sp
1227  Return the value of the rightmost literal byte that must exist in any matched  Return the value of the rightmost literal data unit that must exist in any
1228  string, other than at its start, if such a byte has been recorded. The fourth  matched string, other than at its start, if such a value has been recorded. The
1229  argument should point to an \fBint\fP variable. If there is no such byte, -1 is  fourth argument should point to an \fBint\fP variable. If there is no such
1230  returned. For anchored patterns, a last literal byte is recorded only if it  value, -1 is returned. For anchored patterns, a last literal value is recorded
1231  follows something of variable length. For example, for the pattern  only if it follows something of variable length. For example, for the pattern
1232  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value  /^a\ed+z\ed+/ the returned value is "z", but for /^a\edz\ed/ the returned value
1233  is -1.  is -1.
1234  .sp  .sp
1235      PCRE_INFO_MINLENGTH
1236    .sp
1237    If the pattern was studied and a minimum length for matching subject strings
1238    was computed, its value is returned. Otherwise the returned value is -1. The
1239    value is a number of characters, which in UTF-8 mode may be different from the
1240    number of bytes. The fourth argument should point to an \fBint\fP variable. A
1241    non-negative value is a lower bound to the length of any matching string. There
1242    may not be any strings of that length that do actually match, but every string
1243    that does match is at least that long.
1244    .sp
1245    PCRE_INFO_NAMECOUNT    PCRE_INFO_NAMECOUNT
1246    PCRE_INFO_NAMEENTRYSIZE    PCRE_INFO_NAMEENTRYSIZE
1247    PCRE_INFO_NAMETABLE    PCRE_INFO_NAMETABLE
1248  .sp  .sp
1249  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
1250  names are just an additional way of identifying the parentheses, which still  names are just an additional way of identifying the parentheses, which still
1251  acquire numbers. A convenience function called \fBpcre_get_named_substring()\fP  acquire numbers. Several convenience functions such as
1252  is provided for extracting an individual captured substring by name. It is also  \fBpcre_get_named_substring()\fP are provided for extracting captured
1253  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
1254  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
1255  \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,
1256  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
1257    values.
1258  .P  .P
1259  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
1260  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
1261  entry; both of these return an \fBint\fP value. The entry size depends on the  entry; both of these return an \fBint\fP value. The entry size depends on the
1262  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first  length of the longest name. PCRE_INFO_NAMETABLE returns a pointer to the first
1263  entry of the table (a pointer to \fBchar\fP). The first two bytes of each entry  entry of the table. This is a pointer to \fBchar\fP in the 8-bit library, where
1264  are the number of the capturing parenthesis, most significant byte first. The  the first two bytes of each entry are the number of the capturing parenthesis,
1265  rest of the entry is the corresponding name, zero terminated. The names are in  most significant byte first. In the 16-bit library, the pointer points to
1266  alphabetical order. For example, consider the following pattern (assume  16-bit data units, the first of which contains the parenthesis number. The rest
1267  PCRE_EXTENDED is set, so white space - including newlines - is ignored):  of the entry is the corresponding name, zero terminated.
1268    .P
1269    The names are in alphabetical order. Duplicate names may appear if (?| is used
1270    to create multiple groups with the same number, as described in the
1271    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
1272    .\" </a>
1273    section on duplicate subpattern numbers
1274    .\"
1275    in the
1276    .\" HREF
1277    \fBpcrepattern\fP
1278    .\"
1279    page. Duplicate names for subpatterns with different numbers are permitted only
1280    if PCRE_DUPNAMES is set. In all cases of duplicate names, they appear in the
1281    table in the order in which they were found in the pattern. In the absence of
1282    (?| this is the order of increasing number; when (?| is used this is not
1283    necessarily the case because later subpatterns may have lower numbers.
1284    .P
1285    As a simple example of the name/number table, consider the following pattern
1286    after compilation by the 8-bit library (assume PCRE_EXTENDED is set, so white
1287    space - including newlines - is ignored):
1288  .sp  .sp
1289  .\" JOIN  .\" JOIN
1290    (?P<date> (?P<year>(\ed\ed)?\ed\ed) -    (?<date> (?<year>(\ed\ed)?\ed\ed) -
1291    (?P<month>\ed\ed) - (?P<day>\ed\ed) )    (?<month>\ed\ed) - (?<day>\ed\ed) )
1292  .sp  .sp
1293  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
1294  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 834  bytes shows in hexadecimal, and undefine Line 1300  bytes shows in hexadecimal, and undefine
1300    00 02 y  e  a  r  00 ??    00 02 y  e  a  r  00 ??
1301  .sp  .sp
1302  When writing code to extract data from named subpatterns using the  When writing code to extract data from named subpatterns using the
1303  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
1304  different for each compiled pattern.  different for each compiled pattern.
1305  .sp  .sp
1306      PCRE_INFO_OKPARTIAL
1307    .sp
1308    Return 1 if the pattern can be used for partial matching with
1309    \fBpcre_exec()\fP, otherwise 0. The fourth argument should point to an
1310    \fBint\fP variable. From release 8.00, this always returns 1, because the
1311    restrictions that previously applied to partial matching have been lifted. The
1312    .\" HREF
1313    \fBpcrepartial\fP
1314    .\"
1315    documentation gives details of partial matching.
1316    .sp
1317    PCRE_INFO_OPTIONS    PCRE_INFO_OPTIONS
1318  .sp  .sp
1319  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
1320  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
1321  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
1322  top-level option settings within the pattern itself.  top-level option settings at the start of the pattern itself. In other words,
1323    they are the options that will be in force when matching starts. For example,
1324    if the pattern /(?im)abc(?-i)d/ is compiled with the PCRE_EXTENDED option, the
1325    result is PCRE_CASELESS, PCRE_MULTILINE, and PCRE_EXTENDED.
1326  .P  .P
1327  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
1328  alternatives begin with one of the following:  alternatives begin with one of the following:
# Line 859  For such patterns, the PCRE_ANCHORED bit Line 1339  For such patterns, the PCRE_ANCHORED bit
1339  .sp  .sp
1340    PCRE_INFO_SIZE    PCRE_INFO_SIZE
1341  .sp  .sp
1342  Return the size of the compiled pattern, that is, the value that was passed as  Return the size of the compiled pattern in bytes (for both libraries). The
1343  the argument to \fBpcre_malloc()\fP when PCRE was getting memory in which to  fourth argument should point to a \fBsize_t\fP variable. This value does not
1344  place the compiled data. The fourth argument should point to a \fBsize_t\fP  include the size of the \fBpcre\fP structure that is returned by
1345  variable.  \fBpcre_compile()\fP. The value that is passed as the argument to
1346    \fBpcre_malloc()\fP when \fBpcre_compile()\fP is getting memory in which to
1347    place the compiled data is the value returned by this option plus the size of
1348    the \fBpcre\fP structure. Studying a compiled pattern, with or without JIT,
1349    does not alter the value returned by this option.
1350  .sp  .sp
1351    PCRE_INFO_STUDYSIZE    PCRE_INFO_STUDYSIZE
1352  .sp  .sp
1353  Return the size of the data block pointed to by the \fIstudy_data\fP field in  Return the size in bytes of the data block pointed to by the \fIstudy_data\fP
1354  a \fBpcre_extra\fP block. That is, it is the value that was passed to  field in a \fBpcre_extra\fP block. If \fBpcre_extra\fP is NULL, or there is no
1355  \fBpcre_malloc()\fP when PCRE was getting memory into which to place the data  study data, zero is returned. The fourth argument should point to a
1356  created by \fBpcre_study()\fP. The fourth argument should point to a  \fBsize_t\fP variable. The \fIstudy_data\fP field is set by \fBpcre_study()\fP
1357  \fBsize_t\fP variable.  to record information that will speed up matching (see the section entitled
1358  .  .\" HTML <a href="#studyingapattern">
1359  .  .\" </a>
1360  .SH "OBSOLETE INFO FUNCTION"  "Studying a pattern"
1361  .rs  .\"
1362  .sp  above). The format of the \fIstudy_data\fP block is private, but its length
1363  .B int pcre_info(const pcre *\fIcode\fP, int *\fIoptptr\fP, int  is made available via this option so that it can be saved and restored (see the
1364  .B *\fIfirstcharptr\fP);  .\" HREF
1365  .PP  \fBpcreprecompile\fP
1366  The \fBpcre_info()\fP function is now obsolete because its interface is too  .\"
1367  restrictive to return all the available data about a compiled pattern. New  documentation for details).
 programs should use \fBpcre_fullinfo()\fP instead. The yield of  
 \fBpcre_info()\fP is the number of capturing subpatterns, or one of the  
 following negative numbers:  
 .sp  
   PCRE_ERROR_NULL       the argument \fIcode\fP was NULL  
   PCRE_ERROR_BADMAGIC   the "magic number" was not found  
 .sp  
 If the \fIoptptr\fP argument is not NULL, a copy of the options with which the  
 pattern was compiled is placed in the integer it points to (see  
 PCRE_INFO_OPTIONS above).  
 .P  
 If the pattern is not anchored and the \fIfirstcharptr\fP argument is not NULL,  
 it is used to pass back information about the first character of any matched  
 string (see PCRE_INFO_FIRSTBYTE above).  
1368  .  .
1369  .  .
1370  .SH "REFERENCE COUNTS"  .SH "REFERENCE COUNTS"
# Line 931  is different. (This seems a highly unlik Line 1401  is different. (This seems a highly unlik
1401  .P  .P
1402  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
1403  compiled pattern, which is passed in the \fIcode\fP argument. If the  compiled pattern, which is passed in the \fIcode\fP argument. If the
1404  pattern has been studied, the result of the study should be passed in the  pattern was studied, the result of the study should be passed in the
1405  \fIextra\fP argument. This function is the main matching facility of the  \fIextra\fP argument. You can call \fBpcre_exec()\fP with the same \fIcode\fP
1406  library, and it operates in a Perl-like manner. For specialist use there is  and \fIextra\fP arguments as many times as you like, in order to match
1407  also an alternative matching function, which is described  different subject strings with the same pattern.
1408    .P
1409    This function is the main matching facility of the library, and it operates in
1410    a Perl-like manner. For specialist use there is also an alternative matching
1411    function, which is described
1412  .\" HTML <a href="#dfamatch">  .\" HTML <a href="#dfamatch">
1413  .\" </a>  .\" </a>
1414  below  below
# Line 965  Here is an example of a simple call to \ Line 1439  Here is an example of a simple call to \
1439      ovector,        /* vector of integers for substring information */      ovector,        /* vector of integers for substring information */
1440      30);            /* number of elements (NOT size in bytes) */      30);            /* number of elements (NOT size in bytes) */
1441  .  .
1442    .
1443  .\" HTML <a name="extradata"></a>  .\" HTML <a name="extradata"></a>
1444  .SS "Extra data for \fBpcre_exec()\fR"  .SS "Extra data for \fBpcre_exec()\fR"
1445  .rs  .rs
# Line 977  fields (not necessarily in this order): Line 1452  fields (not necessarily in this order):
1452  .sp  .sp
1453    unsigned long int \fIflags\fP;    unsigned long int \fIflags\fP;
1454    void *\fIstudy_data\fP;    void *\fIstudy_data\fP;
1455      void *\fIexecutable_jit\fP;
1456    unsigned long int \fImatch_limit\fP;    unsigned long int \fImatch_limit\fP;
1457    unsigned long int \fImatch_limit_recursion\fP;    unsigned long int \fImatch_limit_recursion\fP;
1458    void *\fIcallout_data\fP;    void *\fIcallout_data\fP;
1459    const unsigned char *\fItables\fP;    const unsigned char *\fItables\fP;
1460      unsigned char **\fImark\fP;
1461  .sp  .sp
1462    In the 16-bit version of this structure, the \fImark\fP field has type
1463    "PCRE_UCHAR16 **".
1464    .P
1465  The \fIflags\fP field is a bitmap that specifies which of the other fields  The \fIflags\fP field is a bitmap that specifies which of the other fields
1466  are set. The flag bits are:  are set. The flag bits are:
1467  .sp  .sp
1468    PCRE_EXTRA_STUDY_DATA    PCRE_EXTRA_STUDY_DATA
1469      PCRE_EXTRA_EXECUTABLE_JIT
1470    PCRE_EXTRA_MATCH_LIMIT    PCRE_EXTRA_MATCH_LIMIT
1471    PCRE_EXTRA_MATCH_LIMIT_RECURSION    PCRE_EXTRA_MATCH_LIMIT_RECURSION
1472    PCRE_EXTRA_CALLOUT_DATA    PCRE_EXTRA_CALLOUT_DATA
1473    PCRE_EXTRA_TABLES    PCRE_EXTRA_TABLES
1474      PCRE_EXTRA_MARK
1475  .sp  .sp
1476  Other flag bits should be set to zero. The \fIstudy_data\fP field is set in the  Other flag bits should be set to zero. The \fIstudy_data\fP field and sometimes
1477  \fBpcre_extra\fP block that is returned by \fBpcre_study()\fP, together with  the \fIexecutable_jit\fP field are set in the \fBpcre_extra\fP block that is
1478  the appropriate flag bit. You should not set this yourself, but you may add to  returned by \fBpcre_study()\fP, together with the appropriate flag bits. You
1479  the block by setting the other fields and their corresponding flag bits.  should not set these yourself, but you may add to the block by setting the
1480    other fields and their corresponding flag bits.
1481  .P  .P
1482  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a  The \fImatch_limit\fP field provides a means of preventing PCRE from using up a
1483  vast amount of resources when running patterns that are not going to match,  vast amount of resources when running patterns that are not going to match,
1484  but which have a very large number of possibilities in their search trees. The  but which have a very large number of possibilities in their search trees. The
1485  classic example is the use of nested unlimited repeats.  classic example is a pattern that uses nested unlimited repeats.
1486  .P  .P
1487  Internally, PCRE uses a function called \fBmatch()\fP which it calls repeatedly  Internally, \fBpcre_exec()\fP uses a function called \fBmatch()\fP, which it
1488  (sometimes recursively). The limit set by \fImatch_limit\fP is imposed on the  calls repeatedly (sometimes recursively). The limit set by \fImatch_limit\fP is
1489  number of times this function is called during a match, which has the effect of  imposed on the number of times this function is called during a match, which
1490  limiting the amount of backtracking that can take place. For patterns that are  has the effect of limiting the amount of backtracking that can take place. For
1491  not anchored, the count restarts from zero for each position in the subject  patterns that are not anchored, the count restarts from zero for each position
1492  string.  in the subject string.
1493    .P
1494    When \fBpcre_exec()\fP is called with a pattern that was successfully studied
1495    with the PCRE_STUDY_JIT_COMPILE option, the way that the matching is executed
1496    is entirely different. However, there is still the possibility of runaway
1497    matching that goes on for a very long time, and so the \fImatch_limit\fP value
1498    is also used in this case (but in a different way) to limit how long the
1499    matching can continue.
1500  .P  .P
1501  The default value for the limit can be set when PCRE is built; the default  The default value for the limit can be set when PCRE is built; the default
1502  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
# Line 1021  limits the depth of recursion. The recur Line 1511  limits the depth of recursion. The recur
1511  total number of calls, because not all calls to \fBmatch()\fP are recursive.  total number of calls, because not all calls to \fBmatch()\fP are recursive.
1512  This limit is of use only if it is set smaller than \fImatch_limit\fP.  This limit is of use only if it is set smaller than \fImatch_limit\fP.
1513  .P  .P
1514  Limiting the recursion depth limits the amount of stack that can be used, or,  Limiting the recursion depth limits the amount of machine stack that can be
1515  when PCRE has been compiled to use memory on the heap instead of the stack, the  used, or, when PCRE has been compiled to use memory on the heap instead of the
1516  amount of heap memory that can be used.  stack, the amount of heap memory that can be used. This limit is not relevant,
1517    and is ignored, if the pattern was successfully studied with
1518    PCRE_STUDY_JIT_COMPILE.
1519  .P  .P
1520  The default value for \fImatch_limit_recursion\fP can be set when PCRE is  The default value for \fImatch_limit_recursion\fP can be set when PCRE is
1521  built; the default default is the same value as the default for  built; the default default is the same value as the default for
# Line 1032  with a \fBpcre_extra\fP block in which \ Line 1524  with a \fBpcre_extra\fP block in which \
1524  PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit  PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the \fIflags\fP field. If the limit
1525  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.  is exceeded, \fBpcre_exec()\fP returns PCRE_ERROR_RECURSIONLIMIT.
1526  .P  .P
1527  The \fIpcre_callout\fP field is used in conjunction with the "callout" feature,  The \fIcallout_data\fP field is used in conjunction with the "callout" feature,
1528  which is described in the  and is described in the
1529  .\" HREF  .\" HREF
1530  \fBpcrecallout\fP  \fBpcrecallout\fP
1531  .\"  .\"
# Line 1052  called. See the Line 1544  called. See the
1544  \fBpcreprecompile\fP  \fBpcreprecompile\fP
1545  .\"  .\"
1546  documentation for a discussion of saving compiled patterns for later use.  documentation for a discussion of saving compiled patterns for later use.
1547    .P
1548    If PCRE_EXTRA_MARK is set in the \fIflags\fP field, the \fImark\fP field must
1549    be set to point to a suitable variable. If the pattern contains any
1550    backtracking control verbs such as (*MARK:NAME), and the execution ends up with
1551    a name to pass back, a pointer to the name string (zero terminated) is placed
1552    in the variable pointed to by the \fImark\fP field. The names are within the
1553    compiled pattern; if you wish to retain such a name you must copy it before
1554    freeing the memory of a compiled pattern. If there is no name to pass back, the
1555    variable pointed to by the \fImark\fP field is set to NULL. For details of the
1556    backtracking control verbs, see the section entitled
1557    .\" HTML <a href="pcrepattern#backtrackcontrol">
1558    .\" </a>
1559    "Backtracking control"
1560    .\"
1561    in the
1562    .\" HREF
1563    \fBpcrepattern\fP
1564    .\"
1565    documentation.
1566    .
1567  .  .
1568    .\" HTML <a name="execoptions"></a>
1569  .SS "Option bits for \fBpcre_exec()\fP"  .SS "Option bits for \fBpcre_exec()\fP"
1570  .rs  .rs
1571  .sp  .sp
1572  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
1573  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,
1574  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and PCRE_PARTIAL.  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1575    PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_SOFT, and
1576    PCRE_PARTIAL_HARD.
1577    .P
1578    If the pattern was successfully studied with the PCRE_STUDY_JIT_COMPILE option,
1579    the only supported options for JIT execution are PCRE_NO_UTF8_CHECK,
1580    PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, and PCRE_NOTEMPTY_ATSTART. Note in
1581    particular that partial matching is not supported. If an unsupported option is
1582    used, JIT execution is disabled and the normal interpretive code in
1583    \fBpcre_exec()\fP is run.
1584  .sp  .sp
1585    PCRE_ANCHORED    PCRE_ANCHORED
1586  .sp  .sp
# Line 1067  matching position. If a pattern was comp Line 1589  matching position. If a pattern was comp
1589  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
1590  matching time.  matching time.
1591  .sp  .sp
1592      PCRE_BSR_ANYCRLF
1593      PCRE_BSR_UNICODE
1594    .sp
1595    These options (which are mutually exclusive) control what the \eR escape
1596    sequence matches. The choice is either to match only CR, LF, or CRLF, or to
1597    match any Unicode newline sequence. These options override the choice that was
1598    made or defaulted when the pattern was compiled.
1599    .sp
1600      PCRE_NEWLINE_CR
1601      PCRE_NEWLINE_LF
1602      PCRE_NEWLINE_CRLF
1603      PCRE_NEWLINE_ANYCRLF
1604      PCRE_NEWLINE_ANY
1605    .sp
1606    These options override the newline definition that was chosen or defaulted when
1607    the pattern was compiled. For details, see the description of
1608    \fBpcre_compile()\fP above. During matching, the newline choice affects the
1609    behaviour of the dot, circumflex, and dollar metacharacters. It may also alter
1610    the way the match position is advanced after a match failure for an unanchored
1611    pattern.
1612    .P
1613    When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is set, and a
1614    match attempt for an unanchored pattern fails when the current position is at a
1615    CRLF sequence, and the pattern contains no explicit matches for CR or LF
1616    characters, the match position is advanced by two characters instead of one, in
1617    other words, to after the CRLF.
1618    .P
1619    The above rule is a compromise that makes the most common cases work as
1620    expected. For example, if the pattern is .+A (and the PCRE_DOTALL option is not
1621    set), it does not match the string "\er\enA" because, after failing at the
1622    start, it skips both the CR and the LF before retrying. However, the pattern
1623    [\er\en]A does match that string, because it contains an explicit CR or LF
1624    reference, and so advances only by one character after the first failure.
1625    .P
1626    An explicit match for CR of LF is either a literal appearance of one of those
1627    characters, or one of the \er or \en escape sequences. Implicit matches such as
1628    [^X] do not count, nor does \es (which includes CR and LF in the characters
1629    that it matches).
1630    .P
1631    Notwithstanding the above, anomalous effects may still occur when CRLF is a
1632    valid newline sequence and explicit \er or \en escapes appear in the pattern.
1633    .sp
1634    PCRE_NOTBOL    PCRE_NOTBOL
1635  .sp  .sp
1636  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 1091  match the empty string, the entire match Line 1655  match the empty string, the entire match
1655  .sp  .sp
1656    a?b?    a?b?
1657  .sp  .sp
1658  is applied to a string not beginning with "a" or "b", it matches the empty  is applied to a string not beginning with "a" or "b", it matches an empty
1659  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not  string at the start of the subject. With PCRE_NOTEMPTY set, this match is not
1660  valid, so PCRE searches further into the string for occurrences of "a" or "b".  valid, so PCRE searches further into the string for occurrences of "a" or "b".
1661    .sp
1662      PCRE_NOTEMPTY_ATSTART
1663    .sp
1664    This is like PCRE_NOTEMPTY, except that an empty string match that is not at
1665    the start of the subject is permitted. If the pattern is anchored, such a match
1666    can occur only if the pattern contains \eK.
1667    .P
1668    Perl has no direct equivalent of PCRE_NOTEMPTY or PCRE_NOTEMPTY_ATSTART, but it
1669    does make a special case of a pattern match of the empty string within its
1670    \fBsplit()\fP function, and when using the /g modifier. It is possible to
1671    emulate Perl's behaviour after matching a null string by first trying the match
1672    again at the same offset with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then
1673    if that fails, by advancing the starting offset (see below) and trying an
1674    ordinary match again. There is some code that demonstrates how to do this in
1675    the
1676    .\" HREF
1677    \fBpcredemo\fP
1678    .\"
1679    sample program. In the most general case, you have to check to see if the
1680    newline convention recognizes CRLF as a newline, and if so, and the current
1681    character is CR followed by LF, advance the starting offset by two characters
1682    instead of one.
1683    .sp
1684      PCRE_NO_START_OPTIMIZE
1685    .sp
1686    There are a number of optimizations that \fBpcre_exec()\fP uses at the start of
1687    a match, in order to speed up the process. For example, if it is known that an
1688    unanchored match must start with a specific character, it searches the subject
1689    for that character, and fails immediately if it cannot find it, without
1690    actually running the main matching function. This means that a special item
1691    such as (*COMMIT) at the start of a pattern is not considered until after a
1692    suitable starting point for the match has been found. When callouts or (*MARK)
1693    items are in use, these "start-up" optimizations can cause them to be skipped
1694    if the pattern is never actually used. The start-up optimizations are in effect
1695    a pre-scan of the subject that takes place before the pattern is run.
1696    .P
1697    The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations, possibly
1698    causing performance to suffer, but ensuring that in cases where the result is
1699    "no match", the callouts do occur, and that items such as (*COMMIT) and (*MARK)
1700    are considered at every possible starting position in the subject string. If
1701    PCRE_NO_START_OPTIMIZE is set at compile time, it cannot be unset at matching
1702    time.
1703  .P  .P
1704  Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a special case  Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching operation.
1705  of a pattern match of the empty string within its \fBsplit()\fP function, and  Consider the pattern
1706  when using the /g modifier. It is possible to emulate Perl's behaviour after  .sp
1707  matching a null string by first trying the match again at the same offset with    (*COMMIT)ABC
1708  PCRE_NOTEMPTY and PCRE_ANCHORED, and then if that fails by advancing the  .sp
1709  starting offset (see below) and trying an ordinary match again. There is some  When this is compiled, PCRE records the fact that a match must start with the
1710  code that demonstrates how to do this in the \fIpcredemo.c\fP sample program.  character "A". Suppose the subject string is "DEFABC". The start-up
1711    optimization scans along the subject, finds "A" and runs the first match
1712    attempt from there. The (*COMMIT) item means that the pattern must match the
1713    current starting position, which in this case, it does. However, if the same
1714    match is run with PCRE_NO_START_OPTIMIZE set, the initial scan along the
1715    subject string does not happen. The first match attempt is run starting from
1716    "D" and when this fails, (*COMMIT) prevents any further matches being tried, so
1717    the overall result is "no match". If the pattern is studied, more start-up
1718    optimizations may be used. For example, a minimum length for the subject may be
1719    recorded. Consider the pattern
1720    .sp
1721      (*MARK:A)(X|Y)
1722    .sp
1723    The minimum length for a match is one character. If the subject is "ABC", there
1724    will be attempts to match "ABC", "BC", "C", and then finally an empty string.
1725    If the pattern is studied, the final attempt does not take place, because PCRE
1726    knows that the subject is too short, and so the (*MARK) is never encountered.
1727    In this case, studying the pattern does not affect the overall match result,
1728    which is still "no match", but it does affect the auxiliary information that is
1729    returned.
1730  .sp  .sp
1731    PCRE_NO_UTF8_CHECK    PCRE_NO_UTF8_CHECK
1732  .sp  .sp
1733  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
1734  string is automatically checked when \fBpcre_exec()\fP is subsequently called.  string is automatically checked when \fBpcre_exec()\fP is subsequently called.
1735  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
1736  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
1737  \fBpcre_exec()\fP returns the error PCRE_ERROR_BADUTF8. If \fIstartoffset\fP  strings in the
1738  contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.  .\" HREF
1739    \fBpcreunicode\fP
1740    .\"
1741    page. If an invalid sequence of bytes is found, \fBpcre_exec()\fP returns the
1742    error PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
1743    truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In both
1744    cases, information about the precise nature of the error may also be returned
1745    (see the descriptions of these errors in the section entitled \fIError return
1746    values from\fP \fBpcre_exec()\fP
1747    .\" HTML <a href="#errorlist">
1748    .\" </a>
1749    below).
1750    .\"
1751    If \fIstartoffset\fP contains a value that does not point to the start of a
1752    UTF-8 character (or to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is
1753    returned.
1754  .P  .P
1755  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
1756  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
1757  calling \fBpcre_exec()\fP. You might want to do this for the second and  calling \fBpcre_exec()\fP. You might want to do this for the second and
1758  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find  subsequent calls to \fBpcre_exec()\fP if you are making repeated calls to find
1759  all the matches in a single subject string. However, you should be sure that  all the matches in a single subject string. However, you should be sure that
1760  the value of \fIstartoffset\fP points to the start of a UTF-8 character. When  the value of \fIstartoffset\fP points to the start of a character (or the end
1761  PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid UTF-8 string as a  of the subject). When PCRE_NO_UTF8_CHECK is set, the effect of passing an
1762  subject, or a value of \fIstartoffset\fP that does not point to the start of a  invalid string as a subject or an invalid value of \fIstartoffset\fP is
1763  UTF-8 character, is undefined. Your program may crash.  undefined. Your program may crash.
1764  .sp  .sp
1765    PCRE_PARTIAL    PCRE_PARTIAL_HARD
1766  .sp    PCRE_PARTIAL_SOFT
1767  This option turns on the partial matching feature. If the subject string fails  .sp
1768  to match the pattern, but at some point during the matching process the end of  These options turn on the partial matching feature. For backwards
1769  the subject was reached (that is, the subject partially matches the pattern and  compatibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial match
1770  the failure to match occurred only because there were not enough subject  occurs if the end of the subject string is reached successfully, but there are
1771  characters), \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL instead of  not enough subject characters to complete the match. If this happens when
1772  PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is used, there are restrictions on what  PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set, matching continues by
1773  may appear in the pattern. These are discussed in the  testing any remaining alternatives. Only if no complete match can be found is
1774    PCRE_ERROR_PARTIAL returned instead of PCRE_ERROR_NOMATCH. In other words,
1775    PCRE_PARTIAL_SOFT says that the caller is prepared to handle a partial match,
1776    but only if no complete match can be found.
1777    .P
1778    If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this case, if a
1779    partial match is found, \fBpcre_exec()\fP immediately returns
1780    PCRE_ERROR_PARTIAL, without considering any other alternatives. In other words,
1781    when PCRE_PARTIAL_HARD is set, a partial match is considered to be more
1782    important that an alternative complete match.
1783    .P
1784    In both cases, the portion of the string that was inspected when the partial
1785    match was found is set as the first matching string. There is a more detailed
1786    discussion of partial and multi-segment matching, with examples, in the
1787  .\" HREF  .\" HREF
1788  \fBpcrepartial\fP  \fBpcrepartial\fP
1789  .\"  .\"
1790  documentation.  documentation.
1791  .  .
1792    .
1793  .SS "The string to be matched by \fBpcre_exec()\fP"  .SS "The string to be matched by \fBpcre_exec()\fP"
1794  .rs  .rs
1795  .sp  .sp
1796  The subject string is passed to \fBpcre_exec()\fP as a pointer in  The subject string is passed to \fBpcre_exec()\fP as a pointer in
1797  \fIsubject\fP, a length in \fIlength\fP, and a starting byte offset in  \fIsubject\fP, a length in bytes in \fIlength\fP, and a starting byte offset
1798  \fIstartoffset\fP. In UTF-8 mode, the byte offset must point to the start of a  in \fIstartoffset\fP. If this is negative or greater than the length of the
1799  UTF-8 character. Unlike the pattern string, the subject may contain binary zero  subject, \fBpcre_exec()\fP returns PCRE_ERROR_BADOFFSET. When the starting
1800  bytes. When the starting offset is zero, the search for a match starts at the  offset is zero, the search for a match starts at the beginning of the subject,
1801  beginning of the subject, and this is by far the most common case.  and this is by far the most common case. In UTF-8 mode, the byte offset must
1802    point to the start of a UTF-8 character (or the end of the subject). Unlike the
1803    pattern string, the subject may contain binary zero bytes.
1804  .P  .P
1805  A non-zero starting offset is useful when searching for another match in the  A non-zero starting offset is useful when searching for another match in the
1806  same subject by calling \fBpcre_exec()\fP again after a previous success.  same subject by calling \fBpcre_exec()\fP again after a previous success.
# Line 1164  start of the subject, which is deemed to Line 1820  start of the subject, which is deemed to
1820  set to 4, it finds the second occurrence of "iss" because it is able to look  set to 4, it finds the second occurrence of "iss" because it is able to look
1821  behind the starting point to discover that it is preceded by a letter.  behind the starting point to discover that it is preceded by a letter.
1822  .P  .P
1823    Finding all the matches in a subject is tricky when the pattern can match an
1824    empty string. It is possible to emulate Perl's /g behaviour by first trying the
1825    match again at the same offset, with the PCRE_NOTEMPTY_ATSTART and
1826    PCRE_ANCHORED options, and then if that fails, advancing the starting offset
1827    and trying an ordinary match again. There is some code that demonstrates how to
1828    do this in the
1829    .\" HREF
1830    \fBpcredemo\fP
1831    .\"
1832    sample program. In the most general case, you have to check to see if the
1833    newline convention recognizes CRLF as a newline, and if so, and the current
1834    character is CR followed by LF, advance the starting offset by two characters
1835    instead of one.
1836    .P
1837  If a non-zero starting offset is passed when the pattern is anchored, one  If a non-zero starting offset is passed when the pattern is anchored, one
1838  attempt to match at the given offset is made. This can only succeed if the  attempt to match at the given offset is made. This can only succeed if the
1839  pattern does not require the match to be at the start of the subject.  pattern does not require the match to be at the start of the subject.
1840  .  .
1841    .
1842  .SS "How \fBpcre_exec()\fP returns captured substrings"  .SS "How \fBpcre_exec()\fP returns captured substrings"
1843  .rs  .rs
1844  .sp  .sp
# Line 1178  pattern. Following the usage in Jeffrey Line 1849  pattern. Following the usage in Jeffrey
1849  a fragment of a pattern that picks out a substring. PCRE supports several other  a fragment of a pattern that picks out a substring. PCRE supports several other
1850  kinds of parenthesized subpattern that do not cause substrings to be captured.  kinds of parenthesized subpattern that do not cause substrings to be captured.
1851  .P  .P
1852  Captured substrings are returned to the caller via a vector of integer offsets  Captured substrings are returned to the caller via a vector of integers whose
1853  whose address is passed in \fIovector\fP. The number of elements in the vector  address is passed in \fIovector\fP. The number of elements in the vector is
1854  is passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP:  passed in \fIovecsize\fP, which must be a non-negative number. \fBNote\fP: this
1855  this argument is NOT the size of \fIovector\fP in bytes.  argument is NOT the size of \fIovector\fP in bytes.
1856  .P  .P
1857  The first two-thirds of the vector is used to pass back captured substrings,  The first two-thirds of the vector is used to pass back captured substrings,
1858  each substring using a pair of integers. The remaining third of the vector is  each substring using a pair of integers. The remaining third of the vector is
1859  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,  used as workspace by \fBpcre_exec()\fP while matching capturing subpatterns,
1860  and is not available for passing back information. The length passed in  and is not available for passing back information. The number passed in
1861  \fIovecsize\fP should always be a multiple of three. If it is not, it is  \fIovecsize\fP should always be a multiple of three. If it is not, it is
1862  rounded down.  rounded down.
1863  .P  .P
1864  When a match is successful, information about captured substrings is returned  When a match is successful, information about captured substrings is returned
1865  in pairs of integers, starting at the beginning of \fIovector\fP, and  in pairs of integers, starting at the beginning of \fIovector\fP, and
1866  continuing up to two-thirds of its length at the most. The first element of a  continuing up to two-thirds of its length at the most. The first element of
1867  pair is set to the offset of the first character in a substring, and the second  each pair is set to the byte offset of the first character in a substring, and
1868  is set to the offset of the first character after the end of a substring. The  the second is set to the byte offset of the first character after the end of a
1869  first pair, \fIovector[0]\fP and \fIovector[1]\fP, identify the portion of the  substring. \fBNote\fP: these values are always byte offsets, even in UTF-8
1870  subject string matched by the entire pattern. The next pair is used for the  mode. They are not character counts.
1871  first capturing subpattern, and so on. The value returned by \fBpcre_exec()\fP  .P
1872  is the number of pairs that have been set. If there are no capturing  The first pair of integers, \fIovector[0]\fP and \fIovector[1]\fP, identify the
1873  subpatterns, the return value from a successful match is 1, indicating that  portion of the subject string matched by the entire pattern. The next pair is
1874  just the first pair of offsets has been set.  used for the first capturing subpattern, and so on. The value returned by
1875  .P  \fBpcre_exec()\fP is one more than the highest numbered pair that has been set.
1876  Some convenience functions are provided for extracting the captured substrings  For example, if two substrings have been captured, the returned value is 3. If
1877  as separate strings. These are described in the following section.  there are no capturing subpatterns, the return value from a successful match is
1878  .P  1, indicating that just the first pair of offsets has been set.
 It is possible for an capturing subpattern number \fIn+1\fP to match some  
 part of the subject when subpattern \fIn\fP has not been used at all. For  
 example, if the string "abc" is matched against the pattern (a|(z))(bc)  
 subpatterns 1 and 3 are matched, but 2 is not. When this happens, both offset  
 values corresponding to the unused subpattern are set to -1.  
1879  .P  .P
1880  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
1881  string that it matched that is returned.  string that it matched that is returned.
1882  .P  .P
1883  If the vector is too small to hold all the captured substring offsets, it is  If the vector is too small to hold all the captured substring offsets, it is
1884  used as far as possible (up to two-thirds of its length), and the function  used as far as possible (up to two-thirds of its length), and the function
1885  returns a value of zero. In particular, if the substring offsets are not of  returns a value of zero. If neither the actual string matched not any captured
1886  interest, \fBpcre_exec()\fP may be called with \fIovector\fP passed as NULL and  substrings are of interest, \fBpcre_exec()\fP may be called with \fIovector\fP
1887  \fIovecsize\fP as zero. However, if the pattern contains back references and  passed as NULL and \fIovecsize\fP as zero. However, if the pattern contains
1888  the \fIovector\fP is not big enough to remember the related substrings, PCRE  back references and the \fIovector\fP is not big enough to remember the related
1889  has to get additional memory for use during matching. Thus it is usually  substrings, PCRE has to get additional memory for use during matching. Thus it
1890  advisable to supply an \fIovector\fP.  is usually advisable to supply an \fIovector\fP of reasonable size.
1891    .P
1892    There are some cases where zero is returned (indicating vector overflow) when
1893    in fact the vector is exactly the right size for the final match. For example,
1894    consider the pattern
1895    .sp
1896      (a)(?:(b)c|bd)
1897    .sp
1898    If a vector of 6 elements (allowing for only 1 captured substring) is given
1899    with subject string "abd", \fBpcre_exec()\fP will try to set the second
1900    captured string, thereby recording a vector overflow, before failing to match
1901    "c" and backing up to try the second alternative. The zero return, however,
1902    does correctly indicate that the maximum number of slots (namely 2) have been
1903    filled. In similar cases where there is temporary overflow, but the final
1904    number of used slots is actually less than the maximum, a non-zero value is
1905    returned.
1906  .P  .P
1907  Note that \fBpcre_info()\fP can be used to find out how many capturing  The \fBpcre_fullinfo()\fP function can be used to find out how many capturing
1908  subpatterns there are in a compiled pattern. The smallest size for  subpatterns there are in a compiled pattern. The smallest size for
1909  \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
1910  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.
1911    .P
1912    It is possible for capturing subpattern number \fIn+1\fP to match some part of
1913    the subject when subpattern \fIn\fP has not been used at all. For example, if
1914    the string "abc" is matched against the pattern (a|(z))(bc) the return from the
1915    function is 4, and subpatterns 1 and 3 are matched, but 2 is not. When this
1916    happens, both values in the offset pairs corresponding to unused subpatterns
1917    are set to -1.
1918    .P
1919    Offset values that correspond to unused subpatterns at the end of the
1920    expression are also set to -1. For example, if the string "abc" is matched
1921    against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not matched. The
1922    return from the function is 2, because the highest used capturing subpattern
1923    number is 1, and the offsets for for the second and third capturing subpatterns
1924    (assuming the vector is large enough, of course) are set to -1.
1925    .P
1926    \fBNote\fP: Elements in the first two-thirds of \fIovector\fP that do not
1927    correspond to capturing parentheses in the pattern are never changed. That is,
1928    if a pattern contains \fIn\fP capturing parentheses, no more than
1929    \fIovector[0]\fP to \fIovector[2n+1]\fP are set by \fBpcre_exec()\fP. The other
1930    elements (in the first two-thirds) retain whatever values they previously had.
1931    .P
1932    Some convenience functions are provided for extracting the captured substrings
1933    as separate strings. These are described below.
1934    .
1935  .  .
1936  .\" HTML <a name="errorlist"></a>  .\" HTML <a name="errorlist"></a>
1937  .SS "Return values from \fBpcre_exec()\fP"  .SS "Error return values from \fBpcre_exec()\fP"
1938  .rs  .rs
1939  .sp  .sp
1940  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 1256  compiled in an environment of one endian Line 1961  compiled in an environment of one endian
1961  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
1962  not present.  not present.
1963  .sp  .sp
1964    PCRE_ERROR_UNKNOWN_NODE   (-5)    PCRE_ERROR_UNKNOWN_OPCODE (-5)
1965  .sp  .sp
1966  While running the pattern match, an unknown item was encountered in the  While running the pattern match, an unknown item was encountered in the
1967  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 1269  If a pattern contains back references, b Line 1974  If a pattern contains back references, b
1974  gets a block of memory at the start of matching to use for this purpose. If the  gets a block of memory at the start of matching to use for this purpose. If the
1975  call via \fBpcre_malloc()\fP fails, this error is given. The memory is  call via \fBpcre_malloc()\fP fails, this error is given. The memory is
1976  automatically freed at the end of matching.  automatically freed at the end of matching.
1977    .P
1978    This error is also given if \fBpcre_stack_malloc()\fP fails in
1979    \fBpcre_exec()\fP. This can happen only when PCRE has been compiled with
1980    \fB--disable-stack-for-recursion\fP.
1981  .sp  .sp
1982    PCRE_ERROR_NOSUBSTRING    (-7)    PCRE_ERROR_NOSUBSTRING    (-7)
1983  .sp  .sp
# Line 1282  The backtracking limit, as specified by Line 1991  The backtracking limit, as specified by
1991  \fBpcre_extra\fP structure (or defaulted) was reached. See the description  \fBpcre_extra\fP structure (or defaulted) was reached. See the description
1992  above.  above.
1993  .sp  .sp
   PCRE_ERROR_RECURSIONLIMIT (-21)  
 .sp  
 The internal recursion limit, as specified by the \fImatch_limit_recursion\fP  
 field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the  
 description above.  
 .sp  
1994    PCRE_ERROR_CALLOUT        (-9)    PCRE_ERROR_CALLOUT        (-9)
1995  .sp  .sp
1996  This error is never generated by \fBpcre_exec()\fP itself. It is provided for  This error is never generated by \fBpcre_exec()\fP itself. It is provided for
# Line 1299  documentation for details. Line 2002  documentation for details.
2002  .sp  .sp
2003    PCRE_ERROR_BADUTF8        (-10)    PCRE_ERROR_BADUTF8        (-10)
2004  .sp  .sp
2005  A string that contains an invalid UTF-8 byte sequence was passed as a subject.  A string that contains an invalid UTF-8 byte sequence was passed as a subject,
2006    and the PCRE_NO_UTF8_CHECK option was not set. If the size of the output vector
2007    (\fIovecsize\fP) is at least 2, the byte offset to the start of the the invalid
2008    UTF-8 character is placed in the first element, and a reason code is placed in
2009    the second element. The reason codes are listed in the
2010    .\" HTML <a href="#badutf8reasons">
2011    .\" </a>
2012    following section.
2013    .\"
2014    For backward compatibility, if PCRE_PARTIAL_HARD is set and the problem is a
2015    truncated UTF-8 character at the end of the subject (reason codes 1 to 5),
2016    PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2017  .sp  .sp
2018    PCRE_ERROR_BADUTF8_OFFSET (-11)    PCRE_ERROR_BADUTF8_OFFSET (-11)
2019  .sp  .sp
2020  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 checked and found to
2021  of \fIstartoffset\fP did not point to the beginning of a UTF-8 character.  be valid (the PCRE_NO_UTF8_CHECK option was not set), but the value of
2022    \fIstartoffset\fP did not point to the beginning of a UTF-8 character or the
2023    end of the subject.
2024  .sp  .sp
2025    PCRE_ERROR_PARTIAL        (-12)    PCRE_ERROR_PARTIAL        (-12)
2026  .sp  .sp
# Line 1316  documentation for details of partial mat Line 2032  documentation for details of partial mat
2032  .sp  .sp
2033    PCRE_ERROR_BADPARTIAL     (-13)    PCRE_ERROR_BADPARTIAL     (-13)
2034  .sp  .sp
2035  The PCRE_PARTIAL option was used with a compiled pattern containing items that  This code is no longer in use. It was formerly returned when the PCRE_PARTIAL
2036  are not supported for partial matching. See the  option was used with a compiled pattern containing items that were not
2037  .\" HREF  supported for partial matching. From release 8.00 onwards, there are no
2038  \fBpcrepartial\fP  restrictions on partial matching.
 .\"  
 documentation for details of partial matching.  
2039  .sp  .sp
2040    PCRE_ERROR_INTERNAL       (-14)    PCRE_ERROR_INTERNAL       (-14)
2041  .sp  .sp
# Line 1331  in PCRE or by overwriting of the compile Line 2045  in PCRE or by overwriting of the compile
2045    PCRE_ERROR_BADCOUNT       (-15)    PCRE_ERROR_BADCOUNT       (-15)
2046  .sp  .sp
2047  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.
2048    .sp
2049      PCRE_ERROR_RECURSIONLIMIT (-21)
2050    .sp
2051    The internal recursion limit, as specified by the \fImatch_limit_recursion\fP
2052    field in a \fBpcre_extra\fP structure (or defaulted) was reached. See the
2053    description above.
2054    .sp
2055      PCRE_ERROR_BADNEWLINE     (-23)
2056    .sp
2057    An invalid combination of PCRE_NEWLINE_\fIxxx\fP options was given.
2058    .sp
2059      PCRE_ERROR_BADOFFSET      (-24)
2060    .sp
2061    The value of \fIstartoffset\fP was negative or greater than the length of the
2062    subject, that is, the value in \fIlength\fP.
2063    .sp
2064      PCRE_ERROR_SHORTUTF8      (-25)
2065    .sp
2066    This error is returned instead of PCRE_ERROR_BADUTF8 when the subject string
2067    ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD option is set.
2068    Information about the failure is returned as for PCRE_ERROR_BADUTF8. It is in
2069    fact sufficient to detect this case, but this special error code for
2070    PCRE_PARTIAL_HARD precedes the implementation of returned information; it is
2071    retained for backwards compatibility.
2072    .sp
2073      PCRE_ERROR_RECURSELOOP    (-26)
2074    .sp
2075    This error is returned when \fBpcre_exec()\fP detects a recursion loop within
2076    the pattern. Specifically, it means that either the whole pattern or a
2077    subpattern has been called recursively for the second time at the same position
2078    in the subject string. Some simple patterns that might do this are detected and
2079    faulted at compile time, but more complicated cases, in particular mutual
2080    recursions between two different subpatterns, cannot be detected until run
2081    time.
2082    .sp
2083      PCRE_ERROR_JIT_STACKLIMIT (-27)
2084    .sp
2085    This error is returned when a pattern that was successfully studied using the
2086    PCRE_STUDY_JIT_COMPILE option is being matched, but the memory available for
2087    the just-in-time processing stack is not large enough. See the
2088    .\" HREF
2089    \fBpcrejit\fP
2090    .\"
2091    documentation for more details.
2092    .sp
2093      PCRE_ERROR_BADMODE (-28)
2094    .sp
2095    This error is given if a pattern that was compiled by the 8-bit library is
2096    passed to a 16-bit library function, or vice versa.
2097    .sp
2098      PCRE_ERROR_BADENDIANNESS (-29)
2099    .sp
2100    This error is given if a pattern that was compiled and saved is reloaded on a
2101    host with different endianness. The utility function
2102    \fBpcre_pattern_to_host_byte_order()\fP can be used to convert such a pattern
2103    so that it runs on the new host.
2104    .P
2105    Error numbers -16 to -20 and -22 are not used by \fBpcre_exec()\fP.
2106    .
2107    .
2108    .\" HTML <a name="badutf8reasons"></a>
2109    .SS "Reason codes for invalid UTF-8 strings"
2110    .rs
2111    .sp
2112    This section applies only to the 8-bit library. The corresponding information
2113    for the 16-bit library is given in the
2114    .\" HREF
2115    \fBpcre16\fP
2116    .\"
2117    page.
2118    .P
2119    When \fBpcre_exec()\fP returns either PCRE_ERROR_BADUTF8 or
2120    PCRE_ERROR_SHORTUTF8, and the size of the output vector (\fIovecsize\fP) is at
2121    least 2, the offset of the start of the invalid UTF-8 character is placed in
2122    the first output vector element (\fIovector[0]\fP) and a reason code is placed
2123    in the second element (\fIovector[1]\fP). The reason codes are given names in
2124    the \fBpcre.h\fP header file:
2125    .sp
2126      PCRE_UTF8_ERR1
2127      PCRE_UTF8_ERR2
2128      PCRE_UTF8_ERR3
2129      PCRE_UTF8_ERR4
2130      PCRE_UTF8_ERR5
2131    .sp
2132    The string ends with a truncated UTF-8 character; the code specifies how many
2133    bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8 characters to be
2134    no longer than 4 bytes, the encoding scheme (originally defined by RFC 2279)
2135    allows for up to 6 bytes, and this is checked first; hence the possibility of
2136    4 or 5 missing bytes.
2137    .sp
2138      PCRE_UTF8_ERR6
2139      PCRE_UTF8_ERR7
2140      PCRE_UTF8_ERR8
2141      PCRE_UTF8_ERR9
2142      PCRE_UTF8_ERR10
2143    .sp
2144    The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of the
2145    character do not have the binary value 0b10 (that is, either the most
2146    significant bit is 0, or the next bit is 1).
2147    .sp
2148      PCRE_UTF8_ERR11
2149      PCRE_UTF8_ERR12
2150    .sp
2151    A character that is valid by the RFC 2279 rules is either 5 or 6 bytes long;
2152    these code points are excluded by RFC 3629.
2153    .sp
2154      PCRE_UTF8_ERR13
2155    .sp
2156    A 4-byte character has a value greater than 0x10fff; these code points are
2157    excluded by RFC 3629.
2158    .sp
2159      PCRE_UTF8_ERR14
2160    .sp
2161    A 3-byte character has a value in the range 0xd800 to 0xdfff; this range of
2162    code points are reserved by RFC 3629 for use with UTF-16, and so are excluded
2163    from UTF-8.
2164    .sp
2165      PCRE_UTF8_ERR15
2166      PCRE_UTF8_ERR16
2167      PCRE_UTF8_ERR17
2168      PCRE_UTF8_ERR18
2169      PCRE_UTF8_ERR19
2170    .sp
2171    A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes for a
2172    value that can be represented by fewer bytes, which is invalid. For example,
2173    the two bytes 0xc0, 0xae give the value 0x2e, whose correct coding uses just
2174    one byte.
2175    .sp
2176      PCRE_UTF8_ERR20
2177    .sp
2178    The two most significant bits of the first byte of a character have the binary
2179    value 0b10 (that is, the most significant bit is 1 and the second is 0). Such a
2180    byte can only validly occur as the second or subsequent byte of a multi-byte
2181    character.
2182    .sp
2183      PCRE_UTF8_ERR21
2184    .sp
2185    The first byte of a character has the value 0xfe or 0xff. These values can
2186    never occur in a valid UTF-8 string.
2187  .  .
2188  .  .
2189  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"  .SH "EXTRACTING CAPTURED SUBSTRINGS BY NUMBER"
# Line 1342  This error is given if the value of the Line 2195  This error is given if the value of the
2195  .ti +5n  .ti +5n
2196  .B int \fIbuffersize\fP);  .B int \fIbuffersize\fP);
2197  .PP  .PP
 .br  
2198  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,  .B int pcre_get_substring(const char *\fIsubject\fP, int *\fIovector\fP,
2199  .ti +5n  .ti +5n
2200  .B int \fIstringcount\fP, int \fIstringnumber\fP,  .B int \fIstringcount\fP, int \fIstringnumber\fP,
2201  .ti +5n  .ti +5n
2202  .B const char **\fIstringptr\fP);  .B const char **\fIstringptr\fP);
2203  .PP  .PP
 .br  
2204  .B int pcre_get_substring_list(const char *\fIsubject\fP,  .B int pcre_get_substring_list(const char *\fIsubject\fP,
2205  .ti +5n  .ti +5n
2206  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"  .B int *\fIovector\fP, int \fIstringcount\fP, "const char ***\fIlistptr\fP);"
# Line 1360  Captured substrings can be accessed dire Line 2211  Captured substrings can be accessed dire
2211  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings  \fBpcre_get_substring_list()\fP are provided for extracting captured substrings
2212  as new, separate, zero-terminated strings. These functions identify substrings  as new, separate, zero-terminated strings. These functions identify substrings
2213  by number. The next section describes functions for extracting named  by number. The next section describes functions for extracting named
2214  substrings. A substring that contains a binary zero is correctly extracted and  substrings.
2215  has a further zero added on the end, but the result is not, of course,  .P
2216  a C string.  A substring that contains a binary zero is correctly extracted and has a
2217    further zero added on the end, but the result is not, of course, a C string.
2218    However, you can process such a string by referring to the length that is
2219    returned by \fBpcre_copy_substring()\fP and \fBpcre_get_substring()\fP.
2220    Unfortunately, the interface to \fBpcre_get_substring_list()\fP is not adequate
2221    for handling strings containing binary zeros, because the end of the final
2222    string is not independently indicated.
2223  .P  .P
2224  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:
2225  \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 1382  the string is placed in \fIbuffer\fP, wh Line 2239  the string is placed in \fIbuffer\fP, wh
2239  \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
2240  obtained via \fBpcre_malloc\fP, and its address is returned via  obtained via \fBpcre_malloc\fP, and its address is returned via
2241  \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
2242  including the terminating zero, or one of  including the terminating zero, or one of these error codes:
2243  .sp  .sp
2244    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
2245  .sp  .sp
# Line 1398  and builds a list of pointers to them. A Line 2255  and builds a list of pointers to them. A
2255  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
2256  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
2257  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
2258  function is zero if all went well, or  function is zero if all went well, or the error code
2259  .sp  .sp
2260    PCRE_ERROR_NOMEMORY       (-6)    PCRE_ERROR_NOMEMORY       (-6)
2261  .sp  .sp
# Line 1417  a previous call of \fBpcre_get_substring Line 2274  a previous call of \fBpcre_get_substring
2274  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call  \fBpcre_get_substring_list()\fP, respectively. They do nothing more than call
2275  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
2276  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
2277  linked via a special interface to another programming language which cannot use  linked via a special interface to another programming language that cannot use
2278  \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
2279  provided.  provided.
2280  .  .
# Line 1429  provided. Line 2286  provided.
2286  .ti +5n  .ti +5n
2287  .B const char *\fIname\fP);  .B const char *\fIname\fP);
2288  .PP  .PP
 .br  
2289  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,  .B int pcre_copy_named_substring(const pcre *\fIcode\fP,
2290  .ti +5n  .ti +5n
2291  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1438  provided. Line 2294  provided.
2294  .ti +5n  .ti +5n
2295  .B char *\fIbuffer\fP, int \fIbuffersize\fP);  .B char *\fIbuffer\fP, int \fIbuffersize\fP);
2296  .PP  .PP
 .br  
2297  .B int pcre_get_named_substring(const pcre *\fIcode\fP,  .B int pcre_get_named_substring(const pcre *\fIcode\fP,
2298  .ti +5n  .ti +5n
2299  .B const char *\fIsubject\fP, int *\fIovector\fP,  .B const char *\fIsubject\fP, int *\fIovector\fP,
# Line 1450  provided. Line 2305  provided.
2305  To extract a substring by name, you first have to find associated number.  To extract a substring by name, you first have to find associated number.
2306  For example, for this pattern  For example, for this pattern
2307  .sp  .sp
2308    (a+)b(?P<xxx>\ed+)...    (a+)b(?<xxx>\ed+)...
2309  .sp  .sp
2310  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
2311  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
2312  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
2313    pattern, and the second is the name. The yield of the function is the
2314  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
2315  that name.  that name.
2316  .P  .P
# Line 1462  Given the number, you can extract the su Line 2318  Given the number, you can extract the su
2318  functions described in the previous section. For convenience, there are also  functions described in the previous section. For convenience, there are also
2319  two functions that do the whole job.  two functions that do the whole job.
2320  .P  .P
2321  Most of the arguments of \fIpcre_copy_named_substring()\fP and  Most of the arguments of \fBpcre_copy_named_substring()\fP and
2322  \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
2323  functions that extract by number. As these are described in the previous  functions that extract by number. As these are described in the previous
2324  section, they are not re-described here. There are just two differences:  section, they are not re-described here. There are just two differences:
2325  .P  .P
# Line 1473  pattern. This is needed in order to gain Line 2329  pattern. This is needed in order to gain
2329  translation table.  translation table.
2330  .P  .P
2331  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they  These functions call \fBpcre_get_stringnumber()\fP, and if it succeeds, they
2332  then call \fIpcre_copy_substring()\fP or \fIpcre_get_substring()\fP, as  then call \fBpcre_copy_substring()\fP or \fBpcre_get_substring()\fP, as
2333  appropriate.  appropriate. \fBNOTE:\fP If PCRE_DUPNAMES is set and there are duplicate names,
2334    the behaviour may not be what you want (see the next section).
2335    .P
2336    \fBWarning:\fP If the pattern uses the (?| feature to set up multiple
2337    subpatterns with the same number, as described in the
2338    .\" HTML <a href="pcrepattern.html#dupsubpatternnumber">
2339    .\" </a>
2340    section on duplicate subpattern numbers
2341    .\"
2342    in the
2343    .\" HREF
2344    \fBpcrepattern\fP
2345    .\"
2346    page, you cannot use names to distinguish the different subpatterns, because
2347    names are not included in the compiled code. The matching process uses only
2348    numbers. For this reason, the use of different names for subpatterns of the
2349    same number causes an error at compile time.
2350    .
2351    .
2352    .SH "DUPLICATE SUBPATTERN NAMES"
2353    .rs
2354    .sp
2355    .B int pcre_get_stringtable_entries(const pcre *\fIcode\fP,
2356    .ti +5n
2357    .B const char *\fIname\fP, char **\fIfirst\fP, char **\fIlast\fP);
2358    .PP
2359    When a pattern is compiled with the PCRE_DUPNAMES option, names for subpatterns
2360    are not required to be unique. (Duplicate names are always allowed for
2361    subpatterns with the same number, created by using the (?| feature. Indeed, if
2362    such subpatterns are named, they are required to use the same names.)
2363    .P
2364    Normally, patterns with duplicate names are such that in any one match, only
2365    one of the named subpatterns participates. An example is shown in the
2366    .\" HREF
2367    \fBpcrepattern\fP
2368    .\"
2369    documentation.
2370    .P
2371    When duplicates are present, \fBpcre_copy_named_substring()\fP and
2372    \fBpcre_get_named_substring()\fP return the first substring corresponding to
2373    the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING (-7) is
2374    returned; no data is returned. The \fBpcre_get_stringnumber()\fP function
2375    returns one of the numbers that are associated with the name, but it is not
2376    defined which it is.
2377    .P
2378    If you want to get full details of all captured substrings for a given name,
2379    you must use the \fBpcre_get_stringtable_entries()\fP function. The first
2380    argument is the compiled pattern, and the second is the name. The third and
2381    fourth are pointers to variables which are updated by the function. After it
2382    has run, they point to the first and last entries in the name-to-number table
2383    for the given name. The function itself returns the length of each entry, or
2384    PCRE_ERROR_NOSUBSTRING (-7) if there are none. The format of the table is
2385    described above in the section entitled \fIInformation about a pattern\fP
2386    .\" HTML <a href="#infoaboutpattern">
2387    .\" </a>
2388    above.
2389    .\"
2390    Given all the relevant entries for the name, you can extract each of their
2391    numbers, and hence the captured data, if any.
2392  .  .
2393  .  .
2394  .SH "FINDING ALL POSSIBLE MATCHES"  .SH "FINDING ALL POSSIBLE MATCHES"
# Line 1512  will yield PCRE_ERROR_NOMATCH. Line 2426  will yield PCRE_ERROR_NOMATCH.
2426  .B int *\fIworkspace\fP, int \fIwscount\fP);  .B int *\fIworkspace\fP, int \fIwscount\fP);
2427  .P  .P
2428  The function \fBpcre_dfa_exec()\fP is called to match a subject string against  The function \fBpcre_dfa_exec()\fP is called to match a subject string against
2429  a compiled pattern, using a "DFA" matching algorithm. This has different  a compiled pattern, using a matching algorithm that scans the subject string
2430  characteristics to the normal algorithm, and is not compatible with Perl. Some  just once, and does not backtrack. This has different characteristics to the
2431  of the features of PCRE patterns are not supported. Nevertheless, there are  normal algorithm, and is not compatible with Perl. Some of the features of PCRE
2432  times when this kind of matching can be useful. For a discussion of the two  patterns are not supported. Nevertheless, there are times when this kind of
2433  matching algorithms, see the  matching can be useful. For a discussion of the two matching algorithms, and a
2434    list of features that \fBpcre_dfa_exec()\fP does not support, see the
2435  .\" HREF  .\" HREF
2436  \fBpcrematching\fP  \fBpcrematching\fP
2437  .\"  .\"
# Line 1531  here. Line 2446  here.
2446  The two additional arguments provide workspace for the function. The workspace  The two additional arguments provide workspace for the function. The workspace
2447  vector should contain at least 20 elements. It is used for keeping track of  vector should contain at least 20 elements. It is used for keeping track of
2448  multiple paths through the pattern tree. More workspace will be needed for  multiple paths through the pattern tree. More workspace will be needed for
2449  patterns and subjects where there are a lot of possible matches.  patterns and subjects where there are a lot of potential matches.
2450  .P  .P
2451  Here is an example of a simple call to \fBpcre_dfa_exec()\fP:  Here is an example of a simple call to \fBpcre_dfa_exec()\fP:
2452  .sp  .sp
# Line 1554  Here is an example of a simple call to \ Line 2469  Here is an example of a simple call to \
2469  .rs  .rs
2470  .sp  .sp
2471  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be  The unused bits of the \fIoptions\fP argument for \fBpcre_dfa_exec()\fP must be
2472  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,
2473  PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL,  PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2474  PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last three of these are  PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF, PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE,
2475  the same as for \fBpcre_exec()\fP, so their description is not repeated here.  PCRE_PARTIAL_HARD, PCRE_PARTIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART.
2476  .sp  All but the last four of these are exactly the same as for \fBpcre_exec()\fP,
2477    PCRE_PARTIAL  so their description is not repeated here.
2478  .sp  .sp
2479  This has the same general effect as it does for \fBpcre_exec()\fP, but the    PCRE_PARTIAL_HARD
2480  details are slightly different. When PCRE_PARTIAL is set for    PCRE_PARTIAL_SOFT
2481  \fBpcre_dfa_exec()\fP, the return code PCRE_ERROR_NOMATCH is converted into  .sp
2482  PCRE_ERROR_PARTIAL if the end of the subject is reached, there have been no  These have the same general effect as they do for \fBpcre_exec()\fP, but the
2483  complete matches, but there is still at least one matching possibility. The  details are slightly different. When PCRE_PARTIAL_HARD is set for
2484  portion of the string that provided the partial match is set as the first  \fBpcre_dfa_exec()\fP, it returns PCRE_ERROR_PARTIAL if the end of the subject
2485  matching string.  is reached and there is still at least one matching possibility that requires
2486    additional characters. This happens even if some complete matches have also
2487    been found. When PCRE_PARTIAL_SOFT is set, the return code PCRE_ERROR_NOMATCH
2488    is converted into PCRE_ERROR_PARTIAL if the end of the subject is reached,
2489    there have been no complete matches, but there is still at least one matching
2490    possibility. The portion of the string that was inspected when the longest
2491    partial match was found is set as the first matching string in both cases.
2492    There is a more detailed discussion of partial and multi-segment matching, with
2493    examples, in the
2494    .\" HREF
2495    \fBpcrepartial\fP
2496    .\"
2497    documentation.
2498  .sp  .sp
2499    PCRE_DFA_SHORTEST    PCRE_DFA_SHORTEST
2500  .sp  .sp
2501  Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as  Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to stop as
2502  soon as it has found one match. Because of the way the DFA algorithm works,  soon as it has found one match. Because of the way the alternative algorithm
2503  this is necessarily the shortest possible match at the first possible matching  works, this is necessarily the shortest possible match at the first possible
2504  point in the subject string.  matching point in the subject string.
2505  .sp  .sp
2506    PCRE_DFA_RESTART    PCRE_DFA_RESTART
2507  .sp  .sp
2508  When \fBpcre_dfa_exec()\fP is called with the PCRE_PARTIAL option, and returns  When \fBpcre_dfa_exec()\fP returns a partial match, it is possible to call it
2509  a partial match, it is possible to call it again, with additional subject  again, with additional subject characters, and have it continue with the same
2510  characters, and have it continue with the same match. The PCRE_DFA_RESTART  match. The PCRE_DFA_RESTART option requests this action; when it is set, the
2511  option requests this action; when it is set, the \fIworkspace\fP and  \fIworkspace\fP and \fIwscount\fP options must reference the same vector as
2512  \fIwscount\fP options must reference the same vector as before because data  before because data about the match so far is left in them after a partial
2513  about the match so far is left in them after a partial match. There is more  match. There is more discussion of this facility in the
 discussion of this facility in the  
2514  .\" HREF  .\" HREF
2515  \fBpcrepartial\fP  \fBpcrepartial\fP
2516  .\"  .\"
2517  documentation.  documentation.
2518  .  .
2519    .
2520  .SS "Successful returns from \fBpcre_dfa_exec()\fP"  .SS "Successful returns from \fBpcre_dfa_exec()\fP"
2521  .rs  .rs
2522  .sp  .sp
# Line 1613  the three matched strings are Line 2540  the three matched strings are
2540  On success, the yield of the function is a number greater than zero, which is  On success, the yield of the function is a number greater than zero, which is
2541  the number of matched substrings. The substrings themselves are returned in  the number of matched substrings. The substrings themselves are returned in
2542  \fIovector\fP. Each string uses two elements; the first is the offset to the  \fIovector\fP. Each string uses two elements; the first is the offset to the
2543  start, and the second is the offset to the end. All the strings have the same  start, and the second is the offset to the end. In fact, all the strings have
2544  start offset. (Space could have been saved by giving this only once, but it was  the same start offset. (Space could have been saved by giving this only once,
2545  decided to retain some compatibility with the way \fBpcre_exec()\fP returns  but it was decided to retain some compatibility with the way \fBpcre_exec()\fP
2546  data, even though the meaning of the strings is different.)  returns data, even though the meaning of the strings is different.)
2547  .P  .P
2548  The strings are returned in reverse order of length; that is, the longest  The strings are returned in reverse order of length; that is, the longest
2549  matching string is given first. If there were too many matches to fit into  matching string is given first. If there were too many matches to fit into
2550  \fIovector\fP, the yield of the function is zero, and the vector is filled with  \fIovector\fP, the yield of the function is zero, and the vector is filled with
2551  the longest matches.  the longest matches. Unlike \fBpcre_exec()\fP, \fBpcre_dfa_exec()\fP can use
2552    the entire \fIovector\fP for returning matched strings.
2553    .
2554  .  .
2555  .SS "Error returns from \fBpcre_dfa_exec()\fP"  .SS "Error returns from \fBpcre_dfa_exec()\fP"
2556  .rs  .rs
# Line 1643  that it does not support, for instance, Line 2572  that it does not support, for instance,
2572  .sp  .sp
2573    PCRE_ERROR_DFA_UCOND      (-17)    PCRE_ERROR_DFA_UCOND      (-17)
2574  .sp  .sp
2575  This return is given if \fBpcre_dfa_exec()\fP encounters a condition item in a  This return is given if \fBpcre_dfa_exec()\fP encounters a condition item that
2576  pattern that uses a back reference for the condition. This is not supported.  uses a back reference for the condition, or a test for recursion in a specific
2577    group. These are not supported.
2578  .sp  .sp
2579    PCRE_ERROR_DFA_UMLIMIT    (-18)    PCRE_ERROR_DFA_UMLIMIT    (-18)
2580  .sp  .sp
2581  This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP  This return is given if \fBpcre_dfa_exec()\fP is called with an \fIextra\fP
2582  block that contains a setting of the \fImatch_limit\fP field. This is not  block that contains a setting of the \fImatch_limit\fP or
2583  supported (it is meaningless).  \fImatch_limit_recursion\fP fields. This is not supported (these fields are
2584    meaningless for DFA matching).
2585  .sp  .sp
2586    PCRE_ERROR_DFA_WSSIZE     (-19)    PCRE_ERROR_DFA_WSSIZE     (-19)
2587  .sp  .sp
# Line 1663  When a recursive subpattern is processed Line 2594  When a recursive subpattern is processed
2594  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This  recursively, using private vectors for \fIovector\fP and \fIworkspace\fP. This
2595  error is given if the output vector is not large enough. This should be  error is given if the output vector is not large enough. This should be
2596  extremely rare, as a vector of size 1000 is used.  extremely rare, as a vector of size 1000 is used.
2597  .P  .
2598  .in 0  .
2599  Last updated: 18 January 2006  .SH "SEE ALSO"
2600  .br  .rs
2601  Copyright (c) 1997-2006 University of Cambridge.  .sp
2602    \fBpcre16\fP(3), \fBpcrebuild\fP(3), \fBpcrecallout\fP(3), \fBpcrecpp(3)\fP(3),
2603    \fBpcrematching\fP(3), \fBpcrepartial\fP(3), \fBpcreposix\fP(3),
2604    \fBpcreprecompile\fP(3), \fBpcresample\fP(3), \fBpcrestack\fP(3).
2605    .
2606    .
2607    .SH AUTHOR
2608    .rs
2609    .sp
2610    .nf
2611    Philip Hazel
2612    University Computing Service
2613    Cambridge CB2 3QH, England.
2614    .fi
2615    .
2616    .
2617    .SH REVISION
2618    .rs
2619    .sp
2620    .nf
2621    Last updated: 17 January 2012
2622    Copyright (c) 1997-2012 University of Cambridge.
2623    .fi

Legend:
Removed from v.87  
changed lines
  Added in v.887

  ViewVC Help
Powered by ViewVC 1.1.5